2531results about "Fault location" patented technology

The invention relates to the technical field of cable operation state detection, and particularly discloses a multi-source data driven cable operation state comprehensive evaluation method, which comprises the following steps of S1, adopting a layered distributed sensing network architecture, and deploying three types of core sensors at key nodes of a cable, through space-time calibration of the multi-source heterogeneous sensor, data consistency is improved, fusion deviation is eliminated, the problem of data islands of a traditional system is solved, and a precise evaluation foundation is laid; noise suppression and dynamic correlation modeling are adopted, environmental interference is stripped, a vibration and displacement coupling relation is quantified, limitation of a single parameter is broken through, heterogeneous fault features are captured, and evaluation comprehensiveness and sensitivity are improved; a self-adaptive threshold mechanism is constructed based on environment weight and historical data, the bottleneck of a fixed threshold is broken through, an evaluation standard is corrected along with equipment aging and environment change, misjudgment is avoided, and diagnosis robustness in different scenes is enhanced.

The invention relates to the technical field of data processing, and discloses a comprehensive evaluation method and system for the health degree of a distribution network cable. The method comprises the following steps: collecting cable joint multi-source monitoring signals, normalizing the monitoring signals to obtain a degradation degree feature vector, correcting multi-physics field coupling model parameters, obtaining a recessive degradation index through finite element calculation to obtain an enhanced feature vector, and performing time-frequency domain decomposition to extract multi-scale feature parameters to obtain a comprehensive feature matrix; a double attention mechanism calculates a feature weight and a time sequence correlation degree to obtain a deterioration trend prediction value, and fuzzy integral is fused with a multi-classifier output probability to obtain a health degree evaluation grade and an early warning result. According to the invention, the early defect identification accuracy and the degradation trend prediction precision are improved.

The invention relates to the technical field of power distribution protection cabinet fault diagnosis, and discloses an intelligent fault diagnosis and rapid isolation method for a power distribution protection cabinet, and the method comprises the steps: synchronously collecting electrical parameters, environment states and equipment motion signals in the cabinet as multi-source monitoring data when the power distribution protection cabinet operates; parameter fluctuation, state change and action time sequence features are extracted after preprocessing, cross-data association analysis is carried out through a multi-dimensional association model to generate an association feature set, a fault diagnosis information set is formed based on the identified abnormal types, fault positions and development trends, an isolation strategy is formulated according to severity levels and influence ranges, and a fault diagnosis result is obtained. And generating a rapid isolation instruction and dynamically adjusting along with fault evolution. According to the method, multi-source data fusion analysis is realized, faults can be accurately diagnosed and rapidly isolated, the operation safety and reliability of the power distribution protection cabinet are improved, and the method is suitable for fault processing of the power distribution protection cabinet in a power system.

The invention discloses an anomaly detection and intelligent diagnosis method, system and device based on digital power grid multi-source data and a medium, and belongs to the technical field of anomaly detection, and the method comprises the steps: obtaining multi-source operation data from a power grid operation process, carrying out the preprocessing, and generating a standardized data set; time sequence features are extracted based on historical data, a power grid state reference model is constructed, and normal operation states in different load scenes are represented; on the basis of deviation calculation of the standardized data set and the power grid state reference model, abnormal candidate signals are detected, and high-confidence-coefficient abnormal signals are screened and generated; determining an abnormal source based on the high-confidence abnormal signal in combination with a power grid topological structure, and performing analysis to obtain fault type information; and generating a control instruction according to the fault type information and issuing the control instruction to a power grid control system. According to the method, a complete technical scheme of multi-dimensional data fusion, dynamic deviation detection, high-confidence anomaly screening, anomaly source accurate positioning and fault type rapid diagnosis is realized.

