195results about How to "Improve analysis accuracy" patented technology

The invention provides an LED product surface defect detection method, which comprises the following steps: carrying out defect feature extraction on a preprocessed LED surface image by adopting a multi-scale morphological algorithm, and respectively carrying out morphological opening operation and closing operation on surface recesses and fine scratches by setting structural elements with different sizes and shapes to obtain candidate regions of the recesses and the scratches; according to the geometric dimensions, morphological characteristics and spatial distribution detail attributes of the recesses and scratches, constructing defect feature vectors, and performing classification and severity evaluation on defect candidate areas through a support vector machine algorithm to obtain the types, positions and quantitative description parameters of defects; and updating the light source parameter database and the defect feature database according to a quality detection result of the LED product, analyzing an association rule between defects and light source parameters through a data mining algorithm, and optimizing a light source parameter configuration strategy and a defect detection model.

The invention discloses a cargo transportation state monitoring system based on the Internet of Things, and relates to the technical field of logistics management. Electromagnetic signal intensity of each frequency band in a cargo box is recorded in real time through an electromagnetic interference monitoring module, and multi-point temperature data and temperature deviation information collected by a redundant sensor module are combined; the data processing and transmission module is used for preprocessing the signals and uploading the signals to the cloud platform, and the comprehensive analysis module extracts spectrum interference intensity abnormal features and temperature difference fluctuation features and accurately evaluates the influence of electromagnetic interference on the sensor data precision; then, the interference division module divides the interference event into instantaneous interference and continuous interference, the early warning processing module delays the instantaneous interference to trigger the temperature control equipment to adjust and send low-level early warning, and the continuous interference is deeply analyzed to generate a graded early warning signal and guide a coping strategy; and the data reliability, the equipment stability and the cargo protection capability of the cold chain transportation system in a complex electromagnetic environment are effectively improved.

The invention discloses a self-adaptive intelligent symbiotic data ecosystem and a cooperative implementation method thereof. The system comprises an intelligent agent ecosystem module, a self-adaptive domain knowledge graph module, a self-adaptive workflow engine module, a CLID configuration system module, a man-machine cooperation interface module, a data processing pipeline module and an innovation trigger module. Through cooperation of diversified intelligent agents, the dynamic knowledge graph, the self-adaptive workflow and CLID multi-dimensional configuration, high-flexibility and self-adaptive data processing and analysis capacity is achieved. The system can automatically construct a complex workflow, perform multi-modal data fusion, and integrate human judgment and AI capabilities at key points, thereby providing an innovative solution. The method is suitable for complex and dynamic big data processing and analysis scenes, and the data processing efficiency and innovation ability are remarkably improved.

The invention relates to the technical field of offshore wind power operation risk analysis, and discloses an offshore wind power operation risk analysis system based on deep learning, and the system comprises a data collection and preprocessing module which is used for collecting wind power equipment operation state data and environment data of multiple types of sensors, and carrying out the cleaning and preprocessing of the data; the multi-modal data fusion and feature extraction module is used for extracting and fusing multi-modal features from different data sources; and the real-time risk prediction and intelligent decision module is used for analyzing the operation state of the wind power equipment in real time based on a deep learning model, and performing prediction and alarm when a fault trend is detected. By introducing a self-attention mechanism and a Transform architecture, deep fusion analysis is performed on multi-modal data from wind power equipment, and information of different data sources can be weighted and integrated in a unified feature space, so that the system can adaptively focus on important features, and the analysis precision and fault detection capability of the multi-modal data are remarkably improved.

The invention relates to the technical field of electrochemical testing, in particular to an electrochemical impedance spectroscopy testing system based on multi-model combination, which comprises a data acquisition unit, a deep learning analysis unit, a frequency control unit and a feedback regulation unit. The deep learning analysis unit adopts a convolutional neural network and a long-short-term memory network to construct a model, after training, features are extracted, trends are analyzed, and dynamic frequency priority ranking is carried out, the frequency control unit drives the electrochemical workstation to adjust the frequency according to a weight generation strategy, and the feedback adjustment unit calculates predicted and actually measured impedance errors. When the threshold value is exceeded, parameter correction is carried out in each unit, the test frequency is intelligently optimized according to the dynamic change of the electrochemical process, key information is accurately captured, resource waste is avoided, the test efficiency and accuracy are improved, the stability and reliability of the system are ensured through a feedback regulation mechanism, and the research and analysis capability of the electrochemical system is effectively improved.

