141 results about "Conjoint analysis" patented technology

The invention relates to the technical field of electrical fault diagnosis, and discloses an electrical system fault diagnosis method based on multiple parameters. According to the method, multi-dimensional operation parameters of an electrical system are collected, wherein the multi-dimensional operation parameters comprise voltage waveform data, current waveform data and temperature distribution data; time-frequency conjoint analysis is carried out on the multi-dimensional operation parameters, and multi-scale electrical features are extracted and comprise steady-state feature components and transient feature components; based on a historical fault case library, performing mode matching on the multi-scale electrical characteristics to generate an initial fault type set; performing confidence evaluation on the fault types in the initial fault type set by using a dynamic weight distribution algorithm, and screening out high-confidence fault types; and according to the high-confidence fault type, a fault evolution path model is constructed, and the fault evolution path model is used for describing a time sequence of fault features.

The invention relates to the technical field of medical information, in particular to a medical data conjoint analysis system based on medical knowledge graph driving. The system comprises a medical knowledge graph construction module, a medical data quality evaluation module, a medical feature engineering module and a medical knowledge driven analysis module, and heterogeneous knowledge graph modeling can be performed by integrating medical clinical guidelines, biomedical literatures, a drug database and historical medical data so as to generate a medical heterogeneous knowledge graph; obtaining new medical clinical test information and carrying out delayed contradictory learning update to generate a medical dynamic update knowledge graph; the method comprises the following steps: obtaining multi-modal medical data, carrying out quality verification evaluation and medical feature engineering analysis, carrying out knowledge path joint driving analysis at the same time, generating a decision support reasoning path corresponding to medical clinical knowledge, and outputting a corresponding medical knowledge path confidence coefficient. According to the method, fusion reasoning among cross-source data can be realized by constructing the multi-modal medical knowledge graph.

The invention discloses a multi-dimensional natural resource intelligent monitoring method and system based on big data analysis, and relates to the technical field of resource monitoring, and the method comprises the steps: calculating the information sharing degree between different data streams, constructing an abnormal feature credibility distribution map, and carrying out the calculation of the abnormal feature credibility distribution map; and inputting the abnormal feature credibility distribution graph and the multi-source auxiliary information into an integrated classifier, carrying out joint analysis and comprehensive research and judgment on multi-dimensional information to output a target list, constructing an event response relation graph reflecting a monitoring task priority and an execution logic relation according to the attribute features of each element in the target list, and carrying out event response analysis on the monitoring task priority and the execution logic relation. Wherein each node represents a to-be-monitored target, an edge represents a dependency relationship between tasks, and an optimal monitoring execution path is planned for various intelligent monitoring carriers by using an improved A algorithm integrated with a multi-factor cost function. Through natural resource monitoring of multi-modal data interference correction, intelligent classification and path optimization and dynamic response, the monitoring quality and the execution efficiency are effectively improved.

The invention relates to a water pump scheduling method for estimating lift based on data driving, which comprises the following steps: calling structural information and sensing data, calculating friction loss and structural symmetry, extracting differential pressure data flow gradient, judging load matching degree, activating path nodes, identifying differential pressure data change sections, and mapping a calculation path. According to the method, dynamic correction and compensation of input parameters are achieved by means of conjoint analysis of the structural features of the water inlet and the water outlet and flow pressure difference data, and comprehensive judgment of the working condition matching degree and the periodic change state is combined, so that the flow resistance trend is judged, and the calculation result is corrected. The adaptability to cold load fluctuation and loop temperature difference change is enhanced, the response capability of lift calculation to water body characteristic change is improved by introducing flow resistance trend and viscous state factors, and the calculation stability of the model in different operation sections and the calculation accuracy of the model in complex working conditions are optimized by using a path activation configuration and output switching mechanism.