The invention relates to the technical field of power grid monitoring, in particular to a power operation risk identification method, system and device based on multi-modal data fusion and a storage medium. In order to solve the problems of multi-modal data splitting, topological constraint missing and the like in traditional power disturbance analysis, an improved BERT model is constructed, and electrical signal time-frequency features and text semantic information are mapped to a unified vector space through a multi-modal embedding mechanism; a time sequence attention mechanism is adopted to establish a time dependency relationship between signals and texts, and a graph attention network is combined to realize risk propagation modeling under power grid topology constraints; and collaborative optimization of disturbance classification, risk prediction and trend analysis is carried out through a multi-task learning framework. The technical problems that heterogeneous data fusion is difficult and risk identification precision is insufficient are effectively solved, accurate identification and intelligent early warning of electric power operation risks are achieved, and the safe operation level of a power grid is improved.

The invention relates to the technical field of cable asset management and fault prediction, and discloses a cable fault positioning method based on deep learning clustering analysis test waveform characteristics, and the method comprises the steps: collecting waveform and environment data in a cable operation period, and constructing a historical feature library comprising waveform, environment and position features; a self-adaptive detection model is adopted, and parameters are dynamically adjusted to adapt to different working conditions; multi-dimensional feature fusion and matching analysis are combined; a fault point distance is calculated through a signal propagation model and a time difference positioning algorithm, precise positioning is realized by fusing environment compensation and multi-point cross validation, and a three-dimensional geographic coordinate is generated by combining a laying path; and after multiple verifications, a structured report containing a fault type, a risk level, a prediction position, confidence and operation and maintenance suggestions is generated. According to the system, intelligent monitoring, fault risk prediction, asset optimization management and operation and maintenance decision support of a cable operation state are realized, and scientificity and economy of cable management in a complex environment are improved.

The invention provides a power transmission network equipment fault diagnosis and life prediction method and system, and relates to the technical field of fault diagnosis, and the method comprises the steps: obtaining the time sequence electrical characteristic data of a plurality of monitoring nodes, carrying out the window segmentation and statistical characteristic extraction, building a dynamic association graph structure based on a space-time association constraint model, and obtaining the time sequence electrical characteristic data; and calculating the abnormal contribution degree of each node, marking candidate abnormal nodes, determining a fault propagation path through reverse tracing and path analysis, and finally outputting a fault positioning result. According to the invention, abnormal nodes can be accurately identified, a fault propagation path can be accurately tracked, and the accuracy and timeliness of power transmission network fault diagnosis are improved.

The invention relates to the technical field of power equipment state monitoring, and particularly discloses a power cable electrical performance detection method and system, and the method comprises the steps: synchronously collecting a broadband electromagnetic signal, a mechanical vibration signal and a temperature signal at a cable monitoring point; calculating a wavelet coherence coefficient between the signals through continuous wavelet transform, and constructing a multi-modal coupling tensor fusing amplitude and cross-modal time-frequency correlation characteristics; performing time slicing and high-order singular value decomposition on the tensor to obtain a time-varying core tensor sequence, mapping the time-varying core tensor sequence into a high-dimensional manifold curve, and generating a system state fingerprint by calculating local curvature distribution and topology invariants of the curve; inputting the fingerprints into a pre-trained defect prediction model, and directly outputting defect inoculation probability and evolution stage judgment; according to the method, the limitation that early weak defect detection is not sensitive in a traditional method is broken through, and early warning and accurate diagnosis of cable insulation latent defects are achieved.

The invention discloses a high-voltage cable operation state on-line monitoring, intelligent early warning and fault positioning system, particularly relates to the technical field of power equipment insulation monitoring, and is used for solving the problems that transient discharge signals generated in the cable insulation degradation process are difficult to effectively capture and accurate fault positioning cannot be realized in the prior art. The method comprises the following steps: collecting transient leakage current signals and traveling wave propagation characteristic parameters, performing phase correlation analysis on leakage current pulses and power frequency voltage to identify discharge types, and performing variational mode decomposition and Hilbert-Huang transform on the signals to respectively extract complexity characteristics and energy distribution characteristics; signal logic conflicts are judged by analyzing similarity and statistical distance among characteristics of a plurality of monitoring points, and collaborative diagnosis is started or traveling wave distance measurement is directly utilized to carry out insulation state evaluation and fault location. And finally, differential early warning levels are generated according to the diagnosis result, and fault line selection and positioning information is output to realize real-time monitoring, intelligent early warning and accurate positioning of the operation state of the high-voltage cable.