The invention discloses an intelligent monitoring control method and system for conveying equipment based on chain tension, and relates to the technical field of conveying equipment.The intelligent monitoring control method for the conveying equipment based on the chain tension comprises the following steps that chain tension characteristic data are obtained; predicting the change of the chain tension at the future moment by using the tension prediction model to obtain the change trend of the chain tension; identifying a chain tension change mode; and operating parameters of the conveying equipment are adjusted based on the recognized key factors. By monitoring the tensioning force of the chain in real time and predicting the future change trend of the tensioning force, the tensioning state of the chain can be adjusted more accurately, it is ensured that the chain runs in the optimal state all the time, excessive abrasion and damage of the chain and a chain wheel are prevented, operation parameters can be automatically adjusted according to the recognized tensioning force change mode, and the working efficiency is improved. And therefore, the performance and the safety of the conveying equipment are improved.

The invention relates to the technical field of electronic shopping, in particular to a big data analysis method based on an e-commerce platform and an AI system, and the method comprises the following steps: collecting a click rate, a search frequency and a purchase record of a user on the e-commerce platform, carrying out the subdivision of the click rate, the search frequency and the purchase record, and creating a data set; according to the invention, by integrating the equipment fingerprint technology and the social media data, the analysis precision of the user behavior data is improved, and the capabilities of user identity recognition and behavior pattern analysis are further optimized. In-depth analysis is carried out on user interaction data such as click rate, search frequency and purchase records, consumer behavior patterns and preference prediction are refined through clustering and decision tree algorithms, and sentiment analysis and keyword extraction are carried out on social media posting data. Therefore, the enterprise can monitor the market dynamic state in real time and quickly adjust the marketing strategy. And the accuracy of market prediction and the pertinence of marketing activities are improved.

The invention discloses a classroom student attention assessment method and system based on multi-source feature fusion. According to the invention, face recognition is carried out through the face detection unit of the system, occurrence frequencies of different emotional states of students are obtained through a trained student expression recognition neural network model containing four expression discrimination blocks in the expression recognition unit, and fatigue times of the students are obtained through the fatigue detection unit. The number of times of abnormal head postures of the students is obtained through the head posture estimation unit, finally attention evaluation parameters of the students are obtained through an extreme learning machine model of the attention evaluation unit, and then the attention states of the students in the class are evaluated and displayed. According to the method, a more efficient dynamic student expression recognition method can be provided, real-time and accurate classroom feedback is provided for teachers, optimization and implementation of personalized teaching strategies are supported, longitudinal extension of teaching of the teachers and transverse extension of learning modes of the students are facilitated, and new talents are expanded to cultivate a new normal form.

The invention discloses a gas outburst monitoring and early warning method and system based on an improved ConvLSTM model, and relates to the technical field of gas monitoring, and the key points of the technical scheme are that the method comprises the following steps: constructing a mine gas data set; cleaning the gas data set, converting unstructured data into structured data, and performing data cleaning, denoising and standardization on the structured data; performing high-order derived feature extraction on the preprocessed gas data set; an improved ConvLSTM model is constructed; using a historical gas data set to train the improved ConvLSTM model; identifying a potential gas high-risk area; based on an analysis result of the improved ConvLSTM model, identifying a potential gas outburst ignition risk area; and constructing a monitoring platform. By improving the ConvLSTM model, efficient collection and processing of gas data are achieved, monitoring accuracy and real-time performance are improved, and powerful support is provided for mine safety management.

The invention discloses a spoken English man-machine conversation interaction training method and system, and relates to the technical field of spoken English training, spoken English audio signals of a user are monitored in real time through a microphone, and the frequency spectrum information of the audio signals is refined into each frame of spectrogram and cepstrum coefficient feature vectors by using Fourier transform and feature extraction technologies; therefore, the voice signal characteristics of the user can be completely captured; three core dimensions of semantic similarity, grammatical structure similarity and fluency are incorporated into an evaluation model, a comprehensive oral English training evaluation system is formed, a text sequence is generated by analyzing a time sequence dependency relationship through a convolutional neural network, and the oral English training evaluation model of a deep learning technology is combined, so that the oral English training evaluation accuracy is improved. According to the method, the expressions of the user in the aspects of semantics, grammar and fluency can be accurately measured, the oral training level of the user can be comprehensively judged by the system by comparing the comprehensive evaluation coefficient with the preset qualified threshold value, and a personalized learning guide is provided according to a result.