The invention relates to the technical field of multi-modal data, in particular to a multi-modal learning method which comprises the following steps: S1, acquiring multi-modal data from various sensors and operation logs through a data acquisition module; s2, preprocessing the multi-modal data through a data processing and fusion module, and performing multi-modal feature fusion by using a cross-modal attention mechanism to obtain fusion features; when the method is used, through joint feature representation, recognition of various faults is achieved, the multi-modal recognition performance is improved, then the accuracy is improved, the method adapts to the change of an operation environment through a dynamic threshold mechanism, misinformation and missing report are effectively reduced, the response speed of operation and maintenance personnel is improved through visual report and task distribution, the operation and maintenance efficiency is conveniently improved, and the operation and maintenance efficiency is improved. Through conjoint analysis of a thermodynamic diagram and stress data, local overheating is accurately detected, an overheating fault is conveniently recognized in real time, meanwhile, the specific position and time of an overheating area are marked in real time, and accurate fault positioning is facilitated.

The invention discloses a monitoring and prevention method for comprehensive prevention and control of tobacco viruses. The monitoring and prevention method comprises the following steps: S1, collecting tobacco virus monitoring data; s2, analyzing a virus infection risk index through the monitoring data; s3, preparing a comprehensive virus control agent; s4, collecting leaf states of the tobacco plants; s5, judging an infection propagation direction by combining multi-dimensional data; according to the method, concentration data of root system metabolites and leaf VOCs are collected at the same time, and multi-parameter conjoint analysis of virus invasion is achieved; through joint analysis of VOCs concentration data in the orientation, root system metabolites and leaf apparent parameters, the numerical value change trend in the virus propagation direction is quantified, and accurate positioning of a propagation path is achieved; through multi-dimensional data fusion, dynamic model quantitative analysis and physical blocking and medicament cooperative prevention and control, early warning, accurate judgment, propagation direction quantification and efficient prevention and control of tobacco virus infringement are realized, and a complete closed loop from monitoring to intervention is formed.

The invention relates to the technical field of fault prediction, and discloses an excitation online monitoring system, which comprises an information acquisition module, a data processing module, a mechanical wear quantification module, a shaft vibration health analysis module, a feature fusion module, a mechanical health analysis module and an early warning instruction generation module, collecting multi-source operation data in the gas generator set in real time; performing time-frequency domain conjoint analysis on the excitation current waveform to obtain a current high-frequency ripple component; carrying out harmonic distortion rate fusion on the current high-frequency ripple component to obtain a current ripple characteristic coefficient; dynamically correcting a preset rotor vibration threshold value, and comparing the rotor shaft vibration signal with the dynamically corrected rotor vibration threshold value to obtain a shaft vibration health index; performing cross-domain feature fusion to obtain fused features; carrying out the normalization processing of the fusion features, obtaining a comprehensive health state score, and generating a PHM early warning instruction of a fault block in the excitation primitive; according to the invention, the excitation monitoring accuracy of the gas generator set can be improved.

The embodiment of the invention discloses a water resource dynamic evolution simulation analysis method and system based on digital twinning, and the method comprises the steps: calling a target drainage basin water resource monitoring data set from a preset distributed database cluster after obtaining the authorization authentication of a digital twinning server; performing space-time correlation processing on the data set and the meteorological environment data of the target watershed to obtain a conjoint analysis data set with a unified space-time reference system; performing time-space coupled water resource cyclic evolution simulation on the conjoint analysis data set through a preset MIKESHE model to obtain a target drainage basin water resource dynamic evolution process simulation result; according to the result, a quantitative influence simulation result of the climate change on the agricultural irrigation water resource of the target watershed is generated, the result comprises an irrigation water demand change trend, water resource supply and demand balance and agricultural water efficiency fluctuation characteristics, and a scientific basis is provided for agricultural irrigation water resource management.

