Automatic detection method and system for electrical equipment

Abstract

The invention provides an automatic detection method and system for electrical equipment, and the method comprises the steps: arranging a plurality of sensor nodes, and extracting a characteristic harmonic component of a voltage signal of the sensor nodes; performing traceability analysis on a propagation path of voltage fluctuation according to an amplitude attenuation coefficient and a phase difference of a characteristic harmonic component between adjacent sensor nodes and a line impedance parameter to obtain a voltage characteristic propagation characteristic corresponding to each sensor node; based on the similarity deviation of the temperature change rate and the vibration spectrum between the adjacent sensor nodes and the physical connection relationship between the components, constructing an anomaly detection map of the electrical equipment; and performing fuzzy evaluation on the fault probability of the position of the sensor node according to the voltage characteristic propagation characteristics corresponding to the sensor node and the anomaly detection map to obtain the fault membership degree of the position of the sensor node, and performing automatic positioning on the fault position of the electrical equipment based on all the fault membership degrees. By adopting the scheme of the invention, automatic identification of the fault position of the electrical equipment based on the fault membership degree can be realized.

Description

Technical Field

[0001] This application relates to the field of fault detection technology, and more specifically, to an automated detection method and system for electrical equipment. Background Technology

[0002] With the continuous expansion of power systems and industrial automation equipment, the safe and stable operation of electrical equipment has become crucial to ensuring the reliability of production and power supply. During long-term operation, electrical equipment is prone to faults such as local overload, insulation aging, or mechanical loosening. If these faults are not detected and located in time, they may cause large-scale power outages or equipment damage. Traditional manual inspection and periodic testing methods have the drawbacks of low efficiency, limited coverage, and delayed response, making it difficult to meet the real-time and accuracy requirements of modern intelligent operation and maintenance.

[0003] In existing technologies, most fault detection methods rely on single sensor data or local indicators for judgment, such as triggering alarms solely based on abnormal voltage or temperature thresholds. These methods fail to comprehensively reflect the complex changes in the internal state of equipment, especially when multiple nodes and features interact, easily leading to false alarms or missed alarms. Furthermore, existing methods often lack comprehensive analysis tools for fault propagation paths and component relationships, making it difficult to accurately quantify the fault probability of each node, resulting in inaccurate location results and affecting the reliability of maintenance decisions. Therefore, how to achieve automated fault location identification of electrical equipment based on fault membership has become a challenge for the industry. Summary of the Invention

[0004] This application provides an automated detection method and system for electrical equipment, which can realize automated identification of fault location of electrical equipment based on fault membership.

[0005] In a first aspect, this application provides an automated testing method for electrical equipment, comprising the following steps: Sensor nodes are deployed at key components of electrical equipment, and electrical parameters of each sensor node are collected by intelligent sensors. These electrical parameters include voltage signals, temperature signals, and vibration signals. Harmonic analysis is performed on the voltage signals collected by each sensor node to extract characteristic harmonic components. The propagation path of voltage fluctuations is traced by the amplitude attenuation coefficient, phase difference, and line impedance parameters of the characteristic harmonic components between adjacent sensor nodes, and the voltage characteristic propagation characteristics of each sensor node are obtained. Extract the similarity deviation of temperature change rate and vibration spectrum between adjacent sensor nodes, mark each sensor node as abnormal based on the temperature change rate and the similarity deviation, and then construct an abnormal detection map of electrical equipment based on the abnormal judgment marking results and the physical connection relationship between key components. Based on the voltage characteristic propagation properties of the sensor node and the anomaly detection map, a fuzzy evaluation of the fault probability of the sensor node location is performed to obtain the fault membership degree of the sensor node location. Then, based on all fault membership degrees, the fault location of the electrical equipment is automatically located.

[0006] Preferably, harmonic analysis is performed on the voltage signals collected by each sensor node to extract characteristic harmonic components, specifically including: A fast Fourier transform is performed on the voltage signal acquired by each sensor node to obtain the frequency domain distribution of the voltage signal, which includes the frequency domain amplitude and phase distribution. Identify the higher harmonic components in the frequency domain distribution; Based on preset amplitude and phase fluctuation thresholds, high-order harmonic components that are sensitive to the operating status of electrical equipment are screened out and identified as characteristic harmonic components of the corresponding sensor nodes.

[0007] Preferably, the propagation path of voltage fluctuations is traced and analyzed using the amplitude attenuation coefficient and phase difference of characteristic harmonic components between adjacent sensor nodes and line impedance parameters. The resulting voltage characteristic propagation characteristics for each sensor node specifically include: Extract the amplitude attenuation coefficient and phase difference of characteristic harmonic components between adjacent sensor nodes; A transmission model for voltage fluctuations is established based on the amplitude attenuation coefficient, phase difference, and line impedance parameters. Based on the transmission model, the transmission process of characteristic harmonic components is inverted to determine the propagation path of voltage fluctuations in electrical equipment, and then the voltage characteristic propagation characteristics corresponding to each sensor node are extracted.

[0008] Preferably, extracting the similarity deviation between the temperature change rate and vibration spectrum of adjacent sensor nodes specifically includes: Time series difference operation is performed on the temperature signals collected by each sensor node to obtain the temperature change rate, which reflects the dynamic characteristics of temperature. The vibration signals collected by each sensor node are subjected to spectral decomposition to obtain the vibration spectrum containing amplitude and frequency distribution; The correlation coefficient of the vibration spectrum between adjacent sensor nodes is calculated, and then the consistency of the vibration response of the two sensor nodes is quantified based on the correlation coefficient to obtain the similarity deviation of the vibration spectrum.

[0009] Preferably, the anomaly detection and marking of each sensor node based on the temperature change rate and the similarity deviation specifically includes: The temperature change rate is compared with a preset temperature rise threshold to determine temperature anomalies. The similarity deviation of the vibration spectrum is compared with a preset similarity threshold to determine vibration anomalies; The abnormality determination results based on the temperature abnormality and the vibration abnormality are used to assign abnormality determination marks to each sensor node.

