Disclosed are a method for improving a surface quality of an alloy micro-area via a supersaturated film and use thereof. The method includes: adding nickel chloride to a sodium chloride-ethylene glycol electrolyte until the electrolyte is saturated, and conducting electrochemical machining.
22222 supported catalyst that is prone to agglomeration is greatly ameliorated, the utilization rate and stability of metal atoms are greatly improved, and this synthesis strategy can be popularized to preparation of other heterogeneous catalysts.
B01J 31/02 - Catalysts comprising hydrides, coordination complexes or organic compounds containing organic compounds or metal hydrides
C07C 209/36 - Preparation of compounds containing amino groups bound to a carbon skeleton by reduction of nitrogen-to-oxygen or nitrogen-to-nitrogen bonds by reduction of nitro groups by reduction of nitro groups bound to carbon atoms of six-membered aromatic rings
C07C 211/52 - Compounds containing amino groups bound to a carbon skeleton having amino groups bound to carbon atoms of six-membered aromatic rings of the carbon skeleton having amino groups bound to only one six-membered aromatic ring the carbon skeleton being further substituted by halogen atoms or by nitro or nitroso groups
3.
PREPARATION METHOD OF FLAME-RETARDANT ULTRATHIN PEO-BASED SOLID ELECTROLYTE
A preparation method of a flame-retardant ultrathin PEO-based solid electrolyte is disclosed. The method includes the following steps: preparing a CN support layer; synthesizing a flame retardant-loaded multifunctional filler: HNT@TMP; mixing and stirring PEO, LiTFSI, and HNT@TMP in a certain ratio in acetonitrile to obtain PEO-based solid electrolyte slurry; coating both sides of the CN support layer obtained in step S1 with the PEO-based solid electrolyte slurry obtained in step S3, and performing drying; and performing hot pressing to obtain a PEO-based solid electrolyte. By adopting the preparation method of the flame-retardant ultrathin PEO-based solid electrolyte, the electrochemical performance and flame retardance of a PEO-based solid polymer electrolyte are improved through a multifunctional flame-retardant filler (HNT@TMP), and the mechanical strength of the ultrathin PEO electrolyte is ensured through porous cellulose nanopaper (CN) with excellent mechanical flexibility and thermal stability, whereby the development of high energy density is facilitated.
The present disclosure provides a staged shearing and forming method for a T-bar cylindrical member, comprising: controlling a shear working surface of a shear spinning wheel in contact with a surface of a cylindrical member blank; controlling a circumferential rotation of the cylindrical member blank; controlling a working surface of a flow spinning wheel to be in perpendicular contact with a surface of the unsaturated I-bar cylindrical member and controlling the cylindrical member blank to be maintained in a circumferential rotational state; controlling the flow spinning wheel to thin one side of the unsaturated I-bar cylindrical member; controlling a fractal working surface of a fractal spinning wheel in contact with a top of the saturated I-bar structure; controlling a working surface of a flat spinning wheel to be in perpendicular contact with a surface of the Y-bar cylindrical member.
A processing device, communicatively coupled to a microphone array with a first array geometry, may adapt the microphone array to a target beamformer associated with a target microphone array with a second array geometry. The processing device may, responsive to sound sources in a three-dimensional (3D) space, obtain a first plurality of electronic signals generated by the microphone array, divide a set of angles within the 3D space into J subsets of angles and identify a location of each sound source in the 3D space, with respect to the subsets of angles, based on the first plurality of electronic signals. An associated cost function may be evaluated for each frequency sub-band of a plurality of frequency sub-bands and for each subset of angles and a second plurality of electronic signals corresponding to the target microphone array may be generated based on the evaluation of the associated cost function.
H04R 1/40 - Arrangements for obtaining desired frequency or directional characteristics for obtaining desired directional characteristic only by combining a number of identical transducers
Disclosed is a nighttime cooperative positioning method based on an unmanned aerial vehicle (UAV) group, falling within the technical field of aircraft navigation and positioning. According to the present disclosure, the cooperative visual positioning and the collision warning for UAVs are realized by means of light colors of the UAVs, respective two-dimensional turntable cameras and a communication topology network, without adding additional equipment and without relying on an external signal source, avoiding external interference. Compared with the positioning method in a conventional manner, in the present disclosure, the system is effectively simplified, and the cooperative positioning among the interiors of a UAV cluster can be realized relatively simply and at a low cost to maintain the formation of the UAV group.
G05D 1/695 - Coordinated control of the position or course of two or more vehicles for maintaining a fixed relative position of the vehicles, e.g. for convoy travelling or formation flight
B64D 47/06 - Arrangements or adaptations of signal or lighting devices for indicating aircraft presence
G05D 1/249 - Arrangements for determining position or orientation using signals provided by artificial sources external to the vehicle, e.g. navigation beacons from positioning sensors located off-board the vehicle, e.g. from cameras
G05D 1/46 - Control of position or course in three dimensions
Xi'an Thermal Power Research Institute Co., Ltd (China)
Northwestern Polytechnical University (China)
Inventor
Jing, Xiaolei
Liu, Cunliang
Lin, Yuwen
Xu, Dangqi
Zhao, Liang
Liang, Shuting
Abstract
A device for isothermal compression, constant-pressure power generation and physical energy storage includes an air storage tank, a weight, a piston and a piston rod. An inner cavity of the air storage tank is divided into first to third chambers by first and second heat conducting baffles. The piston is arranged in the second chamber. The piston rod has a lower end connected with the piston and an upper end connected with the weight. First and second elastic sealing belts are arranged in the first and third chambers, respectively. The first chamber includes a first water injection port and a first water outlet above the first elastic sealing belt, and the third chamber includes a second water injection port and a second water outlet above the second elastic sealing belt. A bottom of the air storage tank includes an air injection port and a compressed air outlet.
Xi'an Thermal Power Research Institute Co., Ltd (China)
Northwestern Polytechnical University (China)
Inventor
Jing, Xiaolei
Liu, Cunliang
Lin, Yuwen
Xu, Dangqi
Zhao, Liang
Liang, Shuting
Abstract
An air energy storage system for deep level cascade utilization of energy includes a compressor unit, an air storage chamber, a molten salt heat exchanger and a water source heat exchanger. All stages of compressors in the compressor unit are connected in series, a compressed air outlet of the last stage of compressor is connected with an air inlet of the air storage chamber, a pipeline at a compressed air outlet of each stage of compressor in the compressor unit is sequentially provided with the molten salt heat exchanger and the water source heat exchanger along a flow direction of a compressed air, and a hot end of the molten salt heat exchanger and a hot end of the water source heat exchanger are connected to the pipeline at the compressed air outlet of the compressor.
Disclosed in the present invention is an uncertainty visualization method based on representation sampling and for high-dimensional complex ensemble data. The method comprises: first transforming ensemble data into a uniformly distributed space by using a CVT algorithm; then, performing, by using a weighted sample elimination algorithm with a Poisson disk property, random uniform sampling on the ensemble data, which has been transformed into the uniformly distributed space, so as to obtain a representative subset; and finally, comparing a full ensemble with sampled subset data by using a radial basis function (RBF) interpolation algorithm, so as to obtain the difference in a data depth, and thereby verifying the effectiveness of a representation sampling technique. In the present invention, the complexity of ensemble data can be reduced, and the statistical distribution of original ensemble data is reserved, such that the uncertainty of a bottom layer can be conveyed more conveniently, and the visualization of the uncertainty of high-dimensional and high-complexity ensemble data can also be dealt with.
The present application provides a vision and language navigation method and apparatus based on an inference chain autonomous evolution strategy; in an embodiment of the present application, a large language model is deployed on a cloud server, and a plurality of program modules for assisting in completing a vision and language navigation task are deployed at an intelligent robot-side, wherein the intelligent robot can send to the cloud server a target natural language instruction input by a user and function description information of various program modules deployed on the intelligent robot; By means of the large language model, the cloud server can determine various target program modules that need to be scheduled and executed to execute a vision and language navigation task corresponding to the target natural language instruction, and can send a module scheduling sequence containing the target program modules to the intelligent robot, enabling the intelligent robot to sequentially schedule and execute the target program modules to complete the vision and language navigation task.
The embodiments of the present disclosure relate to a coarse-to-fine heterologous image matching method based on edge guidance. In the present disclosure, a heterogeneous feature extraction module based on subspace projection can effectively enhance the fusion of features and the extraction of cross-modal information while processing multi-scale feature spatial information; a global information integration module models global dependency between adjacent layers, and performs multi-level interaction between different areas so as to pay attention to a specific area in more detail and establish connections between different areas, thereby generating coarse-grained features with higher discriminability and accuracy; a coarse matching module uses a bidirectional softmax function to process coarse-grained attention features, so as to generate a confidence matrix; and a local feature refinement-fine regression module designs a local feature window and extracts fine-grained features by means of an encoder, and a fine regression sub-module readjusts a prediction result of coarse matching, and finally realizes accurate heterologous image feature matching. The method can perform coarse-to-fine alignment on heterologous images, thereby improving the matching precision.
G06V 10/75 - Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video featuresCoarse-fine approaches, e.g. multi-scale approachesImage or video pattern matchingProximity measures in feature spaces using context analysisSelection of dictionaries
12.
STEM CELL, PREPARATION METHOD THEREFOR AND USE THEREOF, AND METHOD FOR TREATING BONE DEFECT
A stem cell, a preparation method therefor and a use thereof, and a method for treating a bone defect. A new stem cell is identified and separated, and the stem cell has strong osteogenesis and chondrogenic differentiation capabilities, is stronger than traditional bone mesenchymal stem cells in respect of osteogenic capability, and can thus be used for bone defect repair of animals or humans. The stem cell can be used for effectively and rapidly treating a bone defect, thereby shortening the bone repair time, and remarkably improving a prognosis effect.
Provided are an α-SiAlON porous ceramic, and a preparation method and use thereof. The α-SiAlON porous ceramic has a dielectric constant of 1.19 to 3.21 at 12 GHz, a dielectric loss of 0.33×10−3 to 11.17×10−3, a thermal conductivity of 0.31 W/(m·K) to 0.81 W/(m·K) at room temperature, a thermal conductivity of 0.14 W/(m·K) to 0.59 W/(m·K) at a temperature of 1,500° C., and a bending strength of 72.4 MPa to 184.4 MPa.
