Method and system for detecting fault locations in intelligent GIS equipment based on cloud-edge collaboration

Abstract

This invention discloses a method for detecting fault locations of intelligent GIS equipment based on cloud-edge collaboration, which relates to the field of computer technology. By utilizing cloud-edge collaboration and combining a digital twin model with anomaly change pattern recognition techniques, this invention achieves comprehensive equipment monitoring, fault prediction, and accurate location. Furthermore, the application of a digital twin model to simulate the interaction relationships between GIS equipment modules is a first in the field of GIS equipment fault location, providing new ideas and methods for solving complex equipment fault problems. Furthermore, a cloud-based large-scale interaction detection model built using a graph neural network algorithm can monitor and analyze GIS equipment interaction data in real time, improving the precision and recall of interaction anomaly detection. Furthermore, by using intelligent algorithms and automatic repair techniques, equipment faults can be automatically repaired after fault location is identified, thereby improving the reliability and stability of the equipment.

Description

[Technical Field]

[0001] The present invention relates to the field of computer technology, and in particular to a method and system for detecting fault locations in intelligent GIS equipment based on cloud-edge collaboration. [Background technology]

[0002] Existing GIS equipment troubleshooting typically relies on manual inspection and judgment to locate and handle faults. However, this method has several limitations and drawbacks. First, due to the complex and diverse structure and functions of GIS equipment, traditional methods cannot accurately identify hidden fault signals, which can result in faults going undetected and untreated in real time. Second, traditional methods typically require a large amount of manpower and time, which can reduce troubleshooting efficiency and cause long downtimes and repair delays. Finally, existing technologies are insufficient for comprehensive equipment monitoring and fault prediction, and are unable to detect potential signs of failure in advance, making it difficult to take preventive maintenance measures. Summary of the Invention

[0003] The purpose of this section is to outline some aspects of embodiments of the present invention and to briefly describe some preferred embodiments. This section, the Abstract of the Application, and the Title of the Invention may be abbreviated or omitted to avoid obscuring the purpose, and such abbreviations or omissions shall not be used to limit the scope of the present invention.

[0004] The present invention has been proposed in view of the limitations and deficiencies in troubleshooting existing GIS equipment.

[0005] Therefore, the problem that this invention aims to solve is to overcome the limitations of traditional methods in GIS equipment fault location and processing by applying digital twin models, cloud-edge collaboration ideas, and abnormal change pattern recognition methods, and to improve the capabilities of equipment monitoring, fault prediction, and accurate location.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] According to a first aspect, an embodiment of the present invention provides a fault location detection method for intelligent GIS equipment based on cloud-edge collaboration, which includes: building a digital twin model in the cloud, including all modules of the GIS equipment and their interaction relationships; simulating changes in the interaction relationships under different fault situations to generate a large amount of fault simulation data; extracting interaction data between components; building a cloud-based large-scale interaction detection model based on a graph neural network; optimizing the model hyperparameters; and simultaneously deploying the optimized large-scale interaction detection model on a cloud platform; deploying sensors locally on the GIS equipment to collect component interaction data; and building an edge-based small-scale interaction localization model to analyze the interaction data; and when an interaction anomaly is detected by the edge-based small-scale interaction localization model, The method includes requesting a global analysis from the action detection model, allowing the cloud-based large-scale interaction detection model to comprehensively determine the fault source location and the fault cause, and feeding back the result to the edge-based small-scale interaction localization model; the edge-based small-scale interaction localization model assists in local detailed judgment based on the feedback to achieve accurate fault localization, and sends the localization result to the cloud-based large-scale interaction detection model to request a correlation analysis; the cloud-based large-scale interaction detection model provides a fault type and a root cause based on the localization result, and the edge-based small-scale interaction localization model verifies the root cause to complete the overall fault localization; and based on the fault results output from the multiple models, querying a repair knowledge base to determine an optimal solution, and issuing a control command to automate the repair process, and the system makes a secondary judgment to ensure the fault is resolved.

[0008] In a preferred embodiment of the fault location detection method for intelligent GIS equipment based on cloud-edge cooperation described in the present invention, requesting a global analysis from a cloud-based large-scale interaction detection model includes the following steps: the edge-based small-scale interaction localization model continuously detects interaction data between components; if the interaction data meets a judgment criterion, the edge-based small-scale interaction localization model determines that an interaction anomaly has occurred and obtains a snapshot of the interaction data at the moment the anomaly occurred; the edge-based small-scale interaction localization model packages the snapshot of the interaction data and requests a global analysis from the cloud-based large-scale interaction detection model via a network; the cloud-based large-scale interaction detection model receives the data, reconstructs the fault scenario in a digital twin environment, and simulates the propagation and impact of the interaction data to identify the fault source location; the cloud-based large-scale interaction detection model comprehensively determines the fault source location and conducts experiments and simulations on the fault source location to estimate possible fault causes; and the cloud-based large-scale interaction detection model feeds back and transmits the estimated fault source location and fault cause to the edge-based small-scale interaction localization model.