The invention relates to the technical field of power system fault positioning, in particular to a power distribution network fault accurate positioning method, device and system based on the traveling wave principle, and the method comprises the steps: deploying traveling wave detection equipment at two ends of a certain section of line of a power distribution network, and enabling the traveling wave detection equipment to be used for synchronously collecting potential fault traveling wave signals; analyzing based on the acquired potential fault traveling wave signal, and screening out a fault traveling wave; in the time union set of the local time windows at the two ends, marking the fault traveling wave with the maximum peak value in the single end as a direct wave, analyzing the waveform similarity between the direct wave and the fault traveling wave arriving after the direct wave, and screening out a reflected wave according to the peak value difference; and based on the arrival time of the direct waves and the reflected waves at the two ends, calculating the distance from the fault point to the two ends of the line through a positioning formula to position the fault point. The invention aims to improve the positioning precision of the fault point in the power distribution network by analyzing the distribution characteristics of the pseudo traveling wave signal.

The invention provides a lithium battery internal short circuit fault early warning and positioning method and system based on deep learning. The early warning and positioning method comprises the following steps: step S10, multi-dimensional data semantic acquisition and working condition self-adaptive preprocessing; step S20, collaborative extraction and alignment of cross-scale spatio-temporal features; step S30, performing dynamic threshold adaptive internal short circuit early warning and feedback optimization; s40, fault accurate positioning and verification of topology perception are carried out; and step S50, performing embedded collaborative optimization and self-diagnosis deployment. According to the invention, early weak signals of an internal short circuit fault can be found in time, and the efficiency and timeliness of battery safety monitoring are improved; the sensitivity of early warning is improved; the robustness in high-temperature, low-temperature and rapid charging and discharging environments is enhanced; and the positions of the single battery and the electrode with the fault can be accurately identified.

The invention relates to the technical field of power grid fault positioning, in particular to a rapid power grid fault positioning system based on digital twinning and AI, which obtains multi-dimensional operation data of each node of a power distribution network, constructs a structured data set, reflects the state of the system, and performs joint modeling on node features and a topological structure through a graph neural network. And reasoning to generate candidate fault nodes. And further constructing a digital twinborn model containing load flow calculation and disturbance injection, performing analogue simulation on candidate nodes, extracting an electrical quantity change rate, performing fuzzy similarity matching with operation data in a fault window, and accurately positioning a fault node. Through cooperation of three stages of prediction, simulation and matching, the problems of weak complex topology perception, poor perturbation identification, positioning offset and the like of the existing method are effectively solved, and the fault positioning precision and robustness are improved.

The invention provides a distribution network traveling wave fault point intelligent positioning method and system based on deep time sequence feature learning, and belongs to the technical field of intelligent power distribution detection based on deep learning. Firstly, a high-speed traveling wave sensor is arranged on a distribution line, multi-dimensional three-phase voltage and current signals are collected, and a data set of fault types, phases, grades and positions is constructed; then, fault state features are extracted through topology perception normalization, symmetric component mapping and two-channel time sequence modeling, and accurate recognition of fault types, related phases and grades is achieved through an attention mechanism; furthermore, a fault intelligent positioning model based on a topological graph is constructed, a tower sensing weight and a line information bearing weight are introduced, and space-time embedding is extracted through a graph attention network and an echo state network, so that line fault classification and accurate positioning are realized. Finally, through combination of off-line model training and on-line system deployment, real-time identification and positioning of power distribution network faults are realized, and accuracy, robustness and response speed of fault diagnosis are effectively improved.

