36results about How to "Overcome loss" patented technology

The invention discloses a geological disaster interpretation method, device and equipment under steep terrain and a medium, and relates to the field of remote sensing geological disaster investigation, and the method comprises the steps: obtaining data and radar point cloud data of a target region; extracting a ground point cloud of the target area from the radar point cloud data; performing feature extraction, clustering and point cloud segmentation on the ground point cloud in sequence to determine slope units in the target area; performing rotation transformation on the point clouds in the slope units; classifying and correcting all the converted slope surface units in the target area, and determining abrupt slope point clouds in the target area; and performing three-dimensional visual rendering on the abrupt slope point cloud to obtain an abrupt slope three-dimensional model of the target area, and performing geological disaster interpretation based on the abrupt slope three-dimensional model and the topographic data and geological data in the data to obtain an interpretation result. And the interpretation of geological disasters in steep terrains can be conveniently and efficiently realized.

The invention provides a capacity attenuation trend prediction method of a low-temperature lithium battery, and relates to the technical field of battery health management and prediction. The capacity fading trend prediction method of the low-temperature lithium battery specifically comprises the following steps of S1, initial data acquisition, S2, feature extraction, S3, hybrid prediction model construction, S4, model training and offline verification, S5, online deployment and dynamic correction, and S6, SOH evaluation and early warning output. A hybrid prediction model is constructed by fusing an electrochemical mechanism and deep learning, so that high-precision prediction and full-life-cycle intelligent health management of lithium battery capacity fading in a low-temperature environment are realized, and the problems of accelerated battery aging and prominent safety risk in an alpine region are effectively solved; early warning and intelligent management of the health state of the battery can be realized, and the service life of the battery under low-temperature application can be prolonged.

The invention provides an intelligent driving closed-loop control method based on a pre-training large model. The intelligent driving closed-loop control method comprises the following steps: S1, obtaining full-quantity heterogeneous input data in a vehicle driving process; s2, performing feature coding on the total heterogeneous input data to generate feature data suitable for pre-training large model processing; s3, inputting the encoded feature data into a pre-trained large model for feature extraction, and outputting a key value pair; s4, constructing a plurality of task head sharing key value pairs in a cascade form, and generating intention-level and track-level outputs; s5, outputting and generating a vehicle control instruction according to the intention level and the track level to control the driving state of the vehicle; according to the method, full-quantity heterogeneous input features in a closed-loop form are constructed, so that the use efficiency of data is remarkably improved, and the most comprehensive feature representation is extracted from multi-source input, so that an output result with higher quality is generated; according to the method, the problems of information loss and accumulative errors in a traditional modular system are effectively solved, and the accuracy and reliability of decision planning are improved.

The invention discloses a high-impact-strength polyethylene naphthalate composite material and a preparation method thereof, and belongs to the field of composites.The preparation method includes the steps that 100 parts by mass of polyethylene naphthalate, 10-40 parts by mass of polyacrylate elastomer, 0.5-1.5 parts by mass of heat stabilizer and 0.5-1 part by mass of antioxidant additive are mixed to be uniform, melt extrusion is conducted, and the high-impact-strength polyethylene naphthalate composite material is obtained. The polyethylene naphthalate composite material with high impact strength is obtained. According to the invention, the polyacrylate elastomer with low crosslinking density is dispersed into the PEN resin to form a phase morphology with the characteristic of low matrix ligament thickness, and the morphology can effectively increase the ductility of PEN, so that the PEN shows better energy absorption and deformability under impact, thereby avoiding brittle rupture. The polyethylene naphthalate composite material provided by the invention is excellent in comprehensive performance, simple in preparation process, easily available in raw materials and wide in industrial application prospect.

The invention discloses a preparation method of iso-methyl ionone. According to the method disclosed by the invention, residual citral methylal in pseudo-iso-methyl ionone is decomposed to generate citral in a cyclization process, and is oxidized into C10 acid by air in a refining process, and the C10 acid enters an iso-methyl ionone product, so that the effects of regulating the odor of the product and improving the odor stability of the product are achieved.