The invention relates to the technical field of flood prediction, in particular to a non-consistent regional flood frequency analysis method based on the Bayesian theory. The method comprises the following steps: acquiring flood sequence data, meteorological data and drainage basin characteristic data of hydrological stations in a drainage basin; performing regional flood dynamic change analysis on the flood sequence data and the meteorological data to generate regional flood time sequence dynamic change data; on the basis of the regional flood time sequence dynamic change data, hydrological station flood characteristic parameters in the drainage basin are extracted from the drainage basin characteristic data, and hydrological station flood characteristic parameters in the region are obtained; and performing flood distribution parameter regression coefficient prior information derivation on the flood characteristic parameters of the hydrological stations in the region to obtain a maximum likelihood estimation value of a single hydrological station distribution parameter. According to the method, the Bayesian theory, the spatial correlation analysis and the spectral analysis are introduced, so that the accuracy and the applicability of flood frequency analysis based on the Bayesian theory are improved.

The invention discloses a water quality prediction optimization method based on variational multimodality, and relates to the technical field of variational multimodality, and the method comprises the following steps: carrying out P times of variational modal decomposition of random modal component quantity on a water consumption abnormity analysis request; calculating a coupling energy difference and a signal residual error between the modal component of each group and the water consumption time sequence signal, establishing a signal residual error change curve, and taking the number of the modal components corresponding to the signal residual error at the most prominent change part as the number of the optimal modal components; variational mode decomposition of the optimal mode component number is carried out on the water consumption anomaly analysis request, and optimized mode components of the optimal mode component number are obtained; drawing a modal component oscillogram for each optimized modal component, and performing data analysis on the modal component oscillogram to obtain a water consumption rule; according to the method and the device, the problems of insufficient decomposition and excessive decomposition in decomposition in the prior art are solved.

The embodiment of the invention provides a multi-modal feature-based video segmentation method. The method comprises the following steps of: respectively extracting image features, audio features and text features of a video frame, voice and a text of the voice; aligning the image features, the audio features and the text features through timestamps; fusing the aligned image features, audio features and text features to obtain a feature sequence with a timestamp; analyzing the feature sequence with the timestamp, and extracting feature mutation points in the feature sequence with the timestamp; filtering the feature mutation points to obtain segmentation points; and segmenting the video data based on the segmentation points to obtain a plurality of video clips. Through multi-modal feature extraction and fusion, video semantic structures are deeply observed, high similarity of fragment contents is ensured, videos are comprehensively and deeply understood, the analysis automation and intelligence level is improved, and it is ensured that results are accurate and timely. The problem of video segmentation is solved, precise self-adaptive segmentation is realized, and the subsequent analysis efficiency and precision are improved.

The invention provides a method and a system for judging a spontaneous combustion dangerous area of a goaf of a coal face, and relates to the technical field of coal mine safety monitoring. Geographic information system software is used for generating an initial geographic map, and a regular three-dimensional grid is created on the basis of the initial geographic map; accurate positioning and data consistency management of the goaf of the coal face are effectively realized, the abnormal sampling points can be accurately identified by setting a specific wavelength range, capturing spectral data of each sampling point by using an infrared spectrum and calculating the relative radiation intensity and the radiation deviation degree of each sampling point, and the accuracy of the abnormal sampling points is improved. Through obtaining neighbor sampling point data of an abnormal sampling point, calculating a uniform distribution index, realizing risk grade marking of the abnormal sampling point, comprehensively considering distribution density, relative radiation intensity and geographic position information of the abnormal sampling point, and generating a spontaneous combustion risk index, comprehensive assessment and timely early warning of goaf spontaneous combustion risk are realized.