[0009] In a preferred embodiment of the fault location detection method for intelligent GIS equipment based on cloud-edge collaboration described in the present invention, determining that an interaction anomaly has occurred is performed when the interaction data deviation exceeds μ+3σ and the interaction response time is 2T e , a significant change occurs in the interaction mode, the interaction session is interrupted, or an abnormal change occurs in the statistical characteristics of the interaction data. Regarding the interaction data deviation abnormality, the mean value μ and standard deviation σ of the normal data are calculated to determine the interaction data deviation threshold as μ+3σ. If the real-time data exceeds the interaction data deviation threshold μ+3σ, it is determined to be an interaction abnormality. Regarding the interaction response time abnormality, the mean response time T eis calculated and determined, and the interaction response time is 2T e determining that an interaction abnormality has occurred if the difference exceeds the threshold; determining that a significant change in the interaction mode has occurred using time series analysis, and analyzing the series trend, periodicity, and seasonality characteristics to establish a normal interaction mode, and comparing the real-time interaction data with the normal interaction mode, and determining that an interaction abnormality has occurred if there is a mismatch; determining that an interaction session has been interrupted by setting up a heartbeat detection mechanism for key interaction signals, and determining that an interaction abnormality has occurred if the heartbeat times out; and determining that an abnormal change has occurred in the statistical characteristics of the interaction data by building a statistical characteristic model, and detecting whether the real-time data deviates significantly from the established statistical distribution, and determining that an interaction abnormality has occurred if there is a significant deviation.

[0010] As a preferred embodiment of the fault location detection method for intelligent GIS equipment based on cloud-edge cooperation described in the present invention, the edge-based small-scale interaction localization model supports local detailed judgment based on feedback to achieve accurate fault location, which includes the following steps: the edge-based small-scale interaction localization model obtains the fault source location information fed back from the cloud, and focuses on monitoring this fault location, while collecting more real-time interaction data; the edge-based small-scale interaction localization model uses correlation analysis to identify the interaction signal with the strongest correlation to the fault, builds a time series model of the interaction signal with the strongest correlation, and searches for subtle abnormal change patterns; the signal is traced to identify the signal source, and combined with the location information fed back from the cloud for filtering, to eliminate unlikely components and accurately identify the fault location; and the location identification result and the interaction signal data are used. and transmitting the set of interaction signals to a cloud-based large-scale interaction detection model, and requesting the cloud-based large-scale interaction detection model to perform a correlation analysis based on global data to determine the fault type and root cause. The method for constructing a time series model of the most strongly correlated interaction signals to search for minute abnormal change patterns includes the following steps: obtaining an interaction signal dataset of the fault source location fed back from the cloud; for each interaction signal, calculating a correlation coefficient between each interaction signal and the fault using the Pearson correlation coefficient method; sorting the correlation coefficients and selecting the top five interaction signals from the sorted list as the signals most strongly correlated with the fault; establishing an ARIMA model based on the most strongly correlated interaction signal, and using the ARIMA model to predict the correlation signal, and simultaneously adjusting the model based on the prediction effect of the ARIMA model; and calculating the model prediction error e. t to detect subtle abnormal change patterns, and to analyze the detected abnormal change patterns to obtain more information about the occurrence of a fault therefrom.

[0011] In a preferred embodiment of the fault location detection method for intelligent GIS equipment based on cloud-edge collaboration described in the present invention, the correlation coefficient ρ XY The specific calculation formula is as follows: JPEG2025538066000002.jpg28150where, ρ XY is the correlation coefficient, σ X and σ Y is the standard deviation of variables X and Y, cov is the covariance of variables X and Y, and ρ XY The range of possible values ​​of is between -1 and 1, and ρ XY The closer the absolute value of ρ is to 1, the stronger the linear relationship between the two variables, and XY The closer the absolute value of ρ is to 0, the weaker the linear relationship between the two variables, and XY > 0 indicates that the two variables are positively correlated, i.e., when one variable increases, the other also increases, and ρ XY < 0 indicates that the two variables are negatively correlated, i.e., when the value of one variable increases, the value of the other variable decreases, and ρ XY = 0 indicates that the two variables are nonlinearly correlated but otherwise correlated, and ρ XY =1 and ρ XY =-1 means that the sample points of the two variables X and Y are both on a straight line.

[0012] In a preferred embodiment of the fault location detection method for intelligent GIS equipment based on cloud-edge collaboration described in the present invention, the ARIMA model includes a lagged-difference test model, an intercept test model, and a trend test model, and the trend test model removes serial correlation by adding a time trend, and the specific calculation formula is as follows: JPEG2025538066000003.jpg27150where t is the time variable, X i is an arbitrary time series, εt is white noise, a is a constant term, and βt is a trend term.