The invention discloses a wireless passive thin film type humidity sensor which comprises an LC resonance circuit, a first isolation thin film, a humidity sensitive thin film, a second isolation thin film, a magnetic thin film and a substrate thin film. The LC resonance circuit is composed of a metal coil, a substrate and an interdigital capacitor, and the metal coil and the interdigital capacitor are placed on the two sides of the substrate respectively and connected in series through through holes formed in the substrate to form the LC resonance circuit; the first isolation film is arranged between the LC resonance circuit and the humidity-sensitive film, the second isolation film is arranged between the humidity-sensitive film and the magnetic film, and the substrate film is arranged on the other side of the magnetic film. When the humidity of the external environment changes, the humidity-sensitive film expands or shrinks due to the change of the water content, so that the distance between the LC resonance circuit and the magnetic film is changed, the inductance is changed along with the change, the resonance frequency of the LC resonance circuit is changed, and wireless passive detection of the humidity is realized by measuring the resonance frequency.

The invention belongs to the field of auditory meatus sound wave reflection signal processing, and discloses a mouth shape recognition method and system based on auditory meatus sound wave reflection signal processing, noise and interference components in signals are eliminated, and sound wave signals are subjected to preprocessing such as cutting and tailoring; obtaining an optimal regular path between the two time sequences; useful information is extracted; performing preprocessing and feature extraction on the signals; acoustic wave signal features are extracted through Fourier transform, frequency domain energy calculation, feature value calculation and the like; and carrying out KNN classifier classification. Theoretical and technical supports are provided for improving the precision and robustness of mouth shape recognition, exploring the application potential of ear canal sound wave reflection signal processing in mouth shape recognition and promoting development of a mouth shape recognition technology achieved through ear canal sound wave reflection signal processing in practical application.

The invention discloses a quantum classical mixed image classification method based on group isovariant and delayed aggregation. The method comprises the following steps: lifting a convolution module, inputting a target image, initializing a convolution kernel and rotating the convolution kernel, and carrying out convolution and stacking on the target image to obtain an output image; the group convolution module is used for converting the output image of the convolution module into an image sample and carrying out convolution and reverse remodeling operation to obtain an output image; the group pooling and flattening module is used for carrying out adaptive average pooling on an output image of the group convolution module, compressing the output image into a feature vector and obtaining an output feature vector; the group equivariant quantum feature processing module encodes the feature vectors to quantum bits according to the feature vectors output by the group pooling and flattening module; and the delay aggregation and output module measures the quantum bits, aggregates and classifies measurement results, and completes classification of target images. The method provided by the invention has strong nonlinear feature fusion capability, and can realize more accurate image classification.

The invention relates to a slurry one-way valve, a sampling barrel and a formation tester, the slurry one-way valve comprises a valve body, a first hole section and a second hole section which are communicated with each other are arranged in the valve body, and the first hole section and the second hole section are connected to form a slurry through hole penetrating through the valve body; a valve element and an elastic unit are arranged in the second hole section, the end, away from the first hole section, of the second hole section is connected with a limiting piece, a fluid through hole with the axis parallel to the axis of the mud through hole is formed in the limiting piece, the limiting piece is used for limiting the valve element and the elastic unit in the second hole section, and the elastic unit is used for pressing the valve element on the bottom wall of the first hole section. The valve element can seal the mud through hole after being pressed on the bottom wall of the first hole section. When the mud one-way valve is applied to a sampling barrel of a formation tester, the requirement that a formation fluid sample smoothly enters the sampling barrel can be met, the formation fluid sample cannot reversely flow through the mud one-way valve, and therefore the problem that the formation fluid sample in the sampling barrel is lost is solved.

The present invention is applicable to the field of dynamic intelligent scheduling of logistics, and is specifically a dynamic scheduling method and system for vehicle logistics based on deep learning prediction. In the scheduling method, in the upper model of the scheduling model, the order set to be allocated is allocated downward using the recommended transportation path; in the middle model of the scheduling model, a loading scheme selection model is established to maximize the loading value and determine the loading scheme; in the lower model of the scheduling model, tasks are reasonably allocated to specific FLSPs according to the loading scheme of the middle model and the logistics service capabilities of each FLSP, and a vehicle FLSP selection model is constructed to maximize the satisfaction of the FLSP; the scheduling model is matched with the predicted order flow to obtain the capacity resource satisfaction of the predicted order flow, and before the time window of the predicted order flow is reached, the capacity distribution is dynamically adjusted based on the capacity resource satisfaction. The present invention can make full use of capacity resources.