The invention discloses a permanent magnet generator air gap eccentricity fault identification method and system, relates to the technical field of permanent magnet generators, and uses wavelet transform to extract time-frequency features and generate multi-band feature vectors, so that the system can more sensitively identify subtle features of air gap thickness deviation. And the extraction and analysis precision of characteristic signals of different frequency bands is effectively improved. And secondly, analyzing the air gap thickness deviation condition based on the feature vector, constructing an eccentricity degree index, and quantifying the eccentricity degree, so that the severity of the air gap eccentricity fault of the permanent magnet generator can be intuitively judged. And when the eccentricity degree exceeds a set threshold value, the system automatically sends a fault type checking instruction, so that the fault detection process is more intelligent. Afterwards, abnormal changes of all positions are analyzed through infrared thermal data and temperature gradients, air gap foreign matter faults possibly causing air gap eccentricity are further recognized, and the limitation of single signal analysis in traditional detection is made up.

The invention provides a precise wound medicine spraying robot control method and system based on a neural network. The method comprises the following steps: receiving operating room environment parameters and wound image information; determining the injection pressure of a robot nozzle according to the operating room environment parameters, and analyzing the wound image information by using a pre-trained neural network to obtain an image analysis result; a pre-trained neural network introduces a multi-scale analysis strategy for the wound area; according to an image analysis result, generating a three-dimensional motion trail of the robot nozzle and moving speeds and spraying angles of the robot nozzle at different positions in the three-dimensional motion trail; and according to the injection pressure of the robot nozzle, the three-dimensional motion track of the robot nozzle and the moving speed of the robot nozzle, a corresponding control instruction is generated, so that the robot nozzle performs corresponding operation according to the control instruction. According to the technical scheme, the accuracy of robot control is improved.

The invention provides a magnetic field observation method and system suitable for carrying out high-density deployment, and relates to the technical field of data processing, and the method comprises the steps: obtaining observation planning data, determining the number, interval and position distribution of observation nodes, setting a plurality of observation nodes to designated positions in a target region, magnetic field data in a target area is collected through observation nodes, preliminary screening is performed on the collected data according to background noise generated by equipment characteristics, secondary screening is performed according to spatial distribution characteristics among the observation nodes, and the observation data of different observation nodes are aggregated according to spatial position association rules. Performing cross validation on data consistency among observation nodes of the target area, performing optimization through a precision improvement algorithm to obtain a magnetic field distribution model, identifying a magnetic field abnormal area, performing multi-level detailed analysis on abnormal area data, and generating a magnetic field distribution result and abnormal characteristic data of the target area; the magnetic field observation precision is improved.

The invention discloses a movie and television play intelligent shrinking method based on a large model, and the method comprises the steps: S1, carrying out the division processing of a movie and television play video and a plot introduction document, completing the segment division of the movie and television play video, and the plot paragraph division of the plot introduction document; s2, completing matching between the video clip and the plot paragraph by using a large language model; s3, generating a reduced script according to the plot introduction document by using a large language model; s4, the big language model completes the selection of the video clip in the step S1 based on the reduced script obtained in the step S3 and the matching relationship between the plot paragraph and the video clip obtained in the step S2; and S5, performing cross-modal retrieval in the range of the video clip selected in the step S4 by using the reduced story plot content to obtain a video picture. By intelligently generating the reduced script, the automatic editing process is realized, the manual operation is reduced, and the editing efficiency is improved.

The embodiment of the invention provides a battery system health detection method, electronic equipment, a storage medium and a program product. The method comprises the following steps: acquiring voltage lag data of each battery module in the battery system and state parameters of the battery system; establishing a standard lag model corresponding to each battery module based on the state parameters and the voltage lag data; according to the state parameters and a standard lag model, standard lag time corresponding to each battery module is obtained through calculation; analyzing health state data corresponding to each battery module according to the standard lag time and the voltage lag data of each battery module; the method effectively reduces the calculation and analysis complexity, improves the analysis precision, can achieve the real-time monitoring and calculation, and directly reflects the dynamic performance and health condition of the battery system.

The invention discloses a visual characterization method for carbon emission of a sewage treatment plant, belongs to the field of carbon emission monitoring, and solves the problem that the fluctuation and trend of carbon emission data cannot be analyzed by an existing method. The method comprises the following steps: checking carbon emission intensity; preprocessing the carbon emission intensity data; defining a color mapping function according to the numerical range of the carbon emission, and mapping the numerical value to a color space; based on the color mapping function, creating a color bar for representing a corresponding relationship between the data value and the color; according to the color mapping function and the color bar, converting the carbon emission data at each time point into a corresponding color, and drawing the color on the heat map; and optimizing and outputting the image. According to the method, the calorific value map technology is introduced, so that the visualization effect of carbon emission intensity data is remarkably improved, the change of the carbon emission intensity in a time sequence is visually displayed, complex data becomes clear at a glance, and the peak value and the fluctuation trend of carbon emission can be quickly captured.