[0010] Preferably, constructing the abnormality detection map of the electrical equipment according to the abnormality determination mark results and the physical connection relationship between the key components specifically includes: determining node attributes according to the abnormality determination marks of each sensor node; establishing a graph structure according to the physical connection relationship between the key components; generating an abnormality detection map representing abnormality distribution of the electrical equipment by combining the node attributes and the graph structure.

[0011] Preferably, the fault membership degree of the sensor node position is obtained by fuzzy evaluation of the fault probability of the sensor node position according to the voltage feature propagation characteristics corresponding to the sensor node and the abnormality detection map specifically includes: constructing a joint evaluation matrix of the voltage feature propagation characteristics and the node abnormality mark information in the abnormality detection map, the joint evaluation matrix being used to quantify the fault possibility of the sensor node in different feature dimensions; mapping each index in the joint evaluation matrix to a fuzzy set by a fuzzy membership function to obtain a fuzzy evaluation vector of each sensor node; based on the fuzzy evaluation vector, comprehensively calculating the fault probability of each sensor node according to fuzzy reasoning rules to obtain the fault membership degree of the sensor node position.

[0012] In a second aspect, the present application provides an automatic detection system of an electrical equipment, comprising: a collection module, configured to arrange sensor nodes at key components of the electrical equipment, and collect electrical parameters of each sensor node by intelligent sensors, the electrical parameters including voltage signals, temperature signals and vibration signals; a processing module, configured to perform harmonic analysis on the voltage signals collected by each sensor node, extract feature harmonic components, trace the propagation path of voltage fluctuation by the amplitude attenuation coefficient and the phase difference of the feature harmonic components between adjacent sensor nodes and the line impedance parameter, and obtain the voltage feature propagation characteristics corresponding to each sensor node; the processing module is further configured to extract the similarity deviation of the temperature change rate and the vibration spectrum between adjacent sensor nodes, assign abnormality determination marks to each sensor node based on the temperature change rate and the similarity deviation, and further construct an abnormality detection map of the electrical equipment according to the abnormality determination mark results and the physical connection relationship between the key components; The execution module is used for fuzzy evaluation of the fault probability of the sensor node position according to the voltage feature propagation characteristic corresponding to the sensor node and the abnormality detection graph, to obtain the fault membership of the sensor node position, and then automatically locate the fault position of the electrical equipment based on all the fault memberships.

[0013] In a third aspect, the present application provides a computer device, comprising a memory and a processor, the memory stores a code, and the processor is configured to acquire the code and execute the automatic detection method of the electrical equipment.

[0014] In a fourth aspect, the present application provides a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the automatic detection method of the electrical equipment.

[0015] The technical scheme provided by the embodiments of the present application has the following beneficial effects: In the embodiments of the present application, first, sensor nodes are arranged at key components of the electrical equipment, and electrical parameters of each sensor node are collected by intelligent sensors, including voltage signals, temperature signals and vibration signals; the voltage signals collected by each sensor node are subjected to harmonic analysis to extract characteristic harmonic components, the propagation path of voltage fluctuation is traced and analyzed by the amplitude attenuation coefficient and phase difference of the characteristic harmonic components between adjacent sensor nodes and the line impedance parameter, to obtain the voltage feature propagation characteristic corresponding to each sensor node; the similarity deviation of the temperature change rate and the vibration spectrum between adjacent sensor nodes is extracted, and each sensor node is marked for abnormality determination based on the temperature change rate and the similarity deviation, and then an abnormality detection graph of the electrical equipment is constructed according to the abnormality determination marking result and the physical connection relationship between the key components; the fault probability of the sensor node position is evaluated based on the voltage feature propagation characteristic corresponding to the sensor node and the abnormality detection graph, to obtain the fault membership of the sensor node position, and then the fault position of the electrical equipment is automatically located based on all the fault memberships.

[0016] It can be seen that, by means of the fuzzy evaluation of the fault probability of the sensor node position based on the voltage feature propagation characteristics corresponding to the sensor node and the abnormality detection graph, the fault membership of the sensor node position is obtained, and then the fault position of the electrical equipment is automatically located based on all the fault memberships. Firstly, by arranging sensor nodes on key components of the electrical equipment and collecting multi-dimensional electrical parameters such as voltage, temperature and vibration, the all-around perception of the equipment running state is realized, thereby improving the monitoring accuracy and real-time performance. In the voltage signal processing link, by extracting the characteristic harmonic component and combining the amplitude attenuation coefficient, phase difference and line impedance parameter between adjacent nodes for propagation path tracing analysis, not only the transmission characteristics of the voltage fluctuation can be quantified, but also the propagation law of the fault signal in the equipment can be revealed, thereby providing a reliable basis for accurate positioning. Secondly, by calculating the temperature change rate and vibration frequency spectrum similarity deviation and combining the preset threshold for abnormality judgment, sensitive identification of local abnormalities is realized, the false alarm risk is effectively reduced, and a quantifiable node abnormality mark is formed to provide key input for subsequent topology analysis. Then, a structured graph is established by using the physical connection relationship between the node attributes and the key components, the multi-dimensional node features are fused with the equipment topology, and a data structure that can intuitively reflect the abnormal distribution and correlation is formed, thereby enhancing the visualization and operability of the fault propagation path analysis. Finally, based on the joint analysis of the graph and the voltage feature propagation characteristics, the fault probability of each sensor node is comprehensively evaluated by using the fuzzy membership function and the fuzzy reasoning rule, the node-level fault membership is generated, the quantitative representation of the fault probability in the complex system is realized, the fault position of the electrical equipment is automatically identified by comprehensively considering the fault membership of all nodes, and the positioning accuracy, response speed and detection robustness are improved. The scheme of the present application can realize the automatic identification of the fault position of the electrical equipment based on the fault membership, thereby improving the robustness of the automatic detection of the electrical equipment in complex environments. BRIEF DESCRIPTION OF DRAWINGS

[0017] Figure 1 is an exemplary flowchart of an automatic detection method of an electrical equipment according to some embodiments of the present application; Figure 2 is a schematic diagram of an application scenario of an automatic detection system according to some embodiments of the present application; Figure 3 is a flowchart of generating an abnormality detection graph according to some embodiments of the present application; Figure 4 is a structural schematic diagram of an automatic detection system of an electrical equipment according to some embodiments of the present application; Figure 5 is a structural schematic diagram of a computer device for implementing an automatic detection method of an electrical equipment according to some embodiments of the present application. Detailed Implementation

[0018] To better understand the technical solution of this application, the technical solution of this application will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0019] refer to Figure 1 The figure is an exemplary flowchart of an automated testing method for electrical equipment according to some embodiments of this application. The automated testing method for electrical equipment mainly includes the following steps: In step 101, sensor nodes are arranged at key components of the electrical equipment, and electrical parameters of each sensor node are collected by smart sensors. The electrical parameters include voltage signals, temperature signals and vibration signals.