C04B 35/597 - Shaped ceramic products characterised by their compositionCeramic compositionsProcessing powders of inorganic compounds preparatory to the manufacturing of ceramic products based on non-oxides based on borides, nitrides or silicides based on silicon oxynitrides
C04B 38/00 - Porous mortars, concrete, artificial stone or ceramic warePreparation thereof
C04B 38/02 - Porous mortars, concrete, artificial stone or ceramic warePreparation thereof by adding chemical blowing agents
14.
Nickel-based superalloy formed by selective laser melting and preparation method thereof
Disclosed are a nickel-based superalloy formed by selective laser melting and a preparation method thereof. In the method, CrFeNb alloy powder is used as a grain refiner, and its element composition is within the composition range of a nickel-based superalloy powder to ensure that the prepared nickel-based superalloy has the same element composition with the original alloy; the grain size in the nickel-based superalloy could be refined by the addition of CrFeNb alloy powder, such that the anisotropic columnar grain structure in the alloy is transformed to equiaxed grain structure, thereby improving mechanical properties of the alloy.
B22F 3/105 - Sintering only by using electric current, laser radiation or plasma
B22F 9/04 - Making metallic powder or suspensions thereofApparatus or devices specially adapted therefor using physical processes starting from solid material, e.g. by crushing, grinding or milling
C22C 19/05 - Alloys based on nickel or cobalt based on nickel with chromium
15.
PEDESTRIAN ATTRIBUTE CROSS-MODAL ALIGNMENT METHOD BASED ON COMPLETE ATTRIBUTE IDENTIFICATION ENHANCEMENT
Disclosed in the present invention is a pedestrian attribute cross-modal alignment method based on complete attribute identification enhancement. In the method, a statement structure analysis policy is first introduced to extract attributes included in pedestrians, and high-frequency attributes among the attributes are selected to construct a high-frequency attribute vocabulary; then, an attribute identification and enhancement module based on the high-frequency attribute vocabulary is designed, and complementary context prediction and attribute-level prediction are also performed; next, an attribute complete learning module is designed to determine the corresponding positions of the high-frequency attributes in a feature map and perform screening, the remaining attributes are further regarded as low-frequency attributes, and cross-modal alignment learning is performed between the high-frequency attributes and global context features of another modality and between the low-frequency attributes and the global context features of another modality, respectively; and finally, feature vectors which have a higher discrimination performance and can better model high-frequency and low-frequency attributes at the same time are obtained. The present invention can effectively solve the problem of the accurate semantic alignment of pedestrian attributes in a natural language-based cross-modal pedestrian re-identification task in the case of similar appearances.
Disclosed in the present invention is a key data extraction method driven by user demand. The method comprises: inputting user demand text into a preset text encoder model to extract a text feature vector; inputting each image in a query image data set into an image encoder so as to generate an image feature vector, and combining the obtained image feature vector with an image statistical feature so as to obtain an image quality score; calculating a relevance score sim between each image and the text; multiplying the image quality score by the relevance score, so as to obtain a quality-weighted relevance score; discarding image data, the relevance score of which is less than a threshold value; and clustering the remaining image data, and taking as key data the image closest to a clustering center. By means of the present invention, outputted key data is closely related to user demand, and also has a higher quality itself, and thus same can better meet the user demand, thereby reducing the adverse effect of low-quality data on user decision-making.
G06F 16/532 - Query formulation, e.g. graphical querying
G06V 10/40 - Extraction of image or video features
G06V 10/762 - Arrangements for image or video recognition or understanding using pattern recognition or machine learning using clustering, e.g. of similar faces in social networks
A concentric circular microphone array (CCMA) may include a number of omnidirectional microphones and an equal number of directional microphones, wherein the omnidirectional microphones and the directional microphones form a plurality of concentric rings on a substantially planar platform. Each of the plurality of concentric rings includes a subset of the omnidirectional microphones and a subset of the directional microphones (e.g., arranged in mixed pairs of microphones). Responsive to a sound source, the omnidirectional microphones and the directional microphones may respectively generate first and second electronic signals. A target beampattern of N thorder may be specified for the CCMA. An N th order beamformer for the CCMA, that is steerable in a three-dimensional space including the sound source, may be determined based on the specified target beampattern. The beamformer may be executed to calculate an estimate of the sound source based on the first electronic signals and the second electronic signals.
H04R 1/40 - Arrangements for obtaining desired frequency or directional characteristics for obtaining desired directional characteristic only by combining a number of identical transducers
18.
KEY VIDEO DATA EXTRACTION METHOD BASED ON MULTI-DIMENSIONAL SEMANTIC INFORMATION
A key video data extraction method based on multi-dimensional semantic information. The method comprises: first, carrying out time domain sampling and preprocessing on an inputted video; then constructing a video background based on a Gaussian mixture model; next, in a non-background area, using a single-stage target detection network to extract and screen key targets in a video frame; using a target tracking algorithm to track the key targets and thus obtain a target bounding box sequence; calculating target motion information, calculating a quality score of an image block in each tracking bounding box, and selecting the image block having the largest quality score as a typical target image; using a target fine-grained attribute extraction model to extract target color and model sub-class information; using a Transformer-based video description generation model to generate a textual abstract of the key targets; and finally, constructing a key target multi-dimensional representation structure, and storing as key data the video background and multi-dimensional representations of all targets. In the invention, required storage space can be greatly reduced and data information density is improved.
Disclosed in the present invention is a three-dimensional registration reconstruction method based on a multi-domain multi-dimensional feature map. The method comprises: first, randomly selecting a specified number of key points from a scene point cloud and a model point cloud; then, traversing all the key points, and calculating a corresponding multi-domain multi-dimensional feature map as a local feature descriptor; acquiring correspondences between the key points according to the multi-domain multi-dimensional feature map, and sorting key point pairs according to matching degrees; and in each iteration, selecting a compatible triple in sequence to generate a hypothesis, and verifying whether a model can be correctly registered in a scene by using the hypothesis. The present invention achieves a relatively short calculation time and a relatively low time cost.
A system and method of generating binaural signals includes receiving, by a processing device, a sound signal including speech and noise components, and transforming, by the processing device using a deep neural network (DNN), the sound signal into a first signal and a second signal. The transforming further includes encoding, by an encoding layer of the DNN, the sound signal into a sound signal representation in a latent space, rendering, by a rendering layer of the DNN, the sound signal representation into a first signal representation and a second signal representation in the latent space, and decoding, by a decoding layer of the DNN, the first signal representation into the first signal and the second signal representation into the second signal.
H04S 7/00 - Indicating arrangementsControl arrangements, e.g. balance control
G10L 19/008 - Multichannel audio signal coding or decoding using interchannel correlation to reduce redundancy, e.g. joint-stereo, intensity-coding or matrixing
G10L 25/30 - Speech or voice analysis techniques not restricted to a single one of groups characterised by the analysis technique using neural networks
Disclosed in the present invention is a person image re-identification method based on autonomous model structure evolution. A main structure is divided into three functional modules: a basic feature extraction module, a self-evolution multi-scale feature enhancement module and a multi-part fine-grained alignment module. In the basic feature extraction module, a basic feature of an image is extracted by using a visual convolutional neural network; thereafter, the self-evolution multi-scale feature enhancement module is inserted into the basic feature extraction module, a dynamic routing mechanism is used for being responsible for multi-scale visual feature enhancement, and by using deep and shallow features extracted by the basic feature extraction module, and by means of layer-by-layer scale transform and feature refining, a model is guided to learn how to evolve from a basic visual feature to an adaptive multi-scale feature; and finally, in the multi-part fine-grained alignment module, the model is guided to extract person features of multiple parts by using external semantic knowledge, and the matching of local features is realized. By means of the present invention, the accuracy of a local person image retrieval task can be improved.
Disclosed in the present invention is a fine-grained target recognition method based on a parameter self-evolution policy. The method comprises: constructing a parameter self-evolution module, and sending an input feature into a dynamic convolution of 1*1 size for feature extraction, wherein parameters of the dynamic convolution are generated according to features obtained after self-attention encoding fusion; inserting the parameter self-evolution module into a ResNet-50 model, and naming the parameter self-evolution module as a parameter self-evolution network; and finally, training the parameter self-evolution network by means of using a training set, and realizing fine-grained target recognition after the training is completed. By means of the present invention, backbone network model parameters can be dynamically updated, and parameter self-evolution upon facing unseen data domains is realized, such that generalizable fine-grained target recognition is realized.
A method for preparing an asymmetric wettable polyimide fiber-based photothermal aerogel is provided. The method includes the steps: uniformly mixing polyimide powder and a solvent, then, performing electrostatic spinning, and cutting an obtained fiber felt into pieces for later use; mixing the broken fibers, polyamic acid and tert-butyl alcohol, then, performing shearing to form a stable dispersion liquid for low-temperature directional freezing, and performing freeze-drying and high-temperature thermal imidization to obtain a polyimide fiber-based aerogel material; and soaking the above aerogel material in a hydrophilic monomer solution for a polymerization reaction, and then performing low-temperature directional freezing and freeze-drying to obtain a hydrophilic polyimide fiber-based aerogel. The aerogel is placed under light source irradiation, and dropwise coating is performed on an upper surface of the aerogel with a hydrophobic filler resin mixed solution to obtain the asymmetric wettable fiber-based photothermal aerogel.
A spectral cross-domain transfer super-resolution reconstruction method for a multi-domain image. By means of a spectral image cross-domain transfer super-resolution reconstruction method, which is based on cross-domain transferable knowledge learning and rapid target-domain adaptation learning, for a multi-domain image scenario, spectral super-resolution reconstruction from an RGB image to a hyperspectral image is realized. A model structure design based on a transferable dictionary is used to learn cross-domain transferable features; a source-domain pre-training policy based on a shared learnable mask is used to facilitate a model in learning general knowledge for reconstruction; and a model-agnostic meta-learning fine-tuning method is used to learn a universal model with a strong generalization capability, such that the model can adapt to data of a target domain of a test by means of several iterations of test data. The method can mine cross-domain shared knowledge to improve the generalization capability, thereby improving the effect of spectral cross-domain super-resolution reconstruction.