The invention discloses an electric power marketing service behavior supervision and control method and system, and relates to the technical field of behavior supervision, and the method comprises the steps: collecting audio, video and position information in an electric power access service process, carrying out the multi-modal data fusion analysis of the collected information, and comprehensively judging whether the service behavior is compliant or not; the wearing condition of the customer manager is monitored in real time through the AI image recognition technology, and whether the behavior is correct or not is judged based on the fused data; performing encrypted storage and non-tampering recording on behavior data of a customer manager in each service link based on a block chain technology, and performing intelligent labeling and tracking on key operation steps in a service process; and verifying the audio and video data provided by the customer manager through a deep counterfeiting detection technology, and outputting a final verification result. According to the invention, by introducing a multi-modal data fusion technology, an intelligent behavior analysis technology and a block chain storage technology, brand new improvement of the power management and control system in customer manager service behavior supervision is realized.

The invention discloses a power grid equipment maintenance plan analysis method and device, equipment and a medium, and the method comprises the steps: obtaining to-be-processed power grid equipment data corresponding to a target power supply region, and carrying out the preprocessing of the to-be-processed power grid equipment data, and determining target power grid equipment data; generating an equipment topology ledger corresponding to the target power supply area based on the target power grid equipment data, and determining an equipment maintenance plan corresponding to the target power supply area; and determining a maintenance plan analysis result corresponding to the target power supply area according to the equipment maintenance plan and the equipment topology ledger. Based on the technical scheme, the power grid equipment maintenance plan is analyzed in combination with the equipment topology ledger, so that the analysis accuracy of the maintenance plan is improved, and the power supply reliability of a user and the safe and stable operation of a power system are effectively guaranteed.

The invention provides a mud-phase shale component quantitative analysis and calculation method, which relates to the technical field of geological mineral exploration and mainly comprises the steps of sample collection and pretreatment, X-ray diffraction pattern acquisition, pattern analysis and mineral component determination, quantitative calculation based on an innovative model, result correction and evaluation and the like. Through an innovative quantitative analysis model and a strict sample processing flow, the content of various mineral components in the mud-phase shale can be more accurately determined, analysis errors are reduced, more reliable data support is provided for geological research and resource exploration, and compared with a traditional method, the method has the advantages that operation steps are simplified, analysis efficiency is improved, and the method is suitable for popularization and application. According to the method, the quantitative analysis of the mud-phase shale components can be completed in a shorter time, the progress of research and exploration is accelerated, the dependence on complex and expensive chemical analysis equipment is reduced, the experiment cost and the human input are reduced, and meanwhile the utilization efficiency of resources is improved.

The invention relates to the technical field of lubricating oil analysis, in particular to a method and system for quantitative analysis of a lubricating oil additive. The method comprises the following steps: acquiring base oil data and lubricating oil additive data; performing mixed oil treatment based on the base oil data and the lubricating oil additive data to generate mixed lubricating oil additive data; performing mixed oil concentration detection on the mixed lubricating oil additive data to generate mixed oil concentration detection data; performing chromatographic column selection on the mixed lubricating oil additive data based on the mixed oil concentration detection data to generate chromatographic column selection data; and performing boiling point heating on the chromatographic column selection data to obtain mixed oil boiling point heating data. By optimizing data acquisition, chromatographic separation, component identification and quantitative analysis as well as result analysis and visualization, the accuracy of quantitative analysis of the lubricating oil additive is improved.

The invention relates to the technical field of mechanical testing, in particular to a triaxial mechanical property testing method and device for a sample piece. The method comprises the following steps: determining a buckling critical load of a sample piece according to characteristic parameters of the sample piece; inputting the buckling critical load into a transverse clamping device so as to apply a transverse clamping force to the sample piece through the transverse clamping device; controlling an axial force applying device to apply an axial force to the sample piece; in the process of applying the axial force, parameters of the transverse clamping device subjected to reverse action of the sample piece are collected in real time; the transverse clamping force is adjusted in real time according to the reverse action parameters; acquiring an image of the surface of the sample piece in real time; and identifying speckle deformation on the surface of the sample piece in the image. According to the invention, the mechanical property of the material under the negative stress triaxial degree can be accurately tested and obtained.