[0020] It should be noted that deploying sensor nodes at critical components of electrical equipment refers to installing sensors in parts of the equipment that have a significant impact on the operating status or are prone to failure, in order to collect key electrical parameters such as voltage, temperature, and vibration in real time. Specifically, firstly, critical components are identified based on the structure and operating characteristics of the electrical equipment, such as transformer windings, switch contacts, or busbar connection nodes. These components are sensitive to the overall operating status of the equipment and are prone to anomalies. Then, intelligent sensor nodes are installed at these critical components. These sensors can collect electrical parameters such as voltage, temperature, and vibration in real time, and perform preliminary filtering and amplification through an embedded signal processing unit to ensure the accuracy and stability of the data. Finally, each sensor node transmits the collected data to a centralized monitoring system via wired or wireless communication interfaces, enabling continuous monitoring of the operating status of critical components and providing reliable raw data input for subsequent harmonic analysis, temperature change rate calculation, and vibration spectrum comparison.

[0021] In some embodiments, reference Figure 2 As shown in the figure, this figure is a schematic diagram of the application scenario of the automated detection system shown in some embodiments of the present invention. It includes three main components: acquisition device, server and storage device. The acquisition device is used to acquire electrical parameters and send the acquired electrical parameters to the server through a communication network. The execution code of the automated detection system is run in the server. Finally, the execution result is stored in the storage device through the server for further visualization.

[0022] In step 102, harmonic analysis is performed on the voltage signals collected by each sensor node to extract characteristic harmonic components. The propagation path of voltage fluctuations is traced by the amplitude attenuation coefficient, phase difference, and line impedance parameters of the characteristic harmonic components between adjacent sensor nodes to obtain the voltage characteristic propagation characteristics corresponding to each sensor node.

[0023] In some embodiments, the voltage signal collected by each sensor node is subjected to harmonic analysis, and the extraction of the characteristic harmonic component can be achieved by the following steps: The voltage signal collected by each sensor node is subjected to fast Fourier transform to obtain a frequency domain distribution of the voltage signal, which includes the frequency domain amplitude and phase distribution. In the frequency domain distribution, identify each order of the high harmonic component. According to the preset amplitude threshold and phase fluctuation threshold, filter out the high harmonic component sensitive to the operation state of the electrical equipment, and determine it as the characteristic harmonic component of the corresponding sensor node.

[0024] It should be noted that the high harmonic component in the present application can reflect the local abnormal state of the key components of the electrical equipment and the voltage fluctuation propagation characteristics; the characteristic harmonic component in the present application refers to the high harmonic component which can effectively reflect the operation state of the electrical equipment and the abnormal information of the key components.

[0025] In a specific implementation, first, the voltage signal collected by each sensor node is subjected to fast Fourier transform to obtain the frequency domain distribution of the voltage signal, which can be achieved in the following manner: for the voltage signal collected by each sensor node, frequency domain analysis is performed by using fast Fourier transform to convert the time domain voltage signal into amplitude spectrum and phase spectrum, the amplitude spectrum is used to reflect the energy size of different frequency components, and the phase spectrum is used to represent the phase information of each frequency component. It needs to be further explained that, in order to ensure accurate acquisition of high-order harmonic components, the sampling rate of the voltage signal meets the requirement of the Nyquist sampling theorem and can completely cover the fundamental wave and the required high-order harmonic frequency range; second, the identification of each order of high-order harmonic component in the frequency domain distribution can be achieved in the following manner: in the obtained frequency domain distribution, the fundamental wave frequency is taken as a reference to identify each order of high-order harmonic component, and the specific method includes that in the frequency domain amplitude spectrum, the amplitude peak value at the integer multiple position of the fundamental wave is automatically detected by using a peak value detection algorithm or a sliding window local extremum search, and the corresponding phase information is recorded. By comparing the spectrum amplitude and the noise level, the interference component can be removed to ensure the accuracy of the identification of high-order harmonics; then, the high-order harmonic component sensitive to the running state of the electrical equipment is screened out according to the preset amplitude threshold and phase fluctuation threshold, and the high-order harmonic component is determined as the characteristic harmonic component of the corresponding sensor node, which can be achieved in the following manner: the identified high-order harmonic component is screened according to the preset amplitude threshold and phase fluctuation threshold, the amplitude threshold is used to exclude low-energy harmonics that have no obvious indication effect on the running state of the equipment, and the phase fluctuation threshold is used to determine the harmonic propagation consistency abnormality or local abnormality. Only the high-order harmonic component with an amplitude exceeding the amplitude threshold and a phase deviation exceeding the phase fluctuation threshold is retained as the characteristic harmonic component of the sensor node; it needs to be further explained that the setting of the amplitude threshold can be determined according to the statistical characteristics of each high-order harmonic component when the electrical equipment is normally running. Usually, the mean value and standard deviation of the amplitude of each order of harmonic are calculated by performing frequency spectrum analysis on the historical running data, and the threshold is set by combining engineering experience or safety margin, for example, the mean value plus twice the standard deviation is taken as the determination limit; the setting of the phase fluctuation threshold is based on the phase fluctuation range of the high-order harmonic under the normal running state, the maximum phase deviation is calculated by time series analysis on the historical phase data, and the threshold is determined by combining the characteristics of the equipment and the allowable error, so as to ensure that the screened high-order harmonic can reflect the abnormal characteristics of the key components and avoid misjudgment caused by normal running fluctuation.