The invention relates to the field of rolling, and discloses an anti-backlash device, an anti-backlash control method and rolling equipment, the anti-backlash device comprises two bearing seats, a bearing set, a jacking structure and a hydraulic driving part, the two bearing seats are provided with mounting cavities in a penetrating mode in the first direction, one side face of each bearing seat is provided with a containing cavity, and the containing cavities communicate with the mounting cavities; the two ends, in the first direction, of the roller are each provided with a rotating shaft, the rotating shafts are located in the corresponding installation cavities, each installation cavity is internally provided with a bearing set, each bearing set comprises a third bearing, the jacking structure abuts against the third bearing, the hydraulic driving part is in transmission connection with the jacking structure, and the hydraulic driving part drives the jacking structure to abut against the third bearing. The hydraulic driving part drives the jacking structure to continuously and stably apply the pre-tightening force to the third bearing, a thread jackscrew structure which is prone to loosening and losing efficacy is replaced, the problem that the pre-tightening force is attenuated or lost due to vibration and loads is effectively solved, and therefore the anti-backlash effect can be reliably maintained for a long time.

The invention discloses an abnormal trigger key data identification and priority transmission method supporting QoS, which is suitable for analyzing voltage waveform data output by a photovoltaic system, and comprises the following steps: 1, acquiring a voltage waveform data set output by the photovoltaic system, defining a real category label of each data point, and distinguishing normal waveform data from abnormal waveform data; 2, data decomposition: extracting each frequency component of the data layer by layer through a three-band filter, calculating the energy of each component, and predicting the abnormal probability of the data through LSTM; and 3, constructing a data backoff mechanism, and preferentially transmitting abnormal data. According to the invention, by combining the signal processing technology and the machine learning model, efficient analysis and anomaly detection of the voltage waveform data of the photovoltaic system are realized, and rapid transmission of key data is ensured, so that the performance and stability of the photovoltaic system are improved.

The present invention discloses a financial data processing method, device, electronic device and processing medium, and relates to the field of financial technology. A specific implementation of the method includes: being able to obtain the arrival timestamp of the financial data to be processed and obtaining the current data window; when it is determined that the arrival timestamp exceeds the time range, constructing a new target data window for the financial data to be processed; the target data window is time-continuous with the current data window; and processing the financial data within the target data window. The embodiments of the present invention overcome the problem of data loss or data anomaly caused by receiving and processing financial data in a data window constructed based on the existing method by adaptively adjusting the data window and constructing a time-continuous data window; and improve the integrity and reliability of financial data processing.

This invention discloses a cross-modal data processing system for the safe operation of hydrogen refueling stations, including a raw data acquisition and processing module, a full-variable safety scanning module, a physical relationship coupling diagnosis module, an adaptive operating condition clustering module, a question-answer pair construction module, a hydrogen refueling station time-series command data acquisition module, and a fault type identification module. By constructing a three-layer semantic enhancement logic and model adaptation strategy, this invention effectively solves the technical problems in the prior art, such as the lack of supervision signals in the raw data of hydrogen refueling stations, the difficulty in identifying hidden faults under complex operating conditions, and the difficulty in adapting heterogeneous feature space models.

The present invention relates to the technical field of data transmission, and in particular to an industrial data acquisition and processing system based on wireless transmission technology, comprising: a data acquisition module for collecting industrial characteristic data; a data processing module for determining disguised data according to the industrial characteristic data, including a split component and a disguised component; a data transmission module for transmitting the industrial characteristic data to a target location, including a plurality of intermediate transmission nodes; a controller for determining the transmission location of the disguised data at the intermediate transmission node according to the data volume of the industrial characteristic data, determining the location stamp of the industrial characteristic data encryption according to the transmission location, and determining the type of the industrial characteristic data to be repaired according to the type loss ratio of the split industrial characteristic data. The present invention reduces the risk of data transmission by setting up sub-nodes to transmit data and disguising and encrypting the data.