The invention relates to the technical field of electrical equipment fault diagnosis, in particular to an acoustic-electric joint detection method, system and device for internal defects of GIS equipment and a medium, and the method comprises the steps: synchronously collecting an ultrasonic signal detected by a grating fiber ultrasonic sensing module and an ultrahigh frequency electromagnetic wave signal detected by an ultrahigh frequency sensing module; performing feature extraction on the ultrasonic signal and the ultrahigh-frequency electromagnetic wave signal to obtain a first feature parameter and a second feature parameter respectively; based on the arrival time difference of the ultrahigh-frequency electromagnetic wave signal and the ultrasonic wave signal, calculating the initial space position of the defect by adopting a time difference positioning method; and performing conjoint analysis on the first characteristic parameter and the second characteristic parameter in combination with the initial space position of the defect so as to identify the defect type. Through the arrangement, sound and electricity detection information can be deeply fused, a novel joint detection method with mutual verification and mutual promotion of positioning and identification is realized, and the integrity, accuracy and reliability of GIS internal defect diagnosis are effectively improved.

According to the method, a navigation satellite is randomly selected as a pseudo communication satellite, other navigation satellites are randomly selected to form a prediction combination, multi-dimensional feature vectors of the prediction combination are extracted, the multi-dimensional feature vectors comprise spatial distribution features and correlation features and statistical features of a signal-to-noise ratio time sequence in a sliding window, and according to a scoring model, a prediction result is obtained. And obtaining an optimal prediction combination of the pseudo communication satellites, and further executing deep learning training on the prediction model. In the actual satellite communication process, according to the spatial position of the communication satellite and the optimal prediction combination, multi-dimensional feature vectors are extracted, the trained deep learning prediction model is used for predicting the signal-to-noise ratio time sequence of the communication satellite, whether communication is conducted or not is selected according to the prediction result, a communication strategy is adjusted, and the deep learning prediction model is updated in a self-adaptive mode. According to the method, the combination quality can be rapidly evaluated and predicted without completely training a complex deep learning model, the calculation cost is reduced, the bit error rate is greatly reduced compared with a traditional fixed strategy, and the adaptive capacity to the environment is enhanced.

The invention relates to the technical field of edge calculation data processing, in particular to an edge calculation data processing system for predicting the state of a drainage branch pipe, and the system is provided with a data acquisition module, a data analysis module, a joint analysis module and an edge calculation regulation and control module. The data analysis module is used for screening characteristic monitoring nodes and determining pipeline anchor point paths of a plurality of drainage branch pipes; the joint analysis module is used for screening missing monitoring nodes and judging whether the drainage branch pipes have blockage risks or not; the edge calculation regulation and control module is used for determining risk nodes; and determining a data transmission sequence for performing data transmission on the pipeline operation characterization parameters of the risk nodes. The monitoring nodes are subjected to conjoint analysis, the states of the drainage branch pipes are predicted, the monitoring nodes needing data uploading are screened, the data transmission sequence of the monitoring nodes is adaptively adjusted, and the processing efficiency of related data of the drainage system is improved.

The invention discloses a cable partial discharge accurate positioning method and system based on conjoint analysis, and relates to the technical field of cable discharge positioning, and the method comprises the steps: constructing a multi-source sensing network which is composed of a UHF sensor, an ultrasonic sensor, and a transient voltage / current sensor; obtaining a geometric path, a shielding layer and a dielectric distribution parameter of the cable, executing digital twin modeling of the cable, and establishing a propagation time difference and interference compensation model; accessing a time synchronization network, and reading an acquisition signal set; multi-source signal partial discharge positioning identification is executed, and a fuzzy partial discharge area is established; constructing an error objective function by using the observation time difference and the prediction time difference; and game search is carried out to establish partial discharge positioning. The technical problems that in the prior art, cable information collection is limited, cable structures and medium parameters are difficult to comprehensively analyze, and cable discharge positioning accuracy and reliability are insufficient are solved, and the technical effect of improving cable partial discharge positioning accuracy and running state monitoring reliability is achieved.