The invention discloses a broadband oscillation identification method and system based on a multi-band Nyquist criterion, and the method comprises the steps: dividing the frequency of broadband oscillation into a low frequency band, a middle frequency band and a high frequency band with overlapped frequency bands, and obtaining the oscillation signals of a monitoring node in a power grid in different frequency bands; respectively calculating the equivalent impedance of the non-overlapped frequency band of each frequency band and the equivalent impedance of the overlapped frequency band in the two adjacent frequency bands so as to construct the equivalent impedance of the low frequency band, the middle frequency band and the high frequency band with the overlapped frequency bands; establishing a Nyquist criterion of each frequency band based on a transfer function of dynamic characteristics of each frequency band in an open-loop state of the system; a generalized Nyquist criterion of the overlapped frequency bands is constructed by using the multi-frequency-band coupling matrix; constructing a comprehensive decision function according to the recognition result of the generalized Nyquist criterion on the system stability, the broadband oscillation comprehensive characteristic quantity and the change rate of the high-frequency-band impedance module value along with the frequency; and according to the value of the comprehensive decision function and a threshold value, a broadband oscillation identification result is determined, and the broadband oscillation identification capability is improved.

The invention relates to the field of power distribution control, and particularly discloses a self-healing control system and method based on multi-agent collaboration, and the system comprises a regional agent and a local agent which are disposed in a power distribution network. Wherein the local intelligent agent is configured to obtain real-time operation data of a monitoring point in the power distribution network, and perform rapid fault detection on the real-time operation data; the regional intelligent agent is configured to fuse the fault analysis information of the plurality of local intelligent agents to determine fault position information when the power distribution network has a fault, generate a self-healing control strategy according to the fault position information, and send the self-healing control strategy to the local intelligent agents for execution; through cooperative work of the distributed intelligent agents, the problems of low response speed, low positioning precision, weak autonomous capability and the like in traditional power distribution network fault management are solved. In addition, the system does not need to depend on centralized control of a master station, can still operate autonomously when communication is interrupted, and improves the efficiency and reliability of fault processing through cooperation of multiple agents.

The invention relates to the technical field of cable fault diagnosis and positioning, in particular to a cable fault positioning method and system based on multi-source data space-time correlation, and the method comprises the steps: obtaining partial discharge, traveling wave and temperature multi-source monitoring data from a high-voltage cable-overhead line hybrid power transmission network; on the basis of space-time correlation analysis of partial discharge and temperature data, heat-electricity associated defect points are calibrated; comparing traveling waves of the cable and the overhead line, and identifying and inhibiting cross-domain propagation interference; positioning the traveling wave data which are not identified to obtain temporary coordinates, judging fault characteristics by fusing information such as sheath ring current and the like, and comparing the temporary coordinates with defect points to generate matching identification; and inputting multi-dimensional information such as fault features, defect matching identification and temperature into the fusion model, calculating the confidence coefficient of a fault mode, and outputting a cross validation positioning result. According to the invention, high-reliability positioning of mixed line faults and active early warning of insulation defects are realized by carrying out space-time correlation and fusion decision on multi-source monitoring quantities such as traveling waves, partial discharge, circulation, temperature and the like.

The invention relates to the technical field of fault prediction and health management, in particular to a primary and secondary fusion complete ring main unit fault diagnosis method and system. Comprising the following steps: acquiring three-phase instantaneous voltage and current signals in real time, and converting the signals into digital transient data; performing time window preprocessing on the digital transient data, and executing wavelet packet decomposition to generate a transient feature vector; calculating and analyzing the transient feature vector through a transient zero-sequence power direction method and a support vector machine model to generate a local diagnosis result; when the local diagnosis result is that cooperative positioning needs to be started, a transient current similarity coefficient and a transient waveform intensity difference coefficient are calculated based on the digital transient data, and a comprehensive positioning result is generated; determining a fault line according to the comprehensive positioning result, and generating a remote control command; and executing a fault isolation operation based on the remote control command, and executing a PHM process. According to the method, the double local transient diagnosis and the cross-terminal cooperative positioning are deeply fused, so that the high-precision determination of the boundary of the fault section is realized.