Surface pressure measurements on rotating models are important for flow phenomenon identification, understanding flow mechanisms and model aerodynamic design. The disclosure discloses a periodic pressure field measurement system based on the superposed lifetime of pressure sensitive paint, including pressure sensitive paint, a pulse light source, a camera, a synchronizer and a computer. A lifetime superposition method is disclosed in the disclosure for measuring periodic pressure fields of pressure sensitive paint. The disclosure has the beneficial effects: the disclosure acquires, on the basis of a relationship between the fluorescence lifetime of the pressure sensitive paint and the pressure, fluorescence image pair sequence of the pressure sensitive paint with a high signal to noise ratio under a high-frequency pulsating pressure through the strobe light source and the low-frame-rate CCD camera, and obtains the global dynamic pressure distribution according to the measurement principle of the lifetime method, which can effectively reduce system errors.
H04N 25/773 - Pixel circuitry, e.g. memories, A/D converters, pixel amplifiers, shared circuits or shared components comprising A/D, V/T, V/F, I/T or I/F converters comprising photon counting circuits, e.g. single photon detection [SPD] or single photon avalanche diodes [SPAD]
G01L 27/00 - Testing or calibrating of apparatus for measuring fluid pressure
H04N 25/40 - Extracting pixel data from image sensors by controlling scanning circuits, e.g. by modifying the number of pixels sampled or to be sampled
The disclosure discloses a global pressure acquisition system for a rotating model, including a CCD camera, a signal generator, a stroboscopic pulse LED light source, a photoelectric sensor, a preset counter, and a controller. The disclosure further discloses a global pressure acquisition method for a rotating model. The disclosure has the beneficial effects that, by using the non-contact measuring method disclosed by the present disclosure, a measured model and a flow field are not damaged, a submillimeter-level spatial resolution is achieved, it is ensured that the acquired PSP image of the rotating model is clear, and the signal to noise ratio of the image is increased.
G01L 1/24 - Measuring force or stress, in general by measuring variations of optical properties of material when it is stressed, e.g. by photoelastic stress analysis
27.
MEASURING DEVICE AND METHOD FOR DYNAMIC CHARACTERISTICS OF PRESSURE SENSITIVE PAINT
The disclosure discloses a measuring device for dynamic characteristics of pressure sensitive paint, including a PSP sample wafer, a dynamic pressure sensor, an oscilloscope, a light source, a photomultiplier tube, a bandpass filter, a loudspeaker, a power amplifier, and a signal generator. The disclosure further discloses a measuring method for dynamic characteristics of pressure sensitive paint. The disclosure has the following beneficial effects: by using the measuring device disclosed by the present disclosure, the continuous sinusoidal pressure wave with any frequency can be generated, and has the frequency precision that does not exceed 0.01 Hz, an optical path is not shielded, high stability is achieved, and the amplitude and phase characteristics of the pressure frequency of the PSP can be precisely captured.
G01L 25/00 - Testing or calibrating of apparatus for measuring force, torque, work, mechanical power, or mechanical efficiency
G01N 21/01 - Arrangements or apparatus for facilitating the optical investigation
G01N 21/71 - Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light thermally excited
28.
METHOD FOR SCREENING MOBILE TERMINAL VISUAL ATTENTION ABNORMALITIES IN CHILDREN BASED ON MULTIMODAL DATA LEARNING
The present invention relates to a method for screening mobile terminal visual attention abnormalities in children based on multimodal data learning. A calibration video and a testing video are set up, and a head-face video of children while watching the calibration video and the testing video on smartphones is recorded, respectively. An eye-tracking estimation model is constructed to predict the fixation point location from the head-face video corresponding to the testing video frame by frame and to extract the eye-tracking features. Facial expression features and head posture features are extracted. A Long Short-Term Memory (LSTM) network is used to fuse different modal features and realize the mapping from multimodal features to category labels. In the testing stage, the head-face video of children to be classified while watching the videos on smartphones is recorded, and the features are extracted and input into the post-training model to determine whether they are abnormal.
A61B 5/16 - Devices for psychotechnicsTesting reaction times
G06T 7/73 - Determining position or orientation of objects or cameras using feature-based methods
G06T 7/80 - Analysis of captured images to determine intrinsic or extrinsic camera parameters, i.e. camera calibration
G06V 10/26 - Segmentation of patterns in the image fieldCutting or merging of image elements to establish the pattern region, e.g. clustering-based techniquesDetection of occlusion
G06V 10/77 - Processing image or video features in feature spacesArrangements for image or video recognition or understanding using pattern recognition or machine learning using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]Blind source separation
G06V 10/80 - Fusion, i.e. combining data from various sources at the sensor level, preprocessing level, feature extraction level or classification level
G06V 10/82 - Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
G06V 20/40 - ScenesScene-specific elements in video content
G06V 40/16 - Human faces, e.g. facial parts, sketches or expressions
G06V 40/18 - Eye characteristics, e.g. of the iris
G06V 40/20 - Movements or behaviour, e.g. gesture recognition
29.
MN-MOF-BASED COLD-ADAPTED NANOZYME AND PREPARATION METHOD THEREOF
A preparation method of a Mn-MOF-based cold-adapted nanozyme and a preparation method thereof. The method includes: preparing a Mn(CH3COO)3·2H2O solution by fully dissolving a manganese-containing precursor Mn(CH3COO)3·2H2O in a mixed solution of an alcohol and distilled water; wherein the alcohol is ethanol or methanol; preparing a benzenetricarboxylic acid solution by fully dissolving benzenetricarboxylic acid solid in the mixed solution of the alcohol and distilled water; reacting the Mn(CH3COO)3·2H2O solution with the benzenetricarboxylic acid solution sufficiently using co-precipitation; and centrifuging and removing a supernatant to obtain the MnBTC, where the MnBTC is nano-metal-organic framework (nano-MOF) with a particle size of less than 10 nm.
Disclosed in the present invention is a referring target detection and positioning method based on dynamic adaptive reasoning. A DarkNet pre-training model based on a convolutional neural network is used for an image and a BERT pre-training model is used for text, so as to respectively extract picture and language representations; feature fusion is performed on image and text information by using a multi-modal fusion attention mechanism; and finally dynamic adaptive reasoning is performed by using a reinforcement learning reward mechanism algorithm, thus detecting and positioning the location of a referring target in the image. By means of the present invention, a higher accuracy and a faster operation speed are obtained, and compared with previous models, the models in the present invention have made prominent progress in terms of precision and speed.
Disclosed is a chip-level disc-type acousto-optic standing wave gyroscope including a substrate and a gyroscope structure placed on an upper surface of the substrate; the substrate is in a shape of a circular disc; the gyroscope structure includes an acoustic wave drive module and an optical detection module, the acoustic wave drive module is arranged in a circular shape taking the center of the circular disc as an origin and extending outward radially, and the optical detection module is arranged in the middle of the acoustic wave drive module and is annular; the acoustic wave drive module includes an annular interdigitated transducer, a metal electrode layer group uniformly sputtered on the annular interdigitated transducer, annularly arranged metallic pillars and an annular reflection grating, respectively placed in sequence from center of the disk radially to periphery of the disk; the optical detection module includes a first grating coupler, an optical waveguide at a light source input end, a first coupler, a second coupler, an optical waveguide at a signal output end and a second grating coupler, which are connected in sequence. According to the technical solution of the disclosure, the sensitivity of gyroscope detection can be improved.
G01C 19/72 - Gyrometers using the Sagnac effect, i.e. rotation-induced shifts between counter-rotating electromagnetic beams with counter-rotating light beams in a passive ring, e.g. fibre laser gyrometers
32.
DYNAMIC SCENE STRUCTURE ESTIMATION METHOD BASED ON MULTI-DOMAIN SPATIO-TEMPORAL DATA, AND DEVICE AND STORAGE MEDIUM
The embodiments of the present application relate to the technical fields of three-dimensional computer vision and three-dimensional surveying and mapping. Provided are a dynamic scene structure estimation method based on multi-domain spatio-temporal data, and a device and a storage medium. In the dynamic scene structure estimation solution based on multi-domain spatio-temporal data provided in the present application, a target scene is photographed from different spatial domains, such that a plurality of groups of image sequences are obtained; then, the plurality of groups of image sequences are coded, target discrete three-dimensional meshes of the scene at respective resolutions in different domains are respectively estimated therefrom, and the target discrete three-dimensional meshes are aligned and fused, such that a target discrete three-dimensional mesh of the target scene is obtained; and the offset of the target discrete three-dimensional mesh at each moment is further estimated, thereby realizing the three-dimensional structure estimation of a dynamic scene.
This invention relates to a data dimension reduction method based on maximizing a ratio sum for linear discriminant analysis, which belongs to the fields of image classification and pattern recognition. It includes constructing a data matrix, a label vector and a label matrix; calculating a within-class covariance matrix and a between-class covariance matrix; constructing the optimization problem based on maximizing the ratio sum for the linear discriminant analysis; using the alternating direction method of multipliers to obtain the projection matrix which can maximize an objective function. This invention establishes the objective function based on maximizing the ratio sum for the linear discriminant analysis to avoid the problem that the traditional linear discriminant analysis tends to select features with small variances and weak discriminating ability. It can select features which are more conducive to classification. Moreover, this method does not depend on the calculation of the inverse matrix of the within-class covariance matrix and does not require data preprocessing, which improves the adaptability of the data dimensionality reduction method to the original data feature.
G06V 10/34 - Smoothing or thinning of the patternMorphological operationsSkeletonisation
G06V 10/764 - Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
G06V 10/774 - Generating sets of training patternsBootstrap methods, e.g. bagging or boosting
34.
ENERGY-EFFICIENT SAMPLE SELECTION METHOD BASED ON COMPLEXITY
The invention discloses an energy-efficient sample selection method based on sample complexity, which performs sample selection on the raw data sets through two stages of inter-class sampling and intra-class sampling, the object is to select representative samples from large-scale data sets, thereby reducing the number of samples used for model training and achieving the object of lightweight training. Compared with the prior art, the invention has the following advantages: the invention proposes an energy-efficient sample selection method based on complexity, selects representative samples from large-scale datasets for efficient model training, and proves that sample complexity and model training strategies have a very important impact on the efficient training of deep neural networks. The invention also solves the problem of low efficiency of model training based on sample complexity and model training strategies, which has certain significance for alleviating the problem of low efficiency of deep learning model training.
xslxslxsll; and adjusting pixels of the low-illumination remote sensing image one by one according to the brightness enhancement curve. According to the present invention, a problem in the prior art of low high-dynamic reconstruction precision of low-illumination remote sensing images is solved, and the high-dynamic reconstruction precision of the low-illumination remote sensing images is improved.