The invention discloses a multi-modal learning data conjoint analysis method and system, a medium and a product, and relates to the field of multi-modal learning analysis. Comprising the following steps: acquiring synchronously acquired multi-modal data and extracting a feature vector; and calculating a confidence score of each modal feature vector and a semantic conflict coefficient between the modal feature vectors. When the semantic conflict coefficient is greater than a preset conflict threshold value, determining the modal feature vector with the maximum confidence score as a final state feature vector; and when the semantic conflict coefficient is smaller than or equal to the threshold value, calculating a fusion weight according to the confidence score, the semantic conflict coefficient and the task priority parameter, performing weighted fusion on each modal feature vector to obtain a final state feature vector, and generating an evaluation result according to the final state feature vector. According to the invention, through a decision-making mechanism that the optimal information source is selected in high conflict and context adaptive fusion is carried out in low conflict, the problem of inaccurate evaluation caused by forced fusion of contradictory signals is solved, and the accuracy and robustness of learning state evaluation in a complex scene are improved.

The invention discloses an intelligent code writing and quality assurance system for multi-language code analysis, which relates to the technical field of intelligent code development and quality control and comprises a multi-language code analysis engine, an intelligent engineering management module, a self-adaptive compiler, an interactive console, a multi-dimensional instance library and a personalized configuration center, each module realizes real-time data interaction through a distributed data bus; the system supports more than 15 mainstream programming languages, new language support can be expanded through plug-ins, and the multi-language code analysis engine integrates static analysis, dynamic tracking and cross-language call detection functions. Static analysis identifies grammar errors and undefined variables through abstract syntax tree analysis. According to the intelligent code writing and quality assurance system for multi-language code analysis, the single-language limitation of a traditional system is broken through, conjoint analysis of programming languages is achieved, cross-language defect recognition accuracy is improved, and potential risks of multi-language projects are remarkably reduced.

The invention relates to the field of anomaly detection, in particular to a process execution anomaly detection method based on data driving, and the method comprises the steps: carrying out the statistical modeling analysis of historical normal process cycle characteristic data, and obtaining a historical reference parameter set; obtaining a drift compensation vector by performing robust center estimation and stability evaluation analysis on the current batch of process feature data; performing conjoint analysis on local sparseness and neighborhood eccentricity on the compensated process feature data to obtain a structural relaxation amplification coefficient; performing dynamic self-adaptive equivalent distance construction analysis on the compensated process feature data to obtain a process anomaly clustering result; through comprehensive judgment and analysis of a process anomaly clustering result and a drift compensation vector, an equipment anomaly detection result is obtained, so that the problem that an existing anomaly detection algorithm based on static distance measurement is easy to generate false report and missing report under the coexistence condition of equipment global reference drift and internal structure relaxation is solved.

The invention relates to the technical field of power equipment state monitoring and fault diagnosis, and discloses a fusion type high-voltage switch cabinet insulation state quantitative monitoring method and system, and the method comprises the steps: firstly obtaining a partial discharge ultrahigh frequency and sound wave signal, and calculating a generalized Renyi spectrum entropy and an entropy gradient index; carrying out nonlinear correction on the medium sound wave propagation velocity by using the entropy gradient index, establishing an equivalent sound velocity model in a medium degradation state, and obtaining a standardized equivalent time difference; then, constructing a two-dimensional phase plane state space taking the standardized equivalent time difference and the generalized Renyi spectral entropy as dimensions, and analyzing a differential evolution trajectory of a state vector on a time sequence; and finally, constructing a comprehensive state loss function, mapping to generate an insulation health degree index, and carrying out life prediction. According to the invention, through medium degradation wave velocity constitutive mapping and phase plane conjoint analysis, monitoring errors caused by medium aging are solved, and dynamic accurate quantitative evaluation of the insulation state is realized.

The invention relates to the field of data processing, in particular to an enterprise business metadata intelligent management method based on multi-source heterogeneous data, which comprises the following steps: performing event and blood relationship association analysis on enterprise multi-source heterogeneous data to obtain an enterprise business metadata dynamic monitoring data set; carrying out conjoint analysis on the time neighborhood of the metadata change event and the propagation characteristics of the influence domain to obtain a danger rate regulation factor; performing fusion analysis on the cross-window stability and the structure gradual change feature of the metadata change event to obtain a gradual change regulation factor; carrying out conjoint analysis on the danger rate regulation factor and the slow change regulation factor to obtain a dynamic danger rate; through joint distribution of the dynamic danger rate and the detection posteriori, comprehensive detection output is obtained, and therefore the problem that an existing Bayesian online change point detection algorithm cannot timely and accurately recognize real metadata changes in a multi-source heterogeneous environment is solved.