The invention discloses a wide-frequency-domain weak signal data acquisition method, system, equipment and medium, and relates to the technical field of power system monitoring and fault diagnosis, and the method comprises the steps: collecting an input signal, analyzing the signal characteristics in real time, obtaining the frequency composition and amplitude change information of the signal, and dynamically adjusting the sampling rate according to the signal characteristics. Performing noise reduction processing on the collected signals, eliminating noise interference and retaining effective signal components, performing time alignment processing on the data to form a data set with a unified time reference, extracting multi-category features based on the data set, performing fusion judgment, identifying whether the system is in a fault state, and when it is judged that a fault occurs, judging whether the system is in a fault state or not; if yes, fault analysis and positioning are executed, and an analysis result is generated and output. According to the method, fast Fourier transform analysis is carried out on the signals, the dominant frequency and harmonic components can be accurately recognized, the energy ratio can be calculated, and a reliable data basis is provided for signal feature extraction and fault diagnosis.

The invention relates to the technical field of fault monitoring, in particular to a cable joint fault monitoring system and method, and the system comprises a multi-parameter signal processing module, an instantaneous temperature rise judgment module, a discharge attenuation recognition module, a synchronous instability judgment module and a risk early warning output module. According to the invention, by performing time sequence alignment and analysis on various sensor data such as temperature, current, voltage and partial discharge of the cable joint, the operation state of the cable joint can be monitored in real time, abnormal conditions can be found in time, potential faults such as joint aging, insulation damage and poor contact can be identified, and the abnormal degree of temperature rise can be accurately evaluated; according to the method, the risk level is accurately divided, the attenuation trend of partial discharge is identified, the instability trend of the connector can be effectively judged in combination with the fluctuation of current and voltage, the instability state of the cable connector can be identified in advance under the condition that multiple indexes synchronously change, and the accuracy and timeliness of fault early warning are improved.

The invention relates to a power cable insulation state real-time monitoring and alarm device, and belongs to the technical field of power equipment state monitoring. Comprising a multi-source sensing unit which is installed at a cable body and a joint; the edge computing terminal is fixed in the cable well or the ring main unit; and the cloud early warning platform receives the processing data and outputs an alarm instruction. According to the invention, multi-parameter deep fusion is realized, four-dimensional monitoring of grounding current, temperature, partial discharge and humidity is initiated for the first time, and limitation of a single parameter is broken through; and an insulation health index IHI is proposed, and the insulation deterioration degree is quantitatively evaluated. According to the method, tan delta and IHI are calculated on site through an edge-cloud collaborative architecture edge terminal, so that the data transmission quantity is reduced; the cloud dynamic threshold value is adjusted, and the alarm accuracy is improved. According to the invention, intelligent early warning mechanism three-level alarm linkage protection equipment is arranged; and the self-calibration function ensures long-term reliability.

The invention provides a sheath grounding ring current online monitoring and fault source positioning diagnosis method and system, and relates to the technical field of power system fault diagnosis, and the method comprises the steps: obtaining multi-physical field time domain signals, extracting features, constructing a sheath electrical connectivity topological graph, carrying out the partitioning, building the energy dissipation space-time field distribution, and recognizing a suspected fault region; and analyzing a voltage-circulating current response relationship, and finally judging a fault type and positioning a fault source through a Lyapunov index. The cable sheath grounding circulation state can be accurately monitored, accurate positioning and type identification of a fault source are realized, and the cable operation reliability is improved.