The present invention relates to the technical field of special materials for aluminum alloy additive manufacturing, and provides a special high-strength aluminum alloy for SLM and an SLM forming method therefor. According to the special high-strength aluminum alloy for SLM provided by the present invention, the cracking tendency of the alloy is greatly reduced under the synergistic effect of elements Si and Ti, and meanwhile, a double-reinforced relative Al-Cu-Mg series alloy is introduced under the synergistic effect of the elements Si and Ti for remarkable reinforcement, such that the special high-strength aluminum alloy is more suitable for an SLM process, and the material system for aluminum alloy additive manufacturing is enriched. Experimental results show that the yield strength of the high-strength aluminum alloy prepared from the special high-strength aluminum alloy for SLM, provided by the present invention, subjected to SLM forming can reach 465-481 MPa, the tensile strength can reach 539-543 MPa, the elongation can reach 9.7-12.1%, the density can reach 99.99%, and no defect such as cracks exists.
The present disclosure provides a successive gas path fault diagnosis method with high precision for gas turbine engines and falls within the technical field of fault diagnosis for gas turbine engines, including the following steps: establishing an engine nonlinear component-level model; capturing dynamic effects of an engine transient maneuver; outputting an estimated value of an engine observation parameter by the engine nonlinear component-level model; acquiring a measurement of the engine observation parameter through sensors; and iteratively updating a degradation factor through a solver. The present disclosure captures the dynamic effects of the transient maneuver at consecutive moments through time-series gas path measurement parameters, thereby realizing successive and high-precision diagnosis for health conditions of the gas turbine engines. This technology can provide a new successive and high-precision diagnosis method for the gas turbine engines under steady-state and transient conditions.
RESEARCH & DEVELOPMENT INSTITUTE OF NORTHWESTERN POLYTECHNICAL UNIVERSITY IN SHE (China)
NORTHWESTERN POLYTECHNICAL UNIVERSITY (China)
Inventor
Zhang, Xianggang
Cai, Zhuochen
Yin, Ziang
Wang, Tao
Zhao, Qinghua
Abstract
26255-LiBr, respectively preparing a compact polycrystalline material in accordance with the stoichiometric ratio of CLLB and a solvent region compact polycrystalline material having a raw material ratio that allows a CLLB crystal to be directly precipitated; then adjusting a specific temperature field shape to simply melt the solvent region polycrystalline material; dissolving the polycrystalline material in accordance with the stoichiometric ratio of CLLB at a dissolving interface during melt diffusion and convection, and performing precipitation at a growth interface; slowly moving the solvent region upwards along with the relative movement of a furnace body and the crystal using the THM method; and finally melting all the polycrystalline material in accordance with the stoichiometric ratio of CLLB, and allowing same to pass through the solvent region, followed by precipitation and crystallization at the growth interface.
C30B 9/06 - Single-crystal growth from melt solutions using molten solvents by cooling of the solution using as solvent a component of the crystal composition
G01T 1/202 - Measuring radiation intensity with scintillation detectors the detector being a crystal
39.
First-order differential microphone array with steerable beamformer
A first-order differential microphone array (FODMA) with a steerable beamformer is constructed by specifying a target beampattern for the FODMA at a steering angle θ and then decomposing the target beampattern into a first sub-beampattern and a second sub-beampattern based on the steering angle θ. A first sub-beamformer and a second sub-beamformer are generated to each filter signals from microphones of the FODMA, wherein the first sub-beamformer is associated with the first sub-beampattern, and the second sub-beamformer is associated with the second sub-beampattern. The steerable beamformer is then generated based on the first sub-beamformer and the second sub-beamformer. The decomposing of the target beampattern into a first sub-beampattern and a second sub-beampattern includes dividing the target beampattern into a sum of a first-order cosine (cardioid) first sub-beampattern and a first-order sinusoidal (dipole) second sub-beampattern.
H04R 1/40 - Arrangements for obtaining desired frequency or directional characteristics for obtaining desired directional characteristic only by combining a number of identical transducers
A feature extraction deep neural network (DNN) may be trained based on the minimization of a loss function. A similarity function may be specified to calculate a similarity score for two representations of verbal utterances. A training data set comprising pairs of representations of utterances is received, wherein each one of the pairs of representations of utterances is associated with a corresponding a ground-truth label confirming whether the pair of represented utterances come from a same speaker or not. A respective similarity score may then be calculated for each one of the pairs of representations of utterances. Parameters associated with the DNN may then be updated based on minimizing a loss function associated with an area under a section of a receiver-operating-characteristic (ROC) curve for the similarity scores, wherein the ROC curve section is delimited between a low false positive rate (FPR) value and a high FPR value.
G10L 17/02 - Preprocessing operations, e.g. segment selectionPattern representation or modelling, e.g. based on linear discriminant analysis [LDA] or principal componentsFeature selection or extraction
G10L 17/08 - Use of distortion metrics or a particular distance between probe pattern and reference templates
41.
MULTI-RESOLUTION ENSEMBLE SELF-TRAINING-BASED TARGET DETECTION METHOD FOR SMALL-SAMPLE LOW-QUALITY IMAGE
A multi-resolution ensemble self-training-based target detection method for a small-sample low-quality image, relating to the technical field of image processing. First, preliminary training is performed on a target detection model by means of labeled data, then prediction is performed on unlabeled data by means of the trained model, then the data having experienced the prediction is added into original data, to train the target detection model again, and so on, and iterative updating is continuously performed to obtain a final detection model; moreover, a multi-resolution ensemble self-training mode is used each time the latest model is used for prediction of unlabeled data. According to the method, labeled low-quality image data and unlabeled low-quality image data are effectively combined, and the precision of target detection for small-sample low-quality images is improved.
An annular coupling system suitable for an MEMS modal localization sensor. The main structure of the system comprises an annular coupling beam (203) and coupling stiffness adjusting electrodes (205, 206). The annular coupling beam (203) is a circular ring, a square ring, a rectangular ring, or other closed structures; and the annular coupling beam (203) achieves mechanical coupling by connecting two resonant beams (201, 202). Compared with conventional mechanical coupling beams, the annular coupling system can enable the sensitivity degree of the coupling beam to lateral etching to be greatly reduced, and the stability and consistency of the sensor can be further improved in the case that the processing precision level is not changed; moreover, the coupling stiffness adjusting electrodes (205, 206) for the annular coupling beam (203) are designed, a potential difference between the coupling stiffness adjusting electrodes (205, 206) and the annular coupling beam (203) is generated by adjusting the potential of the coupling stiffness adjusting electrodes (205, 206), so that electrostatic force is generated, stress distribution inside the annular coupling beam (203) is changed, and adjustment of the mechanical coupling stiffness is achieved.
G01P 15/08 - Measuring accelerationMeasuring decelerationMeasuring shock, i.e. sudden change of acceleration by making use of inertia forces with conversion into electric or magnetic values
G01P 15/097 - Measuring accelerationMeasuring decelerationMeasuring shock, i.e. sudden change of acceleration by making use of inertia forces with conversion into electric or magnetic values by vibratory elements
43.
NON-COOPERATIVE TARGET THREE-DIMENSIONAL RECONSTRUCTION METHOD BASED ON BRANCH RECONSTRUCTION REGISTRATION
The present invention relates to a non-cooperative target three-dimensional reconstruction method based on branch reconstruction registration, and belongs to the field of computer vision. The method comprises: when multi-angle image sequences are given, first classifying the image sequences, and obtaining, by using a structure-from-motion algorithm, three-dimensional point cloud data corresponding to various image sequences; and then performing scale unification on various point cloud data, and then performing registration reconstruction on various point clouds by using a point cloud registration algorithm, so as to realize three-dimensional reconstruction of a spatial non-cooperative target. In the present invention, image sequences are classified and then reconstructed in parallel, such that the time consumption in a reconstruction process is reduced, the reconstruction efficiency is improved, and the real-time requirements of spacecraft operation can be met. The various reconstructed point cloud data is registered, the finally reconstructed point cloud is denser, and the reconstruction accuracy is improved.
A high-performance, iron-based, medium-entropy alloy, a preparation method therefor and an application thereof, relating to the technical field of metal materials. The alloy comprises the following components by mole percentage: 12-20 at% Al; 8-12 at% Cr; 35-55 at% Fe; and 25-45 at% Ni.
The present invention relates to the technical field of electrolytic machining in laser additive manufacturing, and specifically relates to a method for improving the surface quality of an alloy micro-region by a salt film method and an application. The method comprises: performing electrolytic machining by using saturated nickel chloride and sodium chloride ethylene glycol mixed electrolyte. By forming a supersaturated salt film on a workpiece surface, the micro-area quality of a workpiece surface is thereby further improved, and the service life of an alloy device is improved.
Source localization method for rumor source based on full-order neighbor coverage strategy includes: constructing a network graph according to the user relationship in the actual target area; mapping an actual relationship into the network graph; determining sensors in the network graph, and deploying users corresponding to the sensors as observation users in an actual target area; executing a source inferring strategy when the number of the observation users in the actual target area who have received the rumor reaches an expected scale; calculating source likelihood score of non-sensor nodes in the network graph corresponding to the non-observation users in the actual target area; processing differentially the source likelihood scores; and outputting the non-observation user corresponding to the minimum source likelihood score as the source.
2 ternary eutectic ceramic coating. The eutectic ceramic thermal barrier material provided by the present disclosure has good high temperature resistance, good oxidation resistance and excellent mechanical properties.
C23C 24/10 - Coating starting from inorganic powder by application of heat or pressure and heat with intermediate formation of a liquid phase in the layer
48.