The application discloses a multi-modal depth perception and grasping system based on transparent objects, and relates to the field of robot operation, comprising a multi-spectral perception module, a depth correction module, a grasping posture generation module and a control module. The application comprehensively acquires visual information and thermal radiation information of the transparent object by combining two perception methods of RGB-D image and thermal imaging (TIR) image, and analyzes systematic errors of the depth map to detect error sources of the RGB-D camera and the TIR camera. The system adopts an encoder-decoder model as a depth correction core, the model extracts complementary features of the RGB-D image and the TIR image through a modal exclusive encoder, and utilizes a feature fusion module to complete feature alignment and integration, thereby effectively improving the depth estimation accuracy on the transparent surface. Meanwhile, a Bayesian optimization method is adopted to optimize hyperparameters of the depth correction model, through hyperparameter optimization and model fitting processing, the system effectively avoids overfitting and underfitting problems, and further improves the robustness of the depth correction model and the grasping accuracy of the transparent object.

The present invention discloses an adaptive storage system and method, the system comprises: an adaptive storage management control module collects first hard disk information of a hard disk through a hard disk port adapter module, when the current hard disk protocol of an FPGA module is not compatible with the hard disk, and / or the current hard disk drive of a CPU module is not compatible with the hard disk, an adapter program compilation instruction is sent to an adaptive storage protocol program compilation module; the adaptive storage protocol program compilation module obtains an adapter program according to the adapter program compilation instruction, and saves it to a program storage unit; after the adaptive storage management control module identifies that the adapter program is successfully saved, an adapter program burning instruction is sent to a program burner to update and burn the adapter program to the FPGA module and / or the CPU module respectively; so that the new hard disk protocol of the FPGA module and / or the new hard disk drive of the CPU module are adapted to the hard disk. The problem that the software and hardware in the traditional technical solution cannot be compatible with multiple types of hard disks at the same time and the compatibility is poor is solved.

The present invention discloses a method for panchromatic sharpening based on detail injection using a dual-branch network. First, the original data is divided into training, verification and test data, and the corresponding data sets are generated and preprocessed; then a dual-branch panchromatic sharpening network model based on detail injection is constructed; the dual-branch panchromatic sharpening network model based on detail injection is trained; the test data is fused using the trained network model; and finally the quality of the fused image is evaluated. The present invention uses different independent subnetworks for feature extraction for source images of different categories, and continuously utilizes the extracted feature maps during feature fusion to make up for the information lost during the fusion process while promoting information flow; and solves the problem that the existing method does not fully utilize the unique information in the two types of source images and the model is not interpretable, resulting in low performance of the fusion result.

The invention relates to the technical field of image processing, and discloses an operation scene real-time dynamic reconstruction method and system based on three-dimensional Gaussian sputtering. The method comprises the following steps: extracting spatial-temporal characteristics of a soft tissue deformation Gaussian subset through a HexPlane spatial-temporal encoder; fusing the decoded feature vectors to generate a soft tissue deformation parameter set; a deformation field Jacobian matrix of Gaussian elements in the soft tissue deformation Gaussian subset is calculated, a local volume strain rate is obtained through determinant operation, when the strain rate exceeds a preset threshold value, self-adaptive densification operation is executed, and new Gaussian elements inherit parent Gaussian element deformation parameters and are updated to a corresponding set; and combining the updated parameters to carry out probabilistic boundary allocation and transmissivity perception rendering to obtain a reconstruction result. The accuracy, the physical authenticity and the stability of real-time dynamic reconstruction of the operation scene are improved.

This invention discloses a configuration control method for vibratory conveying of bulk materials, belonging to the field of industrial automation configuration control technology. The method includes: constructing a three-dimensional configuration monitoring interface based on the physical characteristics of the bulk material, comprising a virtual vibratory feeder, a virtual conveyor belt, and virtual material level detection points, and embedding an adaptive adjustment module for the excitation frequency; collecting real-time changes in the stacking height through the virtual material level detection points and mapping them to a feeding rate correction coefficient; calling a fuzzy rule inference engine to calculate the target excitation frequency and target amplitude; encapsulating the target parameters into configuration control instructions, and sending them to the programmable logic controller of the physical vibratory feeder for execution through the configuration software data communication interface. This method utilizes virtual material level sensing and fuzzy inference to achieve dynamic tuning of vibration parameters, and improves the real-time performance and reliability of feeding control through a structured instruction encapsulation and verification mechanism.