[0026] In some embodiments, the propagation path of the voltage fluctuation is traced by using the amplitude attenuation coefficient and the phase difference of the characteristic harmonic component between adjacent sensor nodes and the line impedance parameter to obtain the voltage characteristic propagation characteristics corresponding to each sensor node, which can be achieved in the following steps: The amplitude attenuation coefficient and the phase difference of the characteristic harmonic component between adjacent sensor nodes are extracted; a transmission model of the voltage fluctuation is established according to the amplitude attenuation coefficient, the phase difference and the line impedance parameter; Based on the transmission model, an inversion calculation is performed on a transmission process of the characteristic harmonic component, a propagation path of the voltage fluctuation in the electrical equipment is determined, and then voltage characteristic propagation characteristics corresponding to each sensor node are extracted.

[0027] It should be noted that the amplitude attenuation coefficient in the present application is an index for measuring the amplitude loss degree of the characteristic harmonic during the propagation process between adjacent sensor nodes; the phase difference in the present application refers to the phase shift amount of the characteristic harmonic during the propagation between adjacent sensor nodes; and the voltage characteristic propagation characteristics in the present application refer to the comprehensive characteristics for quantitatively describing the propagation path, amplitude change and phase shift of the characteristic harmonic in the electrical equipment.

[0028] In a specific implementation, first, the amplitude attenuation coefficient and the phase difference of the characteristic harmonic component between adjacent sensor nodes can be obtained in the following manner: in the process of calculating the amplitude attenuation coefficient of the characteristic harmonic component, the frequency domain amplitudes of the same order of high harmonic at two nodes are first normalized, and then the ratio thereof is taken as the amplitude attenuation coefficient; in the process of calculating the phase difference, the phase spectrum corresponding to the frequency of the two nodes is unwrapped to eliminate the 2π ambiguity effect, and then the phase offset between the nodes is calculated and taken as the phase difference; then, the transmission model of voltage fluctuation based on the amplitude attenuation coefficient, the phase difference and the line impedance parameter can be established in the following manner: according to circuit theory, the line parameters between nodes are associated with the harmonic propagation characteristics, and the lumped parameter model is used to describe the short distance line or the distributed parameter model is used to describe the long distance line; under the lumped parameter model of the line, the line resistance, inductance and capacitance parameters are used to derive the amplitude transmission relationship and the phase change between nodes through the circuit voltage division principle; under the distributed parameter model, the propagation constant is introduced, the amplitude loss is described by exponential decay, and the phase offset in the propagation process is described by the phase term; the model parameters can be calibrated by least square fitting combined with historical normal operating condition data to ensure that the model prediction error is within an acceptable range; finally, the transmission process of the characteristic harmonic component is calculated based on the transmission model, the propagation path of the voltage fluctuation in the electrical equipment is determined, and then the voltage characteristic propagation characteristics of each sensor node can be obtained in the following manner: based on the transmission model, it is assumed that the harmonic source is emitted from a certain node, the theoretical amplitude and phase of each node are calculated, and the theoretical value and the actual measured value are compared in terms of mean square error; the harmonic source position is determined by searching for the node with the minimum error through all nodes or using the particle swarm optimization algorithm; the propagation path of the voltage fluctuation is determined according to the direction from the source node to other nodes, combined with the amplitude attenuation coefficient and the phase difference, and the voltage characteristic propagation characteristics of each node are extracted, including the curve of the characteristic harmonic amplitude changing with the propagation distance, the relationship between the phase difference and the frequency, and the node impedance matching degree, thereby forming a quantitative propagation characteristic data set.

[0029] In step 103, the temperature change rate and the similarity deviation of the vibration frequency spectrum between adjacent sensor nodes are extracted, each sensor node is marked based on the temperature change rate and the similarity deviation, and then the abnormal detection map of the electrical equipment is constructed according to the abnormal judgment mark result and the physical connection relationship between the key components.

[0030] In some embodiments, the temperature change rate and the similarity deviation of the vibration frequency spectrum between adjacent sensor nodes can be obtained in the following steps: The temperature signals collected by each sensor node are subjected to time series difference operation to obtain the temperature change rate reflecting the temperature dynamic characteristics; The vibration signals collected by each sensor node are subjected to frequency spectrum decomposition to obtain vibration frequency spectrum containing amplitude and frequency distribution; The correlation coefficient of the vibration frequency spectrum between adjacent sensor nodes is calculated, and the consistency of the vibration responses of the two sensor nodes is quantified based on the correlation coefficient to obtain the similarity deviation of the vibration frequency spectrum.

[0031] It should be noted that the temperature change rate in the present application refers to the change amplitude of the temperature signal per unit time, which is used to represent the dynamic change characteristics of the temperature with time; the similarity deviation of the vibration frequency spectrum in the present application refers to the difference degree of the vibration frequency spectrum of adjacent sensor nodes in frequency and amplitude distribution, which is used to measure the consistency of the vibration response; the vibration frequency spectrum in the present application refers to the amplitude distribution characteristics of the vibration signal at different frequencies obtained by frequency domain analysis.

[0032] It should also be noted that before extracting the temperature change rate and the similarity deviation of the vibration frequency spectrum between adjacent sensor nodes, the temperature signal and the vibration signal collected by each sensor node are subjected to pretreatment, specifically, the temperature signal is subjected to sampling point alignment based on time stamp linear interpolation, and a low-pass filter is used to filter out high-frequency noise, and before the vibration signal enters the frequency domain analysis, a window function is applied to suppress spectrum leakage, and direct current removal and normalization processing are performed.