The invention discloses a cable fault accurate positioning system based on an intelligent sensor system, and relates to the technical field of power fault positioning, and the cable fault accurate positioning system comprises a cloud monitoring center which is in communication connection with the following modules: a signal acquisition and processing module which is used for deploying the intelligent sensor system along a cable, operation environment data and traveling wave signals in the cable operation process are obtained, and the signals are preprocessed and stored in a cloud data warehouse. According to the invention, the front band-pass filter and the self-adaptive trap filter are combined to filter out high-frequency harmonic waves and specific-frequency interference signals, and the differential sampling technology is adopted to counteract common-mode interference, so that the signal-to-noise ratio of the signals is remarkably improved, a system can accurately capture and process weak traveling wave signals caused by a high-resistance grounding fault in a complex environment, and the fault detection accuracy is improved. Therefore, the fault positioning precision is greatly improved, the positioning error caused by signal interference is reduced, and the determination of the fault position is more reliable and accurate.

The invention relates to the technical field of wires and cables, and particularly discloses a wire and cable test system based on sensor data acquisition, which comprises a data acquisition module, an intelligent analysis module and a function output module. The intelligent analysis module integrates technologies such as multi-modal signal blind source separation, mechanism-data hybrid drive deep learning, federated learning and digital twinborn linkage and self-supervised learning optimization, and forms a layer-by-layer progressive analysis system. Interference components are effectively eliminated through a signal separation technology, meanwhile, the adaptability of the model to different working conditions and different cable types is improved by means of fusion of mechanisms and data, model collaborative optimization under multi-site data privacy protection is realized through federated learning, and a self-supervised learning-driven digital twinborn evolution mechanism is combined, so that the multi-site data privacy protection model is optimized. The problems that a traditional analysis technology is insufficient in generalization ability, simulation precision is prone to drifting, and privacy risks exist in data sharing are solved, and the accuracy of fault recognition and the long-term adaptability of the system are remarkably improved.

The invention discloses a power transmission line inspection abnormity early warning method based on multi-mode perception and digital twinning, and belongs to the technical field of power transmission line monitoring. The problems that in the prior art, inspection efficiency is low, multi-source data utilization is insufficient, early warning lags behind and closed-loop control is lacked are solved, and efficient on-site processing and feature fusion of data are achieved by deploying a multi-modal sensing terminal cluster and a layered edge computing node; a digital twinborn model fusing multi-physical field simulation and data-driven AI is constructed at a cloud end, so that virtual-real interaction and deep analysis of a physical mechanism and real data are realized; the health state of the power transmission line is evaluated based on the deep analysis result, multi-stage early warning can be generated accordingly, a reverse control instruction is triggered, closed-loop control from state perception and intelligent diagnosis to active intervention is finally formed, and the intelligent level and operation and maintenance safety efficiency of power transmission line inspection are remarkably improved.

The invention relates to the technical field of cable fault monitoring, in particular to a power cable fault monitoring method and system based on big data, and the method comprises the following steps: carrying out the optimization and integration of spatial layout and sampling time based on a cable monitoring node, recognizing a data segment with the reverse trend of a current peak and a voltage phase angle, and carrying out the recognition of a current peak; and further analyzing the synchronism of the current-carrying density and the temperature, extracting a periodic intersection distribution characteristic, screening a continuously progressive trend track, and forming a trend directivity factor. According to the method, the linkage accuracy among multi-point signals is guaranteed by combining synchronous acquisition of node data with time sequence calibration, the trend discrimination method performs step-by-step screening for collaborative fluctuations of different characteristic parameters, and a continuous evolution sequence of various types of tiny disturbances under a complex working condition is dynamically extracted; a temperature and current-carrying density correlation mode is adopted to highlight the progressiveness and the positioning capability of a fault precursor, and dynamic risk signals which are difficult to identify at the initial stage of change are effectively captured.