The present application relates to a kind of training methods of clinical auxiliary diagnosis model, belong to medical information processing technical field, solve the problem of poor coordination of size model in prior art affects the accuracy of diagnosis decision.Method includes: obtaining the medical text of multiple patients and constructs sample set;Text compression model and medical diagnosis large model are constructed, and the pre-training of text compression model and medical diagnosis large model constructed respectively is carried out based on sample set;The parameter of medical diagnosis large model is fixed, and the compressed text output by text compression model is input to medical large model, and text compression model is optimized based on the diagnosis task loss of medical diagnosis large model;The parameter of text compression model is fixed, and the compressed text output by text compression model is input to medical large model, and medical diagnosis large model is fine-tuned based on the diagnosis task loss of medical diagnosis large model, and the trained clinical auxiliary diagnosis model is obtained.The accuracy of diagnosis in complex clinical scenario is improved.

The application relates to the field of artificial intelligence and image processing technology, and discloses a multi-modal large model driven complex scene image data processing method and system. The method obtains target image data, extracts multi-scale semantic features, performs feature coding based on a scene context dependency graph, generates optimized feature representation through attention guided feature fusion and multi-level semantic reasoning, and executes a target processing operation. The application can effectively improve the complex scene image understanding accuracy, enhance the feature expression capability, improve the performance of the model in spatial relationship reasoning and semantic analysis, and is suitable for complex visual scene analysis and decision-making.

The invention discloses a method for preparing a pure phosphorus-containing solution based on purification of sludge gasification slag pickle liquor, and belongs to the technical field of sludge treatment. An acid solvent is added into the sludge gasification slag obtained after ore grinding for dissolution, stirring and acid leaching are conducted, and mixed slurry is obtained; carrying out solid-liquid separation on the mixed slurry, and taking phosphorus-containing supernate; diluting the phosphorus-containing supernate, adding strong acidic cationic resin after dilution, and purifying at normal temperature; when the acid solvent is industrial waste acid, strongly acidic cationic resin is added into the feed liquid obtained after the first purification for secondary adsorption and purification; and after purification, carrying out solid-liquid separation on the obtained feed liquid to obtain a pure phosphorus-containing solution. The purification method provided by the invention is simple in process, good in selectivity and low in phosphorus loss, and provides a high-quality raw material solution guarantee for preparing a phosphorus product with high purity and high additional value from the sludge gasification slag.

The present invention relates to a fish-shaped fiber Bragg grating pressure and temperature sensor, which is suitable for detecting the liquid pressure and temperature distribution in a pipeline, and is particularly suitable for accurate detection of the fluid pressure and temperature before and after the pipeline bend or throttle hole. The sensor comprises a skeleton, which is composed of two metal sheets arranged symmetrically with each other. The metal sheets are set as a thin sheet structure that is long in the longitudinal direction and short in the transverse direction. One of the metal sheets is provided with longitudinal grooves symmetrically on both sides of its longitudinal central axis, and is set as a negative metal sheet. According to the bionic principle of fish body, the present invention is designed with a skeleton composed of two metal sheets, a temperature measuring metal wire and a temperature measuring metal cavity in the skeleton, and fiber Bragg gratings are arranged inside and outside the central axis of the negative metal sheet. The positive and negative deformation principle of the metal sheet is used to eliminate the influence of temperature on the grating, and realize the pressure detection with temperature self-compensation; the influence of external compressive stress, internal thermal stress and impact inertia force on the grating is eliminated by encapsulating the temperature measuring metal wire and the temperature measuring cavity with a special structure, and the temperature measurement is realized.