[0033] In a specific implementation, firstly, time series difference operation is performed on the temperature signals collected by each sensor node to obtain the temperature change rate reflecting the temperature dynamic characteristics. The temperature change rate can be obtained in the following manner: the temperature signals are segmented into preset sliding windows, and the trend of temperature change over time is obtained on each window by using a linear least square fitting method. The slope obtained by fitting is the temperature change rate in the window, thereby reflecting the temperature dynamic characteristics of the sensor node in the time period. Secondly, frequency spectrum decomposition is performed on the vibration signals collected by each sensor node to obtain the vibration frequency spectrum containing the amplitude and frequency distribution. The vibration frequency spectrum can be obtained in the following manner: the Welch method is used to perform power spectrum density estimation on the vibration signals to obtain the vibration frequency spectrum containing the amplitude and frequency distribution. In order to ensure the stability of the spectrum estimation, a fixed segment length and overlap rate are set, and each segment of data is weighted by using a Hanning window, thereby obtaining the power spectrum vector of each sensor node. Then, the correlation coefficient of the vibration frequency spectrum between adjacent sensor nodes is calculated, and the consistency of the vibration responses of two sensor nodes is quantified based on the correlation coefficient, thereby obtaining the similarity deviation of the vibration frequency spectrum. The similarity deviation of the vibration frequency spectrum can be obtained in the following manner: after the power spectrum vectors of each node are normalized, the Pearson correlation coefficient or the cosine similarity between the two nodes is calculated to represent the consistency of the overall vibration response. Meanwhile, the cross spectrum and the auto spectrum between the two nodes are further calculated to obtain the amplitude squared coherence function, and the coherence is weighted and averaged in the target frequency band to obtain an index reflecting the frequency resolution consistency. The correlation coefficient and the average coherence are weighted according to a preset weight to obtain a comprehensive similarity index. Further, the average similarity under the historical normal working condition is taken as a reference, and the comprehensive similarity calculated in real time is compared with the reference value to obtain the real-time similarity deviation value. In order to weaken the influence of instantaneous noise on the calculation result, the deviation value can be smoothed, and a threshold is determined in combination with the historical statistical standard deviation, thereby ensuring the stability and reliability of the extracted similarity deviation.

[0034] In some embodiments, the abnormality determination mark for each sensor node based on the temperature change rate and the similarity deviation can be realized in the following steps: Comparing the temperature change rate with a preset temperature rise threshold to determine temperature abnormality; Comparing the similarity deviation of the vibration frequency spectrum with a preset similarity threshold to determine vibration abnormality; Assigning an abnormality determination mark to each sensor node based on the determination results of temperature abnormality and vibration abnormality.

[0035] In a specific implementation, firstly, the temperature change rate is compared with a preset temperature rise threshold, and the temperature abnormality is determined by the following method: the temperature change rate is compared with a preset temperature rise threshold, the temperature rise threshold is determined by statistical fitting of the temperature change rate sample in the normal operation state of the equipment, and when the temperature change rate exceeds the temperature rise threshold, it is determined that the temperature is abnormal; secondly, the similarity deviation of the vibration spectrum is compared with a preset similarity threshold, and the vibration abnormality is determined by the following method: the similarity deviation can be compared with a preset similarity threshold, the similarity threshold is determined by collecting vibration data samples in the normal working condition of the equipment and based on the statistical distribution law, and when the similarity deviation exceeds the similarity threshold, it is determined that the vibration is abnormal; then, the abnormality determination mark is given to each sensor node based on the determination results of the temperature abnormality and the vibration abnormality, which can be realized by the following method: a binary combination coding method can be used, wherein 00 is marked when the temperature and vibration are normal, 01 is marked when the temperature is abnormal and the vibration is normal, 10 is marked when the temperature is normal and the vibration is abnormal, and 11 is marked when the temperature and vibration are abnormal, and the binary mark can be further mapped to the corresponding abnormality level, which is used to indicate the running state of each sensor node and provide a basis for subsequent fault positioning and state evaluation.

[0036] In some embodiments, in some embodiments, reference Figure 3 As shown in the figure, the figure is a flowchart for generating an abnormality detection graph in some embodiments of the present application, and the abnormality detection graph of the electrical equipment can be constructed by the following steps in the present embodiment according to the abnormality determination mark results and the physical connection relationship between the key components: In step 1031, the node attribute is determined according to the abnormality determination mark of each sensor node; In step 1032, the graph structure is established according to the physical connection relationship between the key components; In step 1033, the abnormality detection graph representing the abnormality distribution of the electrical equipment is generated by combining the node attribute and the graph structure.

[0037] It should be noted that the node attribute in the present application is used to represent the running state and key feature information of the sensor node; the abnormality detection graph in the present application is a structured graph used to represent the abnormal state of each key component of the electrical equipment and the mutual correlation.

[0038] In a specific implementation, first, the node attribute can be implemented in the following manner according to the abnormality determination marks of each sensor node: arranging sensor nodes at key components of the electrical equipment, processing the temperature signals and vibration signals collected by each node, and quantifying the state of each sensor node according to the abnormality determination marks of the temperature change rate and vibration spectrum, which includes encoding the temperature abnormality and vibration abnormality and adding the original measurement parameters such as the temperature change rate and vibration similarity deviation to form a complete node attribute set, the node attribute set including the node unique identification, installation location, sensor type, calibration information and abnormality determination result, which can be stored in a database for retrieval and management; then, the graph structure can be implemented in the following manner according to the physical connection relationship between the key components: converting the key components and the connection relationship therebetween into a graph structure based on the physical connection topology of the electrical equipment, the graph structure including nodes and edges, the nodes corresponding to the key components or the sensor installation locations, the node attributes used to describe the component types and abnormal states, the edges corresponding to the physical connections between the components, and the edge attributes used to describe the connection types, line impedance, cable length and wire cross-sectional area and other electrical characteristics, the graph structure integrity is verified through an adjacency matrix to ensure that the connection elements between the nodes in the matrix correctly reflect the actual physical connection relationship between the components, and the matrix elements of 1 represent two components directly connected and 0 represent no direct connection, thereby ensuring that the graph structure is consistent with the actual topology of the electrical equipment; finally, the abnormality detection graph representing the abnormality distribution of the electrical equipment can be generated in the following manner by combining the node attributes and the graph structure: mapping the node attributes to visual elements such as colors and sizes through a visual method to represent the abnormality level and severity, and the color and thickness of the edges represent the connection characteristics and importance, and the graph can be provided with interactive functions to realize the suspended display of the attribute information of the nodes, the display of the connection parameters by clicking the edges, and the labeling of the current direction, so as to quickly locate the abnormality concentration area, and finally, the generated graph can be output as a static graph or a dynamic graph and stored in the operation and maintenance management system in association with the equipment asset number, realizing data traceability and operability.