The invention belongs to the technical field of safety production management, and discloses a safety production management system, which comprises a production data processing module used for storing preprocessed production management data into a database; the production equipment management module is used for extracting the operation data of the production equipment from the database and predicting the health state of each piece of production equipment based on the operation data of the production equipment; the oil depot environment management module is used for analyzing an oil quantity change trend according to the oil depot environment monitoring data; and the production risk control management module is used for formulating and implementing risk control measures in combination with the abnormal behaviors of the personnel in the production field and the potential operation risk factors. Based on periodic law and trend analysis, the future deformation and settlement trend of the oil storage tank can be predicted, and then potential problems of the oil storage tank can be found in time.

According to the method and the system for analyzing the harm influence on the oil and gas pipeline based on the slope, the hidden danger attribute description characteristics of the specified slope matter are extracted from the slope hidden danger data example, and the hidden danger attribute description characteristics have uniqueness and can represent the position information of the specified slope matter. Furthermore, slope risk factors associated with the hidden danger attribute description characteristics of the specified slope matter can be analyzed from the slope hidden danger data based on an artificial intelligence analysis thread, so that an independent oil and gas pipeline hazard influence analysis result of the specified slope matter can be generated based on the slope risk factors obtained through analysis. According to the method, the hidden danger points of the slope can be accurately analyzed, so that the hidden danger points can be treated in advance, the safety of the oil and gas pipeline can be guaranteed, and subsequent potential safety hazards and economic losses caused by damage of the oil and gas pipeline due to geological disasters are avoided.

The invention relates to the field of embedded technologies, in particular to a WCET analysis method and device for an embedded multi-thread program. The method comprises the steps that synchronous operation in a multi-thread program under the POSIX thread standard is recognized, lock-related operation statements are recorded, and an expanded parallel program control flow diagram is generated to describe the synchronous dependency relationship between threads; based on a lock-protected critical region resource mutual exclusion access principle, calculating a program WCST caused by critical resource application of a thread; according to a control flow diagram of a parallel program, determining a parallel segment, running on other cores, of a certain program segment, taking the parallel segment as an interference set, and performing conflict analysis on the shared instruction cache through an abstract interpretation method; and inputting the analysis result as a parameter into a WCET calculation method IPET to obtain a WCET analysis result of the whole parallel program.

The invention provides a psychological health state assessment method and system based on AI, and belongs to the technical field of psychological health assessment. The method comprises the steps that S1, historical assessment questionnaires, historical facial expression images and historical emotion text data are acquired; s2, creating a facial expression recognition model and a text sentiment analysis model; s3, the two models are trained; s4, acquiring an actual evaluation questionnaire, an actual facial expression image and actual emotion text data, evaluating the actual evaluation questionnaire through a psychological evaluation scale to obtain a questionnaire evaluation score, and inputting the actual facial expression image into the facial expression recognition model to obtain an expression category. Inputting the actual sentiment text data into a text sentiment analysis model to obtain a sentiment category; and S5, generating a mental health state assessment result based on the questionnaire assessment score, the expression category and the emotion category. The method has the advantages that the efficiency, the specialty, the objectivity and the flexibility of mental health state assessment are greatly improved.

The embodiment of the invention provides a flying spot generation method and system and a medium, and the method comprises the steps: obtaining monitoring data, extracting monitoring data features, and generating a plurality of monitoring point data features; comparing the monitoring point data characteristics with a set standard monitoring waveform to obtain a data deviation rate; judging whether the data deviation rate is greater than or equal to a set deviation rate threshold value or not; if yes, generating flying spot data, transmitting the flying spot data to the terminal in real time, and generating a monitoring time node of the flying spot data; if yes, fitting the monitoring data to generate a monitoring waveform; by analyzing and comparing the monitoring data and comparing the judged data characteristics with the standard monitoring waveform, the monitoring data deviating from the standard monitoring waveform are accurately obtained, the flying spot data are obtained, the monitoring data are accurately analyzed, and the analysis precision is improved.