The invention provides a power transmission line sheath grounding ring current fault diagnosis method and system, and relates to the technical field of power transmission line fault diagnosis, and the method comprises the steps: obtaining multi-point current measurement data and timestamps; establishing a circulation component coupling relation to solve transfer characteristics; constructing a space-time incidence matrix to identify a propagation path; constructing a multi-dimensional feature space projection node feature vector to carry out fault discrimination; fault types are matched based on the topological constraint relation; and constructing a circulation gradient sequence to calculate a fault index. According to the invention, the grounding circulation abnormal point can be accurately identified, and the fault positioning accuracy and diagnosis efficiency are improved.

The invention belongs to the technical field of wireless communication networks, and particularly discloses a LoRa communication system for power distribution network line fault detection and a use method of the LoRa communication system for power distribution network line fault detection. For each node, continuously monitoring changes in voltage, current and temperature monitoring ranges of a line with high voltage acquisition precision and a specific current signal sampling rate, removing noise influence by combining an environmental interference filtering technology, and generating an original monitoring data set according to a preset data acquisition frequency; the data preprocessing module is used for carrying out standardization processing on the collected signals according to the original monitoring data set; the invention aims to solve the problem of logic association caused by mutual fusion of low monitoring precision, high abnormal false alarm rate, unstable data transmission and slow fault isolation response in a traditional power distribution system in the prior art.

The invention relates to the technical field of power distribution fault positioning, in particular to a power distribution single-phase earth fault positioning method and device based on multi-source information fusion. The device comprises a data acquisition module, a fault line selection module, a grading emergency module, a fault positioning module and a parameter optimization module. According to the method, the zero-sequence current is monitored in real time, the dynamic change curve is generated, the fault line is accurately identified in combination with amplitude and phase characteristics, electrical parameters, environmental factors and load importance are fused to carry out comprehensive risk assessment and start a hierarchical response mechanism, and the fault point of the fault line is accurately positioned by adopting a voltage drop method. And the monitoring frequency and the decision threshold value are adaptively adjusted according to the fault characteristics and historical data, so that response lag and insufficient precision of a traditional fault positioning method are avoided, full-flow intelligent management from fault detection, accurate positioning to parameter self-optimization is realized, the accuracy and efficiency of fault processing are improved, and safe and stable operation of a power grid is ensured.

The invention relates to the technical field of intelligent diagnosis, and discloses a micro-grid fault diagnosis method and system based on data driving and unsupervised learning. The method comprises the following steps: collecting current, voltage, temperature and power data of a micro-grid and constructing a time sequence matrix; inputting a time sequence prediction network and a time sequence reconstruction network, and performing parallel processing to obtain a prediction error and a reconstruction error; carrying out weighted fusion on the two errors and constructing a two-dimensional error space to judge normal fluctuation and fault abnormity; and extracting a state variable to generate a dynamic threshold to judge a fault. The false alarm rate and the missing report rate of fault diagnosis are reduced.

The invention belongs to the technical field of intelligent operation and maintenance and automatic inspection of power equipment, and particularly discloses a power transformation and distribution station room inspection method and device based on multi-mode visual perception. The method comprises the following steps: controlling the intelligent inspection robot to move and synchronously acquiring visible light, infrared thermal imaging, partial discharge ultrasound and other multi-mode sensing data streams; generating an ultrasonic-guided enhanced infrared image, a power equipment structure map and an ambient light field distribution map through deep fusion in combination with ambient illumination parameters; and on the basis of the deeply fused information, generating a refined health state level of the power equipment, and finally automatically generating an inspection report conforming to the regulation. According to the invention, through deep cooperation and mutual verification of the multi-modal data, the perception robustness and diagnosis accuracy of early faults and abnormal states of equipment in a complex environment are improved, intelligent inspection from later judgment to beforehand prediction is realized, and the power supply safety of a power transformation and distribution station room is guaranteed.