According to the comprehensive evaluation method and system for the power production technical improvement project and the medium, expert preferences are quantized through joint analysis, historical project data rules are mined in combination with a deep neural network, deep fusion of subjective and objective information is achieved, authority of expert experience is reserved through a fuzzy integral fusion mechanism, and the comprehensive evaluation efficiency is improved. And objective laws in the data are fully utilized, so that comprehensive, deep and dynamic assessment and evaluation can be carried out on the power production technical improvement project, scientific, objective and prospective assessment and evaluation can be carried out on the project, the defects of a current method are effectively made up, and the method is suitable for popularization and application. And a more powerful, scientific and accurate support is provided for the management decision of a power production technical improvement project. The method is not only suitable for power production technical improvement project evaluation, but also can be expanded to complex project evaluation scenes of other industries, and has good universality and popularization value.

The invention discloses an AI-based spoken language dialogue quality evaluation method, and relates to the technical field of artificial intelligence and natural language processing, and the method comprises the steps: receiving a spoken language dialogue audio stream, extracting text content information and acoustic rhythm information, carrying out the correlation fusion according to a timestamp, and generating a multi-mode dialogue data sequence; carrying out conjoint analysis on the dialogue logic relation graph and the dynamic memory bank, and calculating a dialogue structure consistency index to obtain a comprehensive quality evaluation score; and based on the comprehensive quality evaluation score and the dynamic memory library, positioning contradictory nodes and contexts in the dialogue logic relation graph, constructing a local consistency reconstruction task, and updating the memory enhancement neural network by using the local consistency reconstruction task. According to the method, the memory enhancement neural network is updated by using the local consistency reconstruction task, adaptive learning and sustainable evolution of the memory enhancement neural network based on actual dialogue contradictions are realized, and the consistency detection accuracy and the dynamic ability of adaptability are improved.

The invention provides a transformer operation and maintenance management method and system based on digital twinning, and relates to the technical field of power equipment operation and maintenance, and the method comprises the steps: carrying out the multi-source integration of transformer sampling data, and obtaining traceable state monitoring data through combining with a distributed block chain technology; adaptively decomposing the monitoring data according to frequency bands and extracting time-varying features, and introducing a cross-frequency domain feature fusion device based on mutual information maximization to analyze relevance of different frequency bands to obtain a state index matrix; constructing a state evolution diagram of the transformer parts, analyzing state propagation characteristics among the parts by adopting dynamic causal reasoning, extracting multi-scale time sequence characteristics, and performing layered interactive coupling to realize space-time conjoint analysis; and calculating the reliability index of each component based on probability distribution, tracing degradation time nodes and influence factors, and formulating a maintenance strategy. According to the invention, omnibearing monitoring and accurate prediction of the state of the transformer are realized, and the operation and maintenance efficiency and the equipment reliability are improved.

The invention provides a radar wind profile estimation method based on fusion of multi-domain features and a cross-domain attention mechanism, and the method comprises the steps: converting radar echo data into a time domain, a frequency domain and a fractional order domain, and carrying out the feature extraction through employing a convolutional neural network; and a cross-domain cross attention module (ICAM) is inserted into the feature extraction network to fuse the features of the time domain, the frequency domain and the fractional order domain. The ICAM combines channel attention and self-attention mechanisms, dynamically weights multi-domain features, highlights key information and inhibits redundant components; and inputting the fused feature vectors into a full-connection network, carrying out regression calculation on the weft-wise and warp-wise wind speed components of each height layer, and outputting a vertical wind profile. Through multi-domain conjoint analysis and intelligent feature fusion, the wind profile estimation precision and robustness in a low signal-to-noise ratio and non-stationary signal environment are significantly improved.