The invention belongs to the technical field of radar signal processing, and particularly relates to a multi-time echo signal model analysis method for improving trailing. The method comprises the following steps: step 1, constructing a time domain echo signal model of the pulse laser radar; 2, constructing a single echo signal model according to the time domain echo signal model; 3, constructing a multi-target light spot detection model, and determining a target distance and a light spot proportionality coefficient corresponding to each target according to the multi-target light spot detection model; and 4, according to the single-time echo signal model, and the light spot proportionality coefficient and the target distance of each target, constructing a multi-time echo signal model. According to the multi-time echo signal model analysis method for improving trailing, the multi-time echo signal model is analyzed, a technical basis can be provided for extracting echo information of multiple targets, and the problems of information loss and low ranging precision in pulse laser radar signal processing can be solved.

The invention discloses a multi-modal aspect level sentiment analysis method based on global-local attention multi-granularity semantic fusion, belongs to the technical field of multi-modal aspect level sentiment analysis, and uses a local-global text image interaction module to realize alignment of text features and visual features from coarse granularity to fine granularity. A visually enhanced text representation is obtained. A local-global text aspect interaction module is used for capturing interaction information of different granularities between aspects and contexts, information interaction of the aspects and the texts on different granularities is considered, and global interaction between the texts and the aspects and word-level interaction between aspect words and text words are captured through global and local attention. The problem that information is lost due to the fact that a rough attention mechanism is used for aspect modeling in the past is solved.

The application discloses a method and device for controlling a rolling mill to accurately stop at a strip tail, wherein the method comprises the following steps: obtaining a mark point position of a steel strip; for any mark point in any pass, when the mark point has not left an entry coiler, controlling the entry coiler to track and record position information of the mark point; when the mark point leaves the entry coiler and has not entered an exit coiler, controlling the entry coiler and the exit coiler to synchronously track and record the position information of the mark point; and for any defect point, when the defect point has entered the exit coiler, controlling the entry coiler to clear the position information of the defect point. The application overcomes the situation that the position information of a mark point is lost or inaccurate in the prior art because the position information of the mark point is not tracked in a non-rolling process; effectively improves the reliability of length precision tracking and recording; and realizes accurate stopping at a strip tail in a rolling process.

The application discloses a personal image privacy protection method, system and electronic equipment. The method provided by the application allows a user to specify a certain person in a protected image through language expression, generates multi-scale visual features with sufficient fusion of image and text information by using a lightweight deep neural network to perform parallel coding on input reference information and personal images, generates a stable specified personal privacy protection image mask through a multi-scale feature fusion and mask positioning enhancement module in the decoding process, and in addition, introduces a balanced binary cross-entropy loss to solve the pixel imbalance problem in training, optimize the network performance, and improve the personal image privacy protection effect. The application can solve the problems of excessive protection of content and pixel imbalance in the training of the reference personal image privacy protection network in the existing personal image privacy protection technology.

The invention provides a network threat analysis method and system based on privacy protection, and the method comprises the steps: building a professional channel to issue information, enabling users which can join in after authorization and meet specific and sensitive information, carrying out the modeling analysis of the expected benefits shared by different users, carrying out the quantitative analysis through the learning evolutionary game, and obtaining a reasonable excitation strategy. And each entity user inquires from the open community to obtain the required threat intelligence, analyzes the threat intelligence and deploys own network defense measures, so that the network security defense capability can be improved, the defect that the sensitive information is leaked or misused in the shared threat intelligence in the prior art is overcome, and the user experience is improved. And serious consequences and huge loss can be caused.

This invention discloses a method for semantic detection and multi-frame joint signal reconstruction of radar composite interference, belonging to the field of radar signal processing technology. For signal reconstruction, it includes performing short-time Fourier transform and power-law nonlinear transform on mixed and clean interference signals to obtain enhanced time-frequency maps and binary masks; constructing a U-shaped network guided by physical features, embedding two-dimensional constant false alarm rate (CFAR) detection in skip connections, generating a matrix of interference signal-to-noise ratio (SNR), confidence level, and neighborhood interference proportion as physical features, and outputting a semantic mask; using the CLEAN algorithm to suppress interference, and achieving signal reconstruction based on a corrected dictionary; and solving for sparse vectors in the Doppler domain through sparse Bayesian learning to obtain a high-resolution two-dimensional distribution of the target in the range and Doppler domains. This invention improves the accuracy and anti-interference robustness of high-resolution two-dimensional reconstruction of radar targets in the range and Doppler domains through U-shaped network semantic segmentation guided by physical features and sparse Bayesian learning.