[0013] The actual test starts with the trend test model, if the trend test model is a stationary series, the test stops, otherwise continue testing the intercept test model, if the intercept test model is a stationary series, the test stops, otherwise continue testing whether the lagged difference test model is stationary or not, and as long as the test result of one of the three models can reject the null hypothesis, the time series is considered to be stationary, and if the test results of none of the three models can reject the null hypothesis, the time series is considered to be non-stationary.

[0014] In a preferred embodiment of the fault location method for intelligent GIS equipment based on cloud-edge collaboration described in the present invention, the secondary judgment includes the following steps: querying a repair knowledge base based on the fault result output from the model to find a standard repair solution corresponding to the location result; conducting a simulation based on real-time condition data and the digital twin model to determine the feasibility of the standard solution and individually optimizing and adjusting it to form an optimal repair solution; decomposing the optimized repair solution into a sequence of control commands and issuing the control commands to an edge controller; after receiving the commands, the edge controller checks the rationality and safety of the commands to avoid new faults; the edge controller executes the repair control commands to complete the automatic repair process; after the repair is completed, performing new condition detection and diagnosis to determine whether the fault has been resolved; if the fault has not been resolved, performing localization analysis again until the problem is resolved; recording the current repair event and process and feeding it back to the knowledge base for continuous iterative optimization.

[0015] According to a second aspect, an embodiment of the present invention provides a fault location system for intelligent GIS equipment based on cloud-edge collaboration, including: a digital twin construction module for constructing a digital twin model in the cloud, including all modules of the GIS equipment and their interaction relationships, and simulating changes in the interaction relationships under different fault situations to generate a large amount of fault simulation data; an interaction anomaly detection module for requesting a cloud-based large-scale interaction detection model to perform a global analysis when an interaction anomaly is detected in the edge-based small-scale interaction localization model, so that the cloud-based large-scale interaction detection model comprehensively determines the fault source location and the fault cause, and feeds the result back to the edge-based small-scale interaction localization model; an overall fault localization module for determining a fault type and a root cause based on the localization result, verifying the root cause through the edge-based small-scale interaction localization model, and completing overall fault localization; and a secondary judgment module for performing new condition detection and diagnosis after repair is completed, to confirm whether the fault has been resolved, and if the fault has not been resolved, performing localization analysis again until the problem is resolved.

[0016] According to a third aspect, an embodiment of the present invention provides a computer device comprising: a memory having a computer program stored therein; and a processor, wherein, when the computer program commands are executed by the processor, the steps of the method for fault location detection of intelligent GIS equipment based on cloud-edge cooperation described in the first aspect of the present invention are realized.

[0017] According to a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, the computer program instructions, when executed by a processor, realizing steps of the method for fault location detection of intelligent GIS equipment based on cloud-edge cooperation described in the first aspect of the present invention.

[0018] The beneficial effects of this invention are as follows: By utilizing the idea of ​​cloud-edge collaboration and combining a digital twin model with anomaly change pattern recognition techniques, this invention achieves comprehensive equipment monitoring, fault prediction, and accurate location. Furthermore, the application of a digital twin model to simulate the interaction relationships between GIS equipment modules is a first in the field of GIS equipment fault location, providing new ideas and methods for solving complex equipment fault problems. Furthermore, the cloud-based large-scale interaction detection model built using a graph neural network algorithm can monitor and analyze GIS equipment interaction data in real time, improving the precision and recall of interaction anomaly detection. Furthermore, the use of intelligent algorithms and automatic repair techniques enables automatic repair of equipment faults after fault location, thereby improving the reliability and stability of equipment. [Brief explanation of the drawings]

[0019] In order to more clearly describe the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings used in the description of the embodiments, and it should be understood that the drawings in the following description are only some embodiments of the present invention, and those skilled in the art can derive other drawings based on these drawings without any creative efforts. [Figure 1] 1 is a flowchart of a fault location detection method for intelligent GIS equipment based on cloud-edge collaboration. [Figure 2] This is a diagram showing the computer equipment of the fault location detection method for intelligent GIS equipment based on cloud-edge collaboration. DETAILED DESCRIPTION OF THE INVENTION

[0020] In order to make the above objects, features and advantages of the present invention more clear and understandable, specific embodiments of the present invention will be described in detail below in conjunction with the accompanying drawings of the specification.

[0021] Numerous specific details are set forth below to facilitate a thorough understanding of the present invention; however, the present invention may be practiced in other ways than those described herein, and similar extensions may be made by those skilled in the art without departing from the spirit of the present invention, and therefore the present invention is not limited by the specific examples disclosed below.