The invention relates to the technical field of wireless communication network propagation model intelligent optimization, and discloses a platform wireless network ray tracking propagation optimization method based on deep learning, which comprises the following steps: acquiring train arrival and departure time table data and wireless system base station parameters, and generating a multi-train multi-band joint scene configuration table; evaluating the interference level of each frequency band for each observation point in the slice-level multi-frequency-band field intensity joint distribution matrix, and generating multi-frequency-band interference time sequence distribution data and a cross-frequency-band coverage correlation graph; performing conjoint analysis on the multi-band interference time sequence distribution data to generate a multi-band joint failure risk moment identifier and an interference attribution result; and outputting a heterogeneous network cooperative scheduling suggestion for the multi-band joint failure risk area. According to the method, the factor that the influence of multi-vehicle shielding on the difference of different frequency bands cannot be reflected by simple combination after independent simulation of each frequency band is overcome, and a multi-frequency-band joint evaluation basis is provided for unified scheduling decision-making of a heterogeneous network.

The invention relates to the field of big data, and provides an anomaly detection and repair method and system for a big data task and a medium. The big data task anomaly detection and repair method comprises the steps that task information of a big data task with an abnormal task state is obtained, wherein the task information comprises a task log; obtaining an execution code of the big data task through a decompilation tool, and analyzing the execution code and the task log through an exception analysis module to identify an exception problem; in response to the fact that the exception problems are data source exception and code exception, exception code snippets, data source information and exception types are analyzed through a conjoint analysis module to determine exception root causes, the exception code snippets are recognized through an exception analysis module, the data source information is determined by executing codes, and the exception types are determined according to task logs; and triggering abnormal code repair and / or data processing optimization according to the abnormal root cause.

The invention provides a multi-modal and multi-task-based conjoint analysis system and method, and belongs to the technical field of environmental acoustic perception, a bimodal feature decoupling architecture is adopted, dynamic separation of modal invariant features and modal specific features is realized in text-audio cross-modal fusion, and a scene adapter automatically adjusts bimodal interaction strength; for time sequence positioning characteristics of text-to-audio positioning and semantic understanding characteristics of sound scene classification, an expert network structure is adopted to ensure that scene classification utilizes fused characteristics for reasoning, and meanwhile, a time sequence structure of original audio is reserved for a text-to-audio positioning task. A multi-task learning scheme is adopted for text-to-audio positioning and sound scene classification tasks. According to the invention, key challenges in intelligent city acoustic perception are effectively solved. Dynamic decoupling and recombination of audio-text features are realized; and the problem of negative migration in multi-task learning is solved.

The present application relates to the technical field of data query, in particular to a multi-index horizontal expansion data processing method and system, comprising the following steps: collecting time series and dimension attributes to generate a business detail wide table, matching non-empty measure fields to form independent index logic, combining rule nodes to build a general materialized architecture, and outputting multi-index processing results through scheduling mapping and window aggregation. In the present application, a wide table is constructed in the model layer and atomic and derived indexes are centrally managed, realizing a unified index definition method independent of dimension combination. When a user queries, dimension parameters are dynamically combined by the query engine to automatically build an aggregation statement and calculate the result, avoiding the problem of derived index explosion caused by the expansion of the number of dimensions. Meanwhile, multi-index joint analysis is completed in the same materialized table without the need for multiple table joins, improving query execution efficiency, having high adaptability to index updates and dimension changes, and significantly reducing data development and operation and maintenance burden.

The invention discloses a gene level DNA methylation, transcriptome and proteome conjoint analysis method based on a snkemake framework, which comprises the following steps: preprocessing original data of DNA methylation and / or transcriptome and / or proteome, analyzing gene methylation, analyzing transcriptome and / or proteome, and analyzing the transcriptome and proteome. Based on gene methylation analysis data, the preprocessed transcriptome data and the preprocessed proteome data, correlation analysis and / or difference intersection analysis and / or enrichment pathway conjoint analysis are / is carried out, finally analysis results are sorted, and a visual report is generated. The method has comprehensive results, and relates to quantification, difference analysis and pathway enrichment analysis of gene methylation, and correlation analysis, difference intersection analysis and enrichment pathway joint analysis of integrated DNA methylation and / or transcriptome and / or proteome data; automatic arrangement, verification, visualization and report generation of analysis results are realized; all operation steps can be traced, and corresponding analysis log records are generated. The invention further discloses a related system and application.