[0039] In step 104, the fault probability of the sensor node position is evaluated according to the voltage characteristic propagation characteristics corresponding to the sensor node and the abnormality detection graph, and the fault membership of the sensor node position is obtained, and then the fault position of the electrical equipment is automatically located based on all the fault memberships.

[0040] In some embodiments, the fault probability of the sensor node position can be evaluated according to the voltage characteristic propagation characteristics corresponding to the sensor node and the abnormality detection graph, and the fault membership of the sensor node position can be implemented in the following steps: construct a joint evaluation matrix of the voltage feature propagation characteristics and the node abnormality marked information in the anomaly detection graph, the joint evaluation matrix being used to quantify the failure possibility of the sensor node in different feature dimensions; map each index in the joint evaluation matrix to a fuzzy set through a fuzzy membership function, to obtain a fuzzy evaluation vector of each sensor node; based on the fuzzy evaluation vector, comprehensively calculate the failure probability of each sensor node according to a fuzzy inference rule, to obtain a failure membership degree of the position of each sensor node.

[0041] It should be noted that the joint evaluation matrix in the present application is a quantitative matrix reflecting the failure possibility of the sensor node in each feature dimension; the fuzzy membership function in the present application is a mathematical function mapping each feature index of the sensor node to a failure membership degree; and the failure membership degree in the present application is a technical index measuring the size of the failure possibility of the sensor node.

[0042] In specific implementation, the joint evaluation matrix of voltage characteristic propagation characteristics and node anomaly marking information in the anomaly detection map can be constructed as follows: First, obtain the voltage characteristic propagation characteristics of each sensor node, including amplitude attenuation coefficient, phase difference, propagation path information, etc., and simultaneously obtain the anomaly judgment information of the corresponding node in the anomaly detection map, including temperature anomaly, vibration anomaly, and anomaly level. For these multi-dimensional characteristics, the values ​​are standardized to map indicators of different dimensions to the same range to ensure the comparability and stability of subsequent fuzzy calculations. Standardization can be achieved using a linear normalization method, mapping the amplitude attenuation coefficient, phase difference, and propagation path information to the same range. Position difference and anomaly level are mapped to the interval between 0 and 1. Further, the numerical indicators of each node under each feature dimension are mapped to matrix elements. Matrix rows represent sensor nodes, and matrix columns represent various feature indicators, such as amplitude attenuation coefficient, phase difference, temperature anomaly level, vibration anomaly level, etc. Each element represents the fault sensitivity or fault probability of the node under that feature dimension. The resulting matrix is ​​used as the joint evaluation matrix. This joint evaluation matrix can be used to quantify the fault probability of the corresponding component of the sensor node under different feature dimensions. Then, fuzzy membership functions are used to map each indicator in the joint evaluation matrix to a fuzzy set, obtaining each... The fuzzy evaluation vector of a sensor node can be implemented as follows: For each index in the joint evaluation matrix, a fuzzy membership function is applied for mapping to form the node's fuzzy evaluation vector. The membership functions for amplitude attenuation coefficient and phase difference can use triangular or trapezoidal functions, mapping low attenuation coefficient / small phase difference to low failure probability, and high attenuation coefficient / large phase difference to high failure probability. The discrete levels of temperature anomalies and vibration anomalies are directly mapped to membership values, forming continuous quantities in the range of 0 to 1. In this way, the failure probability of each node under each feature dimension is quantified into a membership degree. Then, the feature indices under each feature dimension are further... The labels are arranged to obtain a vector, which is the fuzzy evaluation vector of the sensor node. The length of the vector is equal to the number of feature indicators, which is used to comprehensively describe the fault sensitivity of the node under different feature dimensions. Finally, based on the fuzzy evaluation vector, the fault probability of each sensor node is comprehensively calculated according to the fuzzy inference rules. The fault membership degree of each sensor node position can be obtained in the following way: based on the fuzzy evaluation vector of each node, the fuzzy inference rules are used to perform logical comprehensive calculation of each indicator. The inference rules can adopt the well-known "if-then" rule. For example, if the amplitude attenuation coefficient is high and the temperature is abnormal, the node has a high probability of failure.If the amplitude attenuation coefficient is low and the vibration is abnormal, the node failure possibility is medium. For multi-index input, fuzzy logic operators can be used for aggregation, such as fuzzy and operation to take the minimum value, fuzzy or operation to take the maximum value, or weighted average method is used to weight and fuse the membership degrees of each index. The weight can be determined according to historical failure data or expert experience to ensure the accuracy of the comprehensive evaluation. Further, the membership degree vector calculated by fuzzy reasoning is de-fuzzied and converted into a specific numerical failure probability to obtain the failure membership of each sensor node position. The de-fuzzification method can use the barycenter method, the maximum membership degree method or the weighted average method to map the membership degree to a continuous value between 0 and 1, where 0 represents a very low failure possibility and 1 represents a very high failure possibility. The failure membership can be directly used for electrical equipment fault location and early warning analysis to provide quantitative decision-making basis for operation and maintenance.

[0043] It should be noted that the scheme of the present application quantitatively processes the voltage feature propagation characteristics of the sensor node and the node abnormality marking information in the anomaly detection graph, effectively solving the problem of lack of multi-dimensional feature comprehensive analysis and uncertainty processing capability in the prior art for electrical equipment fault location. The prior art usually only relies on a single sensor signal or directly judges the abnormal threshold, which is difficult to accurately reflect the complex voltage fluctuation propagation law between nodes and the relevance of multiple types of abnormal signals, leading to false negative or false positive in fault judgment. The present scheme constructs a joint evaluation matrix to quantize multi-dimensional features in the same evaluation framework, and uses a fuzzy membership function to map uncertainty and volatility to a fuzzy set, realizes fuzzy reasoning and comprehensive calculation of failure probability, and obtains the failure membership of each sensor node position. The implementation effect of the technical scheme includes improving the accuracy and robustness of fault location, realizing systematic analysis of multi-source abnormal information in complex electrical systems, quantitatively reflecting the size of failure possibility, and providing reliable data support for subsequent automatic fault diagnosis and operation and maintenance decision-making, significantly enhancing the intelligent level and operation repeatability of electrical equipment state monitoring.