SYSTEM AND METHOD TO USE DEEP NEURAL NETWORK TO GENERATE HIGH-INTELLIGIBILITY BINAURAL SPEECH SIGNALS FROM SINGLE INPUT
A system and method of generating binaural signals includes receiving, by a processing device, a sound signal including speech and noise components (104), and transforming, by the processing device using a deep neural network(DNN), the sound signal into a first signal and a second signal (106). The transforming further includes encoding, by an encoding layer of the DNN, the sound signal into a sound signal representation in a latent space (108), rendering, by a rendering layer of the DNN, the sound signal representation into a first signal representation and a second signal representation in the latent space (110), and decoding, by a decoding layer of the DNN, the first signal representation into the first signal and the second signal representation into the second signal (112).
A chip-scale disc-type acousto-optic standing-wave gyroscope, comprising a substrate and a gyroscope structure placed on an upper surface of the substate; the substrate is disc-shaped; the gyroscope structure comprises an acoustic wave driving module and an optical detection module; the acoustic wave driving module is arranged to be circular outwards in a radial direction using a center of circle of the disc as the origin; the optical detection module is provided in the middle of the acoustic wave driving module and is in the shape of a circular ring; the acoustic wave driving module comprises an annular interdigital transducer, a metal electrode layer group uniformly sputtered on the annular interdigital transducer, an annular metal lattice (3), and an annular reflecting grating (4) which are sequentially arranged from the center of circle of the disc to the periphery in the radial direction; the optical detection module comprises a first grating coupler (5-1), a light source input end optical waveguide (6), a first coupler (8-1), a second coupler (8-2), a signal output end optical waveguide (7), and a second grating coupler (5-2) that are sequentially connected. The chip-scale disc-type acousto-optic standing-wave gyroscope can improve the sensitivity of gyroscope detection.
G01C 19/5656 - Turn-sensitive devices using vibrating masses, e.g. vibratory angular rate sensors based on Coriolis forces using vibrating bars or beams the devices involving a micromechanical structure
50.
MN-MOF-BASED COLD-ADAPTED NANO-ENZYME, PREPARATION METHOD THEREFOR AND USE THEREOF
A nano-Mn-MOF-based cold-adapted nano-enzyme (also referred to as a low-temperature resistant nano-enzyme), a preparation method therefor and use thereof. The nano-MOF mainly comprises nano-MIL-100 (Mn) and Mn-BTC, which are respectively prepared by a hydrothermal method and a co-sedimentation method. The nano-enzyme has more excellent enzymatic activity and cold adaptation characteristics.
The present invention relates to the technical field of electrochemical machining for laser additive manufacturing, and in particular, to a machining method for improving the surface quality of a micro-region of an alloy component. By using a nanosecond pulse electrochemical machining process, the surface treatment effect of the alloy component can be further improved, and a high-quality surface workpiece can be quickly and well obtained.
The present disclosure provides an air combat maneuvering method based on parallel self-play, including the steps of constructing a UAV (unmanned aerial vehicle) maneuver model, constructing a red-and-blue motion situation acquiring model to describe a relative combat situation of red and blue sides, constructing state spaces and action spaces of both red and blue sides and a reward function according to a Markov process, followed by constructing a maneuvering decision-making model structure based on a soft actor-critic (SAC) algorithm, training the SAC algorithm by performing air combat confrontations to realize parallel self-play, and finally testing a trained network, displaying combat trajectories and calculating a combat success rate. The level of confrontations can be effectively enhanced and the combat success rate of the decision-making model can be increased.
The disclosure relates to a method for generating personalized dialogue content, in which an implicit association between personalized characteristics and corresponding dialogue replies is extracted by collecting a set of personalized dialogue data; a vector representation of a dialogue context and texts of the personalized characteristics is learned with a Transformer model; finally, through learning a sequence dependency between natural languages, a subsequent content may be automatically predicted and generated from a previous text, so that the generating of corresponding reply content may be achieved according to the dialogue context. With various optimization algorithms added, a generation probability of universal reply can be reduced and a diversity of the generated dialogue content can be improved.
Disclosed is a chip-level resonant acousto-optic coupling solid-state wave gyroscope based on MEMS technology. A surface acoustic progressive wave mode sensitive structure and a micro-ring resonant cavity optical detection structure are combined in the gyroscope. Through acousto-optic effect, mechanical strain of the device crystal caused by wave vibration of a primary surface acoustic wave and a secondary surface acoustic wave caused by Coriolis force is converted into a variation in the refractive index of an optical waveguide etched on the device, so that the optical signal transmitted in the waveguide diffracts, thereby generating frequency modulation. Meanwhile, a micro-ring resonant cavity using the resonance principle peels off the frequency change introduced by the primary surface acoustic wave, and obtains an output signal containing external angular velocity information. Based on the proportional relationship between the detection resolution and the quality factor of the micro-ring resonant cavity, the order of magnitude of the interface detection resolution is improved, and the performance indicators of the gyroscope are simultaneously optimized in terms of improving sensitivity and resolution, and its precision is improved.
G01C 19/72 - Gyrometers using the Sagnac effect, i.e. rotation-induced shifts between counter-rotating electromagnetic beams with counter-rotating light beams in a passive ring, e.g. fibre laser gyrometers
55.
SEQUENTIAL RECOMMENDATION METHOD BASED ON LONG-TERM AND SHORT-TERM INTERESTS
This disclosure provides a sequential recommendation method based on long-term and short-term interests, in which an interaction sequence between a user and products is obtained by processing a purchase sequence of the user and a question data of the user in a dataset, characteristics of the products are represented with extracted comments of the user on the products; next, a stable long-term preference of the user is learned from a historical purchase sequence of the user with a recursive neural network, and immediate interests of the user are modeled with the question data. For the stable long-term preference and dynamic immediate interests, a dependence of different users on the two characteristics is described with an Attention mechanism, so as to effectively solve a problem of an inaccurate recommendation caused by an evolution of the preference of the user, while different dependence degrees of the different users on the long-term preference and immediate interests can represented effectively.
H04R 1/40 - Arrangements for obtaining desired frequency or directional characteristics for obtaining desired directional characteristic only by combining a number of identical transducers
The present invention relates to the fields of image recognition and classification and pattern recognition, and relates to an unsupervised data dimensionality reduction method based on adaptive nearest neighbor graph embedding. The method comprises: preprocessing data; constructing a nearest neighbor graph and initializing same; and optimizing an objective function by means of alternating iteration. The present invention further provides a face recognition method based on the data dimensionality reduction method, comprising: performing dimensionality reduction on a face image to obtain a projection matrix and low-dimensional data, and clustering the low-dimensional data by using an unsupervised clustering algorithm to obtain clustering centers of categories; and taking, according to the Euclidean distances between an image to be classified and the clustering centers, a clustering center having the minimum Euclidean distance, the category of the clustering center being the category of a new face image. Face recognition is performed in a low-dimensional space, so that the amount of data storage can be reduced, the amount of data calculation can be reduced, the calculation efficiency can be improved, and finally, the real-time performance and recognition precision of a face recognition technology can be improved.
The present invention belongs to the field of image classification and pattern recognition, and relates to a data dimensionality reduction method based on the maximum ratio and linear discriminant analysis. The method comprises: constructing a data matrix, a label vector and a label matrix; calculating an intra-class covariance matrix and an inter-class covariance matrix; constructing an optimization problem based on the maximum ratio and linear discriminant analysis; and by using an alternating optimization iterative algorithm, solving a projection matrix capable of maximizing an objective function. By means of the present invention, an objective function based on the maximum ratio and a linear discriminant analysis method is established, thereby solving the problem of traditional linear discriminant analysis whereby same tends to select features with a small variance and weak discrimination capability, such that features which are more beneficial to classification can be selected. The present invention is independent of calculation of an inverse matrix of an intra-class covariance matrix, data pre-processing is not required, and the adaptability of a data dimensionality reduction method to raw data features is improved.
A first-order differential microphone array (FODMA) with a steerable beamformer is constructed by specifying a target beampattern for the FODMA at a steering angle θ and then decomposing the target beampattern into a first sub-beampattern and a second sub-beampattern based on the steering angle θ. A first sub-beamformer and a second sub-beamformer are generated to each filter signals from microphones of the FODMA, wherein the first sub-beamformer is associated with the first sub-beampattern, and the second sub-beamformer is associated with the second sub-beampattern. The steerable beamformer is then generated based on the first sub-beamformer and the second sub-beamformer. The decomposing of the target beampattern into a first sub-beampattern and a second sub-beampattern includes dividing the target beampattem into a sum of a first-order cosine (cardioid) first sub-beampattern and a first-order sinusoidal (dipole) second sub-beampattern.
The disclosure provides an identification method based on an expert feedback mechanism, in which the expert properly give a feedback to results of a static model, the model is dynamically adjusted and updated according to the feedback of the expert each time, so that identifications for similar objects can be changed from a wrong identification to a correct identification. The model can adapt to dynamic changes of the environment, so that an identification accuracy and robustness of the model under the dynamic environment are improved with an expertise. The accuracy of the identification model is improved without repeated training, which solves a problem that the accuracy of the static model decreases in the dynamic environment, raising an adaptability of the identification model to environmental changes, shortening updating time of the model and improving working efficiency of the identification application system.
This disclosure provides a method for generating a personalized product description based on multi-source crowd data, which includes following steps: collecting data required for the personalized product description, the required data including reviews for crowd products and historical reviews of a crowd of users; portraiting the product and user to obtain a user preference label and a product label, which are then matched to obtain a personalized preference label; and generating the personalized product description in conjunction with the personalized preference labels. For different product attributes, different text generation methods are employed, and with different characteristics of the text generation methods such as extracted text generation and generated text generation, multi-source data are fused, so that the generated product description is smoother.
This disclosure provides an accurate and personalized recommendation method based on a knowledge graph, which includes following steps: acquiring relevant knowledge of objects from a knowledge base according to historical behaviors of a user, and constructing a knowledge graph; initializing a vector representation of each node and its connection, and determining a receptive field of the node; generating training samples according to the historical behaviors of the user, and initializing a vector representation of all users and objects; acquiring a receptive field of an entity in the knowledge graph corresponding to the object in the training sample, then inputting the receptive field and the training sample to a graph neural network model to obtain predicted values of a possibility of an interaction between the user and the object. According to the disclosure, a sparsity of the historical behavior information of the original user is compensated with the knowledge graph information, and the user and objects are depicted in multi-dimension, so that the personalized recommendation is more accurate.