The invention relates to the field of soil classification, in particular to a saline-alkali soil intelligent classification method based on an improvement measure effect, and the method comprises the steps: carrying out the collection and adaptive calculation of land parcel multi-source feature data, and obtaining a basic parameter set; performing difference analysis on the plot obstacle factor structure to obtain a macroscopic obstacle mode correction factor; carrying out conjoint analysis on the land parcel improvement demand degree and the adjustment characteristics to obtain a microscopic improvement feasibility correction factor; performing fusion calculation on the Euclidean distance and the correction factor to obtain an improved effect correction distance; and performing clustering analysis on the improvement effect correction distance to obtain an intelligent classification result of the saline-alkali soil, thereby solving the problems of land parcel improvement mode identification distortion and measure matching error caused by distance measurement isotropy in an existing clustering algorithm.

The invention provides a brain dynamic mode classification method based on a graph auto-encoder and soft and hard clustering. The brain dynamic mode classification method comprises the steps that S1, data preprocessing and blood oxygen level dependence signal extraction are carried out; s2, constructing a dynamic brain network; s3, nonlinear dimensionality reduction of the graph auto-encoder is carried out; s4, soft and hard clustering conjoint analysis and dynamic mode feature extraction; s5, performing feature screening and validity verification; and S6, classifier training verification and classification result output. According to the method, rs-fMRI data is taken as core input, accurate classification of brain dynamic modes is realized through a whole-process design of data preprocessing, brain network construction, nonlinear dimension reduction, clustering analysis and classification verification, and the problem of low classification accuracy of a traditional magnetic resonance image data classification method is solved.

The invention discloses a power transmission line traveling wave fault positioning method and system based on data analysis, and relates to the technical field of big data analysis, and the method comprises the steps: firstly collecting fault original records, temperature, humidity and weather states, and building a historical library; carrying out time-frequency domain conjoint analysis on the traveling wave signal, and calculating a phase difference and a time difference; the signals are adaptively decomposed to different frequency bands to calculate fractal dimensions, and advanced features are extracted in combination with a deep convolutional neural network; associating the phase difference, the instantaneous electric field intensity and the magnetic field rotation direction to construct a three-dimensional matrix rotation graph structure, and forming a high-dimensional feature vector by using a graph attention network; and calculating a fault posterior probability according to the high-dimensional vector, dynamically adjusting a threshold value according to an environment parameter, and generating an adversarial network virtual sample to match a fault type when the probability is insufficient. The system comprises a data acquisition module, a time-frequency analysis module, a feature extraction module, a feature association module and a fault determination module, all the modules cooperate to realize data acquisition, feature analysis and fault positioning, and the positioning reliability is enhanced.

The invention discloses a concrete filled steel tube interface void detection method based on head and tail wave conjoint analysis. A plurality of piezoelectric sensors are uniformly arranged on the inner wall of a steel pipe, modulation pulse or sine wave signals are adopted to cyclically excite the piezoelectric sensors, and response signals of the opposite side and the corresponding adjacent sensors are collected. Under the assumption that both the steel pipe and the concrete are uniform media, the characteristic change of the head wave signal mainly reflects the interface state on the propagation path, and can be used for identifying the interface void of the area where the excitation or receiving sensor is located; as the tail wave signal contains multiple reflection and scattering components, the symmetrical path contrastive analysis is carried out on the tail wave signal, and the medium change of an indirect area between the sensors can be effectively sensed, so that the void condition of a corresponding position is judged. According to the method, quantitative recognition of the head wave on the path defect and high-sensitivity sensing of the tail wave on the change of the adjacent region are combined, so that a completely-covered void detection system is constructed, and the technical problem that global monitoring is difficult to realize by a traditional method is solved.