[0044] It should also be noted that the automated fault location of electrical equipment based on all fault membership degrees in this application refers to calculating and integrating the fault membership degrees of each sensor node, mapping voltage characteristic propagation and abnormal distribution information onto the equipment topology, in order to accurately identify the key components or line sections most likely to experience faults. Specifically, firstly, fault membership degree data is collected from all sensor nodes deployed in the electrical equipment, and associated with the installation location of the corresponding nodes and the collected electrical parameters to form a complete dataset. Secondly, based on the physical topology of the equipment, the fault membership degrees of each sensor node are mapped to key components and their connected lines, constructing a node-component mapping matrix. Then, the fault membership degrees of adjacent nodes or nodes with electrical connections are superimposed or averaged to obtain a comprehensive fault index for each key component or line section. Finally, the comprehensive fault index is sorted or a threshold is set for filtering, thereby automatically identifying the key components or line sections most likely to experience faults. A fault distribution map can be generated using a visual interface, enabling maintenance personnel to intuitively and quickly locate and handle abnormal areas. Through the above steps, this embodiment can realize the automatic identification of electrical equipment fault location based on fault membership, ensuring the accuracy and operability of fault location.

[0045] On the other hand, in some embodiments, this application provides an automated detection system for electrical equipment, with reference to... Figure 4 The figure is a schematic diagram of the structure of an automated detection system for electrical equipment according to some embodiments of this application. The automated detection system 400 for electrical equipment includes: a data acquisition module 401, a processing module 402, and an execution module 403, which are described below: The acquisition module 401 in this application is mainly used to arrange sensor nodes at key components of electrical equipment and acquire electrical parameters of each sensor node through intelligent sensors. The electrical parameters include voltage signals, temperature signals and vibration signals. Processing module 402, in this application, is used to perform harmonic analysis on the voltage signals collected by each sensor node, extract characteristic harmonic components, and perform source analysis on the propagation path of voltage fluctuations based on the amplitude attenuation coefficient, phase difference, and line impedance parameters of the characteristic harmonic components between adjacent sensor nodes, so as to obtain the voltage characteristic propagation characteristics corresponding to each sensor node. In this application, the processing module 402 is also used to extract the similarity deviation of the temperature change rate and vibration spectrum between adjacent sensor nodes, and to mark each sensor node as abnormal based on the temperature change rate and the similarity deviation, and then to construct an abnormal detection map of electrical equipment based on the abnormal judgment marking results and the physical connection relationship between key components. The execution module 403 is mainly used for fuzzy evaluation of the fault probability of the sensor node position according to the voltage feature propagation characteristics of the sensor node and the abnormal detection atlas, obtaining the fault membership of the sensor node position, and then automatically positioning the fault position of the electrical equipment based on all the fault memberships.

[0046] In addition, the application further provides a computer device, which comprises a memory and a processor, the memory stores code, and the processor is configured to acquire the code and execute the automatic detection method of the electrical equipment.

[0047] In some embodiments, the reference Figure 5 The figure is a structural schematic diagram of a computer device for implementing the automatic detection method of the electrical equipment according to some embodiments of the application. The automatic detection method of the electrical equipment in the above embodiments can be implemented by the computer device shown in the figure, which comprises at least one processor 501, a communication bus 502, a memory 503 and at least one communication interface 504. Figure 5

[0048] The processor 501 can be a general central processing unit (CPU) or an application specific integrated circuit (ASIC).

[0049] The communication bus 502 can be used to transmit information between the above components.

[0050] ​The memory 503 can be a readonly memory (ROM) or other type of static storage device that can store static information and instructions, a random access memory (RAM), or other type of dynamic storage device that can store information and instructions, an electrically erasable programmable readonly memory (EEPROM), a compact disc readonly memory (CDROM) or other optical disk storage, a magneto-optical disk storage, a magnetic disk or other magnetic storage device, or any other medium capable of storing desired program code in the form of instructions or data structures and that can be accessed by a computer, but is not limited to this. The memory 503 can exist independently, and is connected to the processor 501 through the communication bus 502. The memory 503 can also be integrated with the processor 501.

[0051] The memory 503 is configured to store program codes for implementing the solutions of the present application, and the processor 501 is configured to control the execution of the program codes. The processor 501 is configured to execute the program codes stored in the memory 503. The program codes can include one or more software modules. The automatic detection method of the electrical equipment in the above embodiments can be implemented by one or more software modules in the program codes in the processor 501 and the memory 503.

[0052] The communication interface 504 is configured to communicate with other devices or communication networks, such as an Ethernet, a radio access network (RAN), a wireless local area network (WLAN), etc., using any transceiver-like mechanism.

[0053] In specific implementations, as an example, the computer device can include multiple processors, each of which can be a single CPU processor or a multi-CPU processor. The processor herein can refer to one or more devices, circuits, and / or processing cores for processing data (e.g., computer program instructions).

[0054] The computer device described above can be a general-purpose computer device or a special-purpose computer device. In a specific implementation, the computer device can be a desktop computer, a laptop computer, a network server, a personal digital assistant (PDA), a mobile phone, a tablet computer, a wireless terminal device, a communication device, or an embedded device. The embodiments of the present application do not limit the type of the computer device.

[0055] In addition, the present application also provides a computer readable storage medium, which stores a computer program. The computer program is executed by a processor to implement the automatic detection method of the electrical device.

[0056] Although the preferred embodiments of the present application have been described, those skilled in the art who are informed of the basic inventive concept can make further changes and modifications to the embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present application.

[0057] Obviously, those skilled in the art can make various modifications and variations to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalent technologies, the present application also intends to include these modifications and variations.