The disclosure relates to a video anomaly detection method based on human-machine cooperation, in which video frames and traditional descriptors of optical stream of an image are utilized as an input for auto-encoder neural network coding, and converted into a representation content of a hidden layer, and then the representation content of the hidden layer is decoded, reconstructed and output. The auto-encoder network is trained with normal samples. In a test stage, if an input is a normal sample, a final reconstructed error keeps high similarity with an input sample; on the contrary, if the input is an abnormal sample, the final reconstructed error deviates greatly from the input sample.
G06V 10/778 - Active pattern-learning, e.g. online learning of image or video features
G06F 16/78 - Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
G06F 18/21 - Design or setup of recognition systems or techniquesExtraction of features in feature spaceBlind source separation
G06F 18/2411 - Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
A binaural beamformer comprising two beamforming filters may be communicatively coupled to a microphone array to generates two beamforming outputs, one for the left ear and the other for the right ear. The beamforming filters may be configured in such a way that they are orthogonal to each other to make white noise components in the binaural outputs substantially uncorrelated and desired signal components in the binaural outputs highly correlated. As a result, the human auditory system may better separate the desired signal from white noise and intelligibility of the desired signal may be improved.
G10K 11/178 - Methods or devices for protecting against, or for damping, noise or other acoustic waves in general using interference effectsMasking sound by electro-acoustically regenerating the original acoustic waves in anti-phase
The present disclosure relates to the technical field of unmanned aerial vehicle (UAV) tests, and more particularly, to an electromagnetic release device for use in vertical falling tests of tri-rotor UAVs and including a mounting frame and multiple clamping and release modules arranged on the mounting frame. movable kits, which include a ferromagnetic plate matching and connected with the electromagnetic adsorption assembly, one end of the ferromagnetic plate is hinged with the electromagnet mounting frame, and the other end of the ferromagnetic plate is connected with the UAV connecting plate; The present disclosure uses electromagnetic control to accurately control the simultaneous opening of three clamping and release modules of a UAV, realizes the release and landing of the UAV in a horizontal status, and is characterized by simple structure and easy operation.
An unsupervised data dimensionality reduction method based on noise suppression, comprising: initializing a global divergence matrix, a graph matrix, a projection matrix, a Laplacian matrix, and a regularization parameter, and then updating a matrix F, updating the projection matrix P, repeatedly iterating until a target function converges, and realizing unsupervised data dimensionality reduction. According to the method, the computing complexity is reduced, the computing time is reduced, and high-dimensional data can be quickly and effectively subjected to dimensionality reduction.
The present invention provides an artificial intelligence-assisted printed electronics self-guided optimization method, which integrates machine learning technology with printed electronics. According to variables that impact printing quality of a microelectronic printer, a user sets up experimental groups, prints samples with the microelectronic printer according to parameters in the experiment groups, characterizes printing effects, and evaluates the printing quality. The characterization result is analyzed by machine learning, and printing parameters that correspond to a best printing effect are obtained; then, the parameters are fed back to the user, and the user configures the printer according to the fed-back parameters, thereby improving printing quality. By using the present invention, optimal printing parameters can be obtained by simply setting up a few simple experiments according to a number of factors that impact printing effects, which reduces the time for a printer user to test out printing effects in an early stage, and provides a good practicability.
H05K 3/12 - Apparatus or processes for manufacturing printed circuits in which conductive material is applied to the insulating support in such a manner as to form the desired conductive pattern using printing techniques to apply the conductive material
The present disclosure relates to a method and system for determining transportation safety of pulverized coal. The method includes: acquiring coal particle data during transportation of pulverized coal, where the coal particle data is size data of a coal particle accumulation; determining a particle model of the coal particle accumulation during the transportation of the pulverized coal according to the size data; establishing a constitutive theoretical model to describe all flow regimes of a coal granular medium; numerically discretizing the constitutive theoretical model by using a numerical method to obtain discrete equations; calculating a movement process of the coal granular medium according to the discrete equations and the particle model of the coal granular medium to obtain a calculation result; plotting the calculation result by using post-processing software Tecplot to obtain relevant information of a coal particle flow; and determining whether the pulverized coal transportation process is safe.
The present disclosure relates to a method for preparing a self-cleaning anti-icing coating based on brushlike organosilicon. In this method, a brushlike organosilicon-modified polyurethane coating is prepared by subjecting a thiolactone, a diamine compound and monovinyl-terminated polydimethylsiloxane to a simple multi-component click reaction to obtain a dihydroxy-terminated block, and introducing the dihydroxy-terminated block into a polyurethane matrix.
A mobile terminal-based method for identifying and monitoring the emotions of a user, which is different from carrying out single prediction in the traditional sense purely on the basis of expression, behavior and language, rather, in said method, various pieces of data generated by a user, such as communication, sleep, APP usage and the like, are collected by means of a mobile phone and a bracelet that the user carries daily, a data set that covers the user's life to a high degree, and then an accurate perception of the emotions of the user is completed. During data collection, the collection process is simplified and user-friendliness is improved, emotional features of a user are described from a multi-dimensional perspective to improve the accuracy of recognition results; meanwhile, in a data processing stage, key features are extracted for modeling, reducing the complexity of a subsequent machine learning algorithm and thus shortening the algorithm running time.
A method, system and machine-readable storage medium that trains a deep neural network (DNN) based on the minimization of a loss function is disclosed. The method comprises: specifying, by a processing device, a similarity function for calculating a similarity score for two representations of utterances (104); receiving, by the processing device, a training data set comprising pairs of representations of utterances, wherein each of the pairs of representations of utterances is associated with a corresponding ground-truth label (106); calculating, by the processing device, a respective similarity score for each of the pairs of representations of utterances (108); and updating, by the processing device, parameters associated with the DNN based on minimizing a loss function associated with an area under a section of a receiver-operating-characteristic (ROC) curve for the similarity scores, wherein the section is delimited between a low false positive rate (FPR) value and a high FPR value (110).
This invention relates to a robust optimal design method for photovoltaic cells. Firstly, the deterministic optimal model is established, which is solved by Monte Carlo method to obtain the maximum output power value of optimization objective and its corresponding design variable value, and then the design variable value obtained from deterministic optimization is deemed as the initial point of the mean value of the robust optimal design variable. Later, the robust optimal model is solved by Monte Carlo method in order to obtain the mean value of design variable, and then appropriate materials and manufacturing techniques are selected for corresponding photovoltaic components according to the design variable obtained, so as to achieve the robust optimal design of photovoltaic cells. In fact, this invention improves the output stability and reliability of photovoltaic cells.
G06F 119/02 - Reliability analysis or reliability optimisationFailure analysis, e.g. worst case scenario performance, failure mode and effects analysis [FMEA]
73.
Flexible differential microphone arrays with fractional order
A beamformer, for a differential microphone array (DMA) including a number M of microphones, is constructed based on a specified target directivity factor (DF) value for the DMA. An N order beampattern is generated for the DMA, wherein N is an integer and a first DF value corresponding to the N order beampattern is greater than the target DF value. An N−1 order beampattern is generated for the DMA, wherein a second DF value corresponding to the N−1 order beampattern is greater than the target DF value. A fractional order beampattern is generated for the DMA, wherein a third DF value corresponding to the fractional order beampattern matches the target DF value and the fractional order beampattern comprises a first fractional contribution from the N order beampattern and a second fractional contribution from the N−1 order beampattern.
A method for piercing a titanium alloy solid billet, the method including: 1) providing a Mannesmann rotary piercer including two rollers, a feed channel, a plurality of centering devices, and a mandril including a plug; fixing the mandril using the plurality of centering devices, where the Mannesmann rotary piercer has a feeding angle of 6-18°, a cross angle of 15°, and a roll speed of 30-90 rpm; 2) heating a titanium alloy solid billet to 930-990° C.; 3) transferring the titanium alloy solid billet to the feed channel of the Mannesmann rotary piercer; and 4) aligning the titanium alloy solid billet with the plug of the mandril, and driving the titanium alloy solid billet to pass through the plug of the mandril, thereby piercing the titanium alloy solid billet and yielding a titanium alloy tube.
The invention relates to the field of micro-sensors, and particularly to a novel MEMS technology-based, chip-level, resonant, acousto-optically coupled, solid-state wave gyroscope. The gyroscope integrates an acoustic surface wave-sensitive structure and a resonant microring resonator optical measurement structure, and applies said structures in a gyroscope. The acousto-optical effect is used to convert mechanical strain on a device crystal caused by a primary acoustic surface wave and a secondary acoustic wave caused by the Coriolis force into a change in refractive index of an optical waveguide etched on the device, causing an optical signal transmitted in the waveguide to diffract, thereby producing a frequency modulation. Meanwhile, a microresonator that uses the principle of resonance separates the frequency change introduced into the primary acoustic surface wave to obtain an output signal containing external angular velocity information. The quality factor of the resonant microring resonator forms a proportional relationship with the measurement resolution, increasing the magnitude of interface measurement resolution. At the same time that aspects such as sensitivity are increased and resolution improved, the performance indexes of the gyroscope are optimized, achieving improvement in the accuracy thereof.
G01C 19/5698 - Turn-sensitive devices using vibrating masses, e.g. vibratory angular rate sensors based on Coriolis forces using acoustic waves, e.g. surface acoustic wave gyros
A device and a method for continuous temperature gradient heat treatment of a rod-shaped material are disclosed. The furnace body of the device includes an upper heating zone and a lower heating zone inside, which are independently controlled in temperature by means of an upper heating power supply and a lower heating power supply. Moreover, both the upper heating zone and the lower heating zone are closed heating zones. The closed heat insulation plates could prevent heat loss and ensure precise temperature control of the upper heating zone and the lower heating zone. In the device, a vacuum pumping equipment is included; an annular radiation screen is configured between the upper heating zone and the lower heating zone, and the rod-shaped material is not in contact with the annular radiation screen The rod-shaped material conducts one-dimensional heat transfer along the axial direction.
C21D 11/00 - Process control or regulation for heat treatments
C21D 1/773 - Methods of treatment in inert gas, controlled atmosphere, vacuum or pulverulent material under reduced pressure or vacuum
C21D 9/00 - Heat treatment, e.g. annealing, hardening, quenching or tempering, adapted for particular articlesFurnaces therefor
C21D 9/52 - Heat treatment, e.g. annealing, hardening, quenching or tempering, adapted for particular articlesFurnaces therefor for wiresHeat treatment, e.g. annealing, hardening, quenching or tempering, adapted for particular articlesFurnaces therefor for strips
A binaural beamformer comprising two beamforming filters may be communicatively coupled to a microphone array to generates two beamforming outputs, one for the left ear and the other for the right ear. The beamforming filters may be configured in such a way that they are orthogonal to each other to make white noise components in the binaural outputs substantially uncorrelated and desired signal components in the binaural outputs highly correlated. As a result, the human auditory system may better separate the desired signal from white noise and intelligibility of the desired signal may be improved.
Disclosed is an aqueous ammonium ion battery electrode based on a pyrazine-fused ring semiconductor, wherein a coordination effect between ammonium ions and aromatic nitrogen atoms in a pyrazine-fused ring semiconductor is used, and the pyrazine-fused ring semiconductor having multiple ion storage sites acts as an ammonium ion storage material to improve the specific capacity of an aqueous rechargeable ammonium ion battery. The aqueous ammonium ion battery electrode based on the pyrazine-fused ring semiconductor and having a specific capacity reaching 355 mAhg-1at a current density of 50 mAg-1is prepared for the first time, and the specific capacity of the electrode during rapid discharge at a current density of 600 mAg-1can still reach 190 mAhg-1.
The present invention provides an identity recognition method based on an expert feedback mechanism. Domain experts are introduced to properly feed back the result of a static model, the model is dynamically adjusted and updated according to the situation fed back by the experts each time, and the model is adjusted so that similar recognition objects can be converted from incorrect recognition to correct recognition. According to the present invention, the model can adapt to a dynamic change of an environment, thereby improving the intelligence of an identity recognition algorithm by using expert knowledge, and improving the accuracy of identity recognition in a dynamic environment and the robustness of the model. An identity recognition model based on a tree structure is combined with expert feedback, and the structure of the model is adjusted in real time according to an expert feedback result, so that the recognition accuracy of the recognition model is improved on the premise that repeated training is not needed, the problem that the accuracy of a static identity recognition model is reduced in a dynamically changing environment is solved, the adaptability of the recognition model to the environment change is improved, updating time of the model is shortened, and the operating efficiency of an identity recognition application system is improved.
Provided by the present invention is a ubiquitous operating system, specifically CrowdOS, that is oriented towards group intelligence perception. By means of the in-depth analysis of the complex environment and diversified features of group intelligence tasks, a set of comprehensive processing mechanisms and core functional components are designed, comprising three core mechanisms, i.e. task semantic analysis and user scheduling, system resource management, and in-depth feedback interaction of task results. The present invention uses CrowdOS to solve the problems of the lack of a unified system structure of existing mobile group intelligence perception or a crowdsourcing platform and the incompatibility of algorithms or modules in related research. The task analysis and scheduling mechanism establishes a bridge between tasks and OS kernels by means of a task resource map, and adaptively selects a reasonable allocation strategy for heterogeneous tasks. The resource management mechanism abstracts heterogeneous physical and virtual resources in the system to provide unified software definition and management. A result quality optimization mechanism quantifies and optimizes the quality of the results.
A human-machine collaborative video anomaly detection method. The method comprises: by using video frames and traditional image optical flow descriptors as input data, encoding an autoencoder neural network, converting into hidden-layer representation content, and outputting the hidden-layer representation content by means of decoding reconstruction; training an autoencoder network by using a normal sample, wherein at a test stage, if the normal sample is inputted, high similarity is maintained between a final reconstruction result and an inputted sample, and on the contrary, if an abnormal sample is inputted, a final reconstruction error has a large deviation from the inputted sample; according to the reconstruction error, setting an appropriate threshold value for a test result, considering that the result is normal if the result is less than the threshold value, and considering the result is abnormal if the result is greater than the threshold value; and then requesting feedback at a certain probability, determining, by a person, a video frame for which feedback is initiated, directly outputting the video frame if the video frame is correctly detected, marking the video frame if a detection error occurs, marking the video frame as 1 if the video frame is normal, marking the video frame as 0 if the video frame is abnormal, and then returning the sample having the detection error to a model for input.
A hierarchical porous honeycombed nickel oxide microsphere and a preparation method thereof are disclosed. The method includes mixing nickel sulfate hexahydrate, urea, water and glycerol, to obtain a mixed solution; subjecting the mixed solution to a hydrothermal reaction, to obtain a precursor; and calcining the precursor, to obtain the hierarchical porous honeycombed nickel oxide microspheres.
Side chain liquid crystal epoxy monomer (S-LCEM) and preparation method thereof, and side chain liquid crystal epoxy resin (S-LCER) with high intrinsic thermal conductivity
The present disclosure provides a side chain liquid crystal epoxy monomer (S-LCEM) and a preparation method thereof, and a side chain liquid crystal epoxy resin (S-LCER) with high intrinsic thermal conductivity, and belongs to the technical field of epoxy resin materials. There are biphenyl mesogenic groups with strong rigidity in a molecular structure of the S-LCEM provided in the present disclosure and there are also flexible connections among chain segments, which promotes the ordered arrangement of S-LCEM molecular chains during a curing process. The highly-ordered arrangement of such mesogenic units is conducive to the formation of a local crystalloid structure, so that a heat flow is transferred along a direction of the ordered molecular chain, which effectively inhibits the scattering of phonons in the S-LCER with high intrinsic thermal conductivity and greatly improves the intrinsic thermal conductivity of the S-LCER with high intrinsic thermal conductivity.
C09K 19/12 - Non-steroidal liquid crystal compounds containing at least two non-condensed rings containing at least two benzene rings at least two benzene rings directly linked, e.g. biphenyls
85.
SEQUENTIAL RECOMMENDATION METHOD BASED ON LONG-TERM INTEREST AND SHORT-TERM INTEREST
Provided is a sequential recommendation method based on long-term interest and short-term interest, the method comprising: processing user purchase sequence data and user questioning data in a data set to obtain sequential interaction data of a user and a commodity, and extracting comment content provided by the user regarding the commodity to represent a feature of the commodity; then, using a recurrent neural network to learn stable long-term preferences of the user from historical purchase sequence data of the user, and using questioning data to perform modeling on instant interest of the user; and finally, for the stable long-term preferences and dynamic instant interest, using an attention mechanism to portray the degrees of dependence of different users with regard to the two features. Accordingly, the problem of inaccurate recommendation caused by user preference evolution can be effectively solved, and the different degrees of dependence of different users with regard to long-term preferences and instant interest can be effectively represented.
A differential microphone array includes a plurality of microphones situated on a substantially planar platform and a processing device, communicatively coupled to the plurality of microphones, to receive a plurality of electronic signals generated by the plurality of microphones responsive to a sound source and execute a minimum-norm beamformer to calculate an estimate of the sound source based on the plurality of electronic signals, wherein the minimum-norm beamformer is determined subject to a constraint that an approximation of a beampattern associated with the differential microphone array substantially matches a target beampattern.
H04R 1/40 - Arrangements for obtaining desired frequency or directional characteristics for obtaining desired directional characteristic only by combining a number of identical transducers
NINGBO INSTITUTE OF NORTHWESTERN POLYTECHNICAL UNIVERSITY (China)
NORTHWESTERN POLYTECHNICAL UNIVERSITY (China)
Inventor
Wang, Minqing
Wang, Shuai
Abstract
The present invention relates to the field of marine ranches, and particularly relates to an offshore acoustic ranch farming platform. The platform comprises an offshore working platform, an energy supply device is provided on the offshore working platform, and an underwater acoustic system is provided under the offshore working platform; the underwater acoustic system is located in sea water; pile foundations are excited by excitation devices to generate acoustic waves having different frequency characteristics, and using fish-school-sensitive acoustic waves, fish schools are attracted, driven, and fed. The purposes are to use an acoustic adjustment and control technology for reducing the requirements of a marine ranch on conventional marine mechanical equipment, for greatly reducing dependency on existing marine equipment and cost input, for reducing pollution to a marine ecological environment, and for promoting building of marine ranches.
Provided is a personalized accurate recommendation method driven by a knowledge graph. The method comprises: according to historical behaviors of users, obtaining related knowledge of articles from a knowledge base so as to construct a knowledge graph, initializing vector representation of each node and connection, and determining a receptive field of each node; generating a training sample according to the historical behaviors of the users, and initializing vector representations of all users and articles; acquiring a receptive field of each article in the training sample corresponding to an entity in the knowledge graph, and inputting the receptive field thereof and the sample as a graph neural network model, so as to obtain prediction values of the possibility of interaction between the users and the articles; optimizing model parameters by minimizing a loss function; and after a model optimization process is finished, sorting the prediction values of the possibility of interaction between a certain user and all the articles, so as to obtain a recommendation list for the user. In the present invention, knowledge graph information is used to make up for a sparsity of original user historical behavior information, and to depict users and articles from multiple dimensions and perspectives, such that a personalized recommendation result is more accurate.
NINGBO INSTITUTE OF NORTHWESTERN POLYTECHNICAL UNIVERSITY (China)
NORTHWESTERN POLYTECHNICAL UNIVERSITY (China)
Inventor
Wang, Minqing
Wang, Shuai
Abstract
The present invention belongs to the field of marine ranches, and particularly relates to a marine acoustic ranch breeding method, including an underwater acoustic robot cluster, an energy supply system and an information transmission system. On the basis of biological hydroacoustic principles, the underwater acoustic robot cluster lures and expels a shoal of fish by means of actively emitting acoustic waves of a particular frequency, and controls the shoal of fish where no marine structure exists. The aims are to realize the migration of a shoal of fish and dispose of the pattern of feeding in a single fixed-point area, so as to realize dynamic fish-farming and fishing in a broader sea area.