Claims

1. An automated testing method for electrical equipment, characterized in that, Includes the following steps: Sensor nodes are deployed at key components of electrical equipment, and electrical parameters of each sensor node are collected by intelligent sensors. These electrical parameters include voltage signals, temperature signals, and vibration signals. Harmonic analysis is performed on the voltage signals collected by each sensor node to extract characteristic harmonic components. The propagation path of voltage fluctuations is traced by the amplitude attenuation coefficient, phase difference, and line impedance parameters of the characteristic harmonic components between adjacent sensor nodes, and the voltage characteristic propagation characteristics of each sensor node are obtained. Extract the similarity deviation of temperature change rate and vibration spectrum between adjacent sensor nodes, mark each sensor node as abnormal based on the temperature change rate and the similarity deviation, and then construct an abnormal detection map of electrical equipment based on the abnormal judgment marking results and the physical connection relationship between key components. Based on the voltage characteristic propagation properties of the sensor node and the anomaly detection map, a fuzzy evaluation of the fault probability of the sensor node location is performed to obtain the fault membership degree of the sensor node location. Then, based on all fault membership degrees, the fault location of the electrical equipment is automatically located.

2. The method as described in claim 1, characterized in that, Harmonic analysis is performed on the voltage signals collected by each sensor node to extract characteristic harmonic components, specifically including: A fast Fourier transform is performed on the voltage signal acquired by each sensor node to obtain the frequency domain distribution of the voltage signal, which includes the frequency domain amplitude and phase distribution. Identify the higher harmonic components in the frequency domain distribution; Based on preset amplitude and phase fluctuation thresholds, high-order harmonic components that are sensitive to the operating status of electrical equipment are screened out and identified as characteristic harmonic components of the corresponding sensor nodes.

3. The method as described in claim 1, characterized in that, The propagation path of voltage fluctuations is traced and analyzed using the amplitude attenuation coefficient and phase difference of characteristic harmonic components between adjacent sensor nodes and line impedance parameters. The specific voltage propagation characteristics corresponding to each sensor node include: Extract the amplitude attenuation coefficient and phase difference of characteristic harmonic components between adjacent sensor nodes; A transmission model for voltage fluctuations is established based on the amplitude attenuation coefficient, phase difference, and line impedance parameters. Based on the transmission model, the transmission process of characteristic harmonic components is inverted to determine the propagation path of voltage fluctuations in electrical equipment, and then the voltage characteristic propagation characteristics corresponding to each sensor node are extracted.

4. The method as described in claim 1, characterized in that, Extracting the similarity deviation between the temperature change rate and vibration spectrum of adjacent sensor nodes specifically includes: Time series difference operation is performed on the temperature signals collected by each sensor node to obtain the temperature change rate, which reflects the dynamic characteristics of temperature. The vibration signals collected by each sensor node are subjected to spectral decomposition to obtain the vibration spectrum containing amplitude and frequency distribution; The correlation coefficient of the vibration spectrum between adjacent sensor nodes is calculated, and then the consistency of the vibration response of the two sensor nodes is quantified based on the correlation coefficient to obtain the similarity deviation of the vibration spectrum.

5. The method as described in claim 1, characterized in that, The anomaly detection and marking of each sensor node based on the temperature change rate and the similarity deviation specifically includes: The temperature change rate is compared with a preset temperature rise threshold to determine temperature anomalies. The similarity deviation of the vibration spectrum is compared with a preset similarity threshold to determine vibration anomalies; Based on the results of temperature and vibration anomalies, anomaly detection labels are assigned to each sensor node.

6. The method as described in claim 1, characterized in that, The anomaly detection map of electrical equipment is constructed based on the anomaly identification and marking results and the physical connection relationships between key components. Specifically, this includes: The node attributes are determined based on the anomaly detection markers of each sensor node; Establish a graph structure based on the physical connection relationships between key components; An anomaly detection map representing the abnormal distribution of electrical equipment is generated by combining node attributes and graph structure.

7. The method as described in claim 1, characterized in that, Based on the voltage characteristic propagation properties of the sensor node and the anomaly detection spectrum, a fuzzy evaluation is performed on the fault probability of the sensor node location to obtain the fault membership degree of the sensor node location, specifically including: A joint evaluation matrix is ​​constructed based on voltage feature propagation characteristics and node anomaly marking information in anomaly detection maps. The joint evaluation matrix is ​​used to quantify the failure probability of sensor nodes under different feature dimensions. By mapping each index in the joint evaluation matrix to a fuzzy set through a fuzzy membership function, the fuzzy evaluation vector of each sensor node is obtained. Based on the fuzzy evaluation vector, the fault probability of each sensor node is comprehensively calculated according to the fuzzy inference rules to obtain the fault membership degree of each sensor node location.

8. An automated testing system for electrical equipment, characterized in that, include: The data acquisition module is used to deploy sensor nodes at key components of electrical equipment and acquire electrical parameters of each sensor node through intelligent sensors. The electrical parameters include voltage signals, temperature signals and vibration signals. The processing module is used to perform harmonic analysis on the voltage signals collected by each sensor node, extract characteristic harmonic components, and perform source analysis on the propagation path of voltage fluctuations based on the amplitude attenuation coefficient, phase difference, and line impedance parameters of the characteristic harmonic components between adjacent sensor nodes, so as to obtain the voltage characteristic propagation characteristics corresponding to each sensor node. The processing module is also used to extract the similarity deviation of the temperature change rate and vibration spectrum between adjacent sensor nodes, mark each sensor node as abnormal based on the temperature change rate and the similarity deviation, and then construct an abnormal detection map of electrical equipment based on the abnormality marking results and the physical connection relationship between key components. The execution module is used to perform a fuzzy evaluation of the fault probability of the sensor node location based on the voltage characteristic propagation characteristics of the sensor node and the anomaly detection spectrum, obtain the fault membership degree of the sensor node location, and then automatically locate the fault location of the electrical equipment based on all fault membership degrees.

9. A computer device comprising a memory and a processor, the memory storing code, characterized in that, The processor is configured to acquire the code and execute the automated detection method for electrical equipment as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the automated detection method for electrical equipment as described in any one of claims 1 to 7.

Images