The present disclosure relates to an autonomous underwater vehicle (AUV) launch and recovery device driven by an elastic linkage mechanism for an extra-large unmanned underwater vehicle (XLUUV). The AUV launch and recovery device includes a hydraulic device, a push plate and a tubular device box, where the tubular device box adopts a frame-type tubular structure with a closed end; the push plate is fixed to a hydraulic rod, the hydraulic rod is controlled to stretch, and furthermore, the push plate is controlled to radially slide in a groove; and as the push plate is controlled to move radially, an inner diameter of a ring part of the inelastic linkage rope is narrowed or enlarged, so that inelastic hauling ropes are pulled to move axially, and the front end of the elastic rubber plates is further pulled to achieve an expanding or contracting state of an recovery/launch opening.
The present invention provides a personalized dialogue content generating method, comprising: a multi-round dialogue content generating model, a personalized multi-round dialogue content generating model, and a diversified personalized dialogue content generating model. Efficient vector representation of each word in a sequence is obtained according to context information by means of a Transformer model, subsequent text content can be automatically predicted and generated according to preceding text by learning a sequential dependency relationship between natural languages, and thus, corresponding reply content can be generated according to dialogue context; moreover, multiple optimization algorithms are added, so that the generation probability of universal replies can be reduced, thereby improving the diversity of generated dialogue content.
Disclosed in the present invention is a personalised product description generating method based on multi-source crowd intelligence data: collecting data required for personalised product description; making a product portrait and making a user portrait to obtain a corresponding user label; on the basis of the product portrait, obtaining a user portrait label set; obtaining a user preference label and product label from the user portrait and the product portrait and matching same to obtain a user personalised preference label; and generating a personalised product description. Using a word vector model, the text content can be expressed in a machine-computable vector form; by means of a codec structure, inputted user information and product information are encoded after matching, and the vector result obtained by encoding is decoded to generate word-by-word a personalised product recommendation text. By means of employing different text generation methods for different product attributes, using different characteristics of text generation methods such as extractive text generation and generative text generation, and incorporating multi-source data, the generated product description is smoother and more on-topic.
Process for the preparation of a ceramic nanowire preform, in particular, a process for the preparation of a ceramic nanowire preform by combining a template technique and a preceramic polymer conversion technique. The process uses carbonaceous material as a template, and prepares an isotropic ceramic nanowire preform by controlling the ratio of a precursor to a solvent, the amount of a catalyst and the ratio of a prepared precursor solution to the carbonaceous template, wherein the preform is isotropic and has lower bulk density and higher volume fraction.
NINGBO INSTITUTE OF NORTHWESTERN POLYTECHNICAL UNIVERSITY (China)
NORTHWESTERN POLYTECHNICAL UNIVERSITY (China)
Inventor
Wang, Minqing
Wang, Shuai
Abstract
A fan apparatus integrating functions of power generation and air supply, comprising expansion-type rotors (1), a rotation shaft (18), a reversible motor (5), a forward and reversal rotation controller (9), and a control system (8). Wind energy is converted into electric energy under wind action. Conversely, the electric energy may be outputted by a power grid (11), and the electric energy is converted into wind energy.
A label-free biochemical reaction detection method, comprising: S100, filling a pipe (4) with an electrolyte solution; S200, selecting receptor molecule A (6) capable of reacting with target molecule B (7) and modifying receptor molecule A (6) on the inner wall of the pipe; S300, introducing a dispersed phase into the pipe (4) for same to form a dispersed phase area (3), and retaining the dispersed phase area (3) in the pipe (4); S400, measuring and recording a contact angle of the dispersed phase area (3) in the pipe (4) and the total resistance of the solution in the pipe (4) before reacting with the target molecule (7); S500, introducing a solution or fluid possibly containing target molecule B (7) into the pipe (4) and reacting with receptor molecule A (6); S600, repeating step S300 to step S400, comparing changes in the contact angle and resistance of the dispersed phase before and after the reaction; if the change is great, then the solution or fluid contains target molecule B (7); and if not, same does not contain target molecule B (7). This serves as an inexpensive, highly sensitive, quick, and label-obviating label-free detection method for a biochemical molecular reaction.
332333; adding a PEG solid to the aqueous solution, with the molar ratio of PEG to Fe3+being 1:1 to 1:50, and sufficiently stirring same until dissolved; stirring and heating the resulting solution in a water bath at 75°C for 10-50 minutes, immediately removing same, and cooling same in an ice bath; and subjecting the cooled mixed solution to low-temperature high-speed centrifugation and washing three times, and then discarding a supernatant, with the resulting precipitate being PEG-modified ferrihydrite nanoparticles (PEG-Fns). The PEG-Fns synthesized in the present invention can undergo a blue-ray-controllable induced reduction to release Fe2+, and a Fenton reaction of Fe2+222 thus occurs in a cell to produce ·OH, and this induces oxidative damage to the cell, thus achieving the aim of controllably counteracting cancer and bacteria.
A61K 41/00 - Medicinal preparations obtained by treating materials with wave energy or particle radiation
A61K 47/60 - Medicinal preparations characterised by the non-active ingredients used, e.g. carriers or inert additivesTargeting or modifying agents chemically bound to the active ingredient the non-active ingredient being chemically bound to the active ingredient, e.g. polymer-drug conjugates the non-active ingredient being a modifying agent the modifying agent being an organic macromolecular compound, e.g. an oligomeric, polymeric or dendrimeric molecule obtained otherwise than by reactions only involving carbon-to-carbon unsaturated bonds, e.g. polyureas or polyurethanes the organic macromolecular compound being a polyoxyalkylene oligomer, polymer or dendrimer, e.g. PEG, PPG, PEO or polyglycerol
A61K 47/69 - Medicinal preparations characterised by the non-active ingredients used, e.g. carriers or inert additivesTargeting or modifying agents chemically bound to the active ingredient the non-active ingredient being chemically bound to the active ingredient, e.g. polymer-drug conjugates the conjugate being characterised by physical or galenical forms, e.g. emulsion, particle, inclusion complex, stent or kit
A composite loading device for a compressive/tensile stress wave-torsional stress wave, comprising: a torsional stress wave generator, a compressive/tensile stress wave generator, and a stress wave synchronization device (10). The torsional stress wave generator performs torsional wave loading on a sample (8), and the torsional stress wave generator comprises: a torsion ratchet mechanism (1), a clamping mechanism (2) and a torsion rod (3); the torsion ratchet mechanism (1) applies torque to the torsion rod (3) so that torque energy is stored from the proximal end of the torsion rod (3) to the clamping position; a torsion main coil of the torsional stress wave generator is discharged, so that the torque energy stored in the torsion rod (3) is released, thereby generating a torsional stress wave on the torsion rod (3); the compressive/tensile stress wave generator performs compressive/tensile stress wave loading on the sample (8), and the stress wave synchronization device (10) sends control signals to the torsional stress wave generator and the compressive/tensile stress wave generator, so that the torsional stress wave generated by the torsional stress wave generator and the compressive/tensile stress wave generated by the compressive/tensile stress wave generator simultaneously load the sample (8).
A method for evaluating the reliability of a sealing structure in a multi-failure mode based on an Adaboost algorithm. The Adaboost algorithm is adopted to carry out a classification iterative training on the seal ring failure related data of a small sample until a classification error of set classifier meets a precision requirement; then the failure probability of the sealing structure is calculated under the fluctuation condition of related parameters by adopting the important sampling method, and further the reliability of the sealing structure is evaluated in the multi-failure mode. The present invention solves the problems of long time consumption and complex calculation process of reliability evaluation in multi-failure mode of the complex structure.
G06F 30/27 - Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
G06T 17/20 - Wire-frame description, e.g. polygonalisation or tessellation
G06F 18/211 - Selection of the most significant subset of features
G06F 18/2415 - Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
G06N 7/01 - Probabilistic graphical models, e.g. probabilistic networks
G06F 119/02 - Reliability analysis or reliability optimisationFailure analysis, e.g. worst case scenario performance, failure mode and effects analysis [FMEA]
99.
EXPERIMENTAL TEACHING SYSTEM, METHOD FOR OPERATING EXPERIMENTAL TEACHING SYSTEM, AND STORAGE MEDIUM
An experimental teaching system (100), a method for operating the experimental teaching system (100), and a storage medium. The experimental teaching system (100) comprises: an experiment board (101) comprising a single-chip microcomputer; and an expansion board part (102) comprising a plurality of expansion boards (103, 104, 105) separated from each other, wherein each of the plurality of expansion boards (103, 104, 105) being configured to be in signal connection to the experiment board (101) during use, and implement a preset function under the control of the single-chip microcomputer.
G09B 23/18 - Models for scientific, medical, or mathematical purposes, e.g. full-sized device for demonstration purposes for physics for electricity or magnetism
100.
Control mechanism of vehicle-mounted system for electromagnetic launch of fire extinguishing bombs for high-rise buildings
A control mechanism of a vehicle-mounted system for electromagnetic launch of fire extinguishing bombs for high-rise buildings is provided. A stress wave amplifier is fixedly connected to a secondary coil, a primary coil is fixedly connected to a coil base, a guide shaft passes through a central hole in the primary coil and the coil base, and a head of the guide shaft is fixedly connected to the secondary coil. A step-up transformer TM1 boosts a 380-V alternating current and changes it to direct current to charge a pulse capacitor C1 and store energy in the pulse capacitor C1. A discharge thyristor M3 is triggered, and the pulse capacitor C1 releases the energy instantaneously. A stress wave is generated between the primary coil and the secondary coil. The stress wave is transmitted to a fire extinguishing bomb through the stress wave amplifier, to launch the fire extinguishing bomb.
A62C 3/02 - Fire prevention, containment or extinguishing specially adapted for particular objects or places for area conflagrations, e.g. forest fires, subterranean fires
F42B 12/46 - Projectiles, missiles or mines characterised by the warhead, the intended effect, or the material characterised by the warhead or the intended effect for dispensing materialsProjectiles, missiles or mines characterised by the warhead, the intended effect, or the material characterised by the warhead or the intended effect for producing chemical or physical reactionProjectiles, missiles or mines characterised by the warhead, the intended effect, or the material characterised by the warhead or the intended effect for signalling for dispensing gases, vapours, powders or chemically-reactive substances
A62C 19/00 - Hand fire-extinguishers in which the extinguishing substance is expelled by an explosionExploding containers thrown into the fire
patents