[0022] Also, as used herein, "one embodiment" or "embodiment" refers to a particular feature, structure, or characteristic that may be included in at least one embodiment of the present invention. The appearances of the phrase "in one embodiment" in different places in this specification do not all refer to the same embodiment, nor are they mutually exclusive embodiments separate or alternative from other embodiments.

[0023] Example 1 Referring to FIGS. 1 and 2, a first embodiment of the present invention provides a fault location detection method for intelligent GIS equipment based on cloud-edge collaboration, which includes the following steps S1 to S7.

[0024] In S1, a digital twin model including all modules of the GIS equipment and their interactions is built in the cloud, and a large amount of failure simulation data is generated by simulating changes in the interactions under different failure conditions.

[0025] Specifically, the design drawings of the GIS equipment, including mechanical structure diagrams, circuit principle diagrams, control logic diagrams, etc., are collected, and a mechanical model of the equipment, including coordinate systems, motion subdivisions, structural constraints, etc., is constructed using 3D modeling software according to the structure diagram. An electrical model is established using electrical simulation software according to the circuit principle diagram, and parameters for electrical components such as power supplies, sensors, and actuators are set. Further, a control model is constructed using related software according to the control logic, a control program is created, and control cycles and logical judgments are set. The above submodels are integrated to establish a digital twin model and debug the normal operating state of the system.

[0026] In S2, interaction data between components is extracted, a cloud-based large-scale interaction detection model based on a graph neural network is constructed, and the model hyperparameters are optimized. At the same time, the optimized large-scale interaction detection model is deployed on the cloud platform.

[0027] Specifically, from the normal and faulty samples collected by the digital twin, the interaction features between components, including timing data such as voltage, current, temperature, and vibration, are extracted, a connection topology structure diagram between the components is constructed, and the direction of information and energy flow is labeled. A graph neural network algorithm is used to build a model with components as nodes and interactions as edges. Hyperparameters such as the number of network layers and node type are adjusted, a network structure capable of handling temporal relationships is designed, a graph format of the sample set is constructed, the interaction features are configured as node vector inputs, model training is performed to minimize the classification loss function, an interaction anomaly detection model is obtained, indicators such as the model's classification precision and recall are evaluated on the test set, the parameters are repeatedly adjusted, and the optimal hyperparameter configuration is obtained by comparing different structures. The optimized large-scale interaction detection model is placed on a cloud server and packaged as a prediction API interface. The cloud-based large-scale interaction detection model receives interaction data uploaded in real time from the edge-based small-scale interaction location model, enabling prediction of interaction anomalies.

[0028] In S3, sensors are placed locally on the GIS equipment to collect component interaction data, and an edge-based small-scale interaction localization model is constructed to analyze the interaction data.

[0029] Preferably, according to the structural drawing of the GIS equipment, determine the placement points of the main components, select appropriate sensors such as current, voltage, and vibration sensors, formulate a sensor distribution plan, place various sensors in the corresponding locations, set the sensor range and accuracy parameters, and write a driver program to collect real-time monitoring data for each sensor.

[0030] Furthermore, an edge-based small-scale interaction localization model is constructed at the edge of the equipment to aggregate and analyze the collected data, a topological connection structure between components is constructed, a link model of information and control flow is set up, an interaction monitoring algorithm is designed to analyze the real-time interaction data between components, and if the interaction data exceeds the judgment criterion, it is determined to be abnormal, a localization analysis program is launched, the topological structure and real-time data are integrated, fault isolation and localization estimation is performed, the faulty components are accurately identified, and the results are output.

[0031] In S4, when the edge-based small-scale interaction localization model detects an interaction abnormality, it requests the cloud-based large-scale interaction detection model to perform a global analysis, and the cloud-based large-scale interaction detection model comprehensively determines the fault source location and fault cause, and feeds back the results to the edge-based small-scale interaction localization model.

[0032] Specifically, the process includes the following steps S4.1 to S4.5.

[0033] In S4.1, the edge-based small-scale interaction localization model continuously detects interaction data between components, and if the interaction data meets the judgment criteria, the edge-based small-scale interaction localization model determines that an interaction anomaly has occurred and obtains a snapshot of the interaction data at the moment the anomaly occurred.

[0034] Specifically, the interaction data deviation must exceed μ+3σ, and the interaction response time must be 2T eThe criteria for determining whether the interaction mode exceeds the threshold, a significant change occurs in the interaction mode, the interaction session is interrupted, or an abnormal change occurs in the statistical characteristics of the interaction data.

[0035] Specifically, for interaction data deviation anomalies, the mean value μ and standard deviation σ of normal data are calculated, and the interaction data deviation threshold is determined as μ + 3σ. If the real-time data exceeds the interaction data deviation threshold μ + 3σ, it is determined to be an interaction anomaly. For interaction response time anomalies, the mean response time T e is calculated and determined, and the interaction response time is 2T e If the average response time T e is the average value of the historical response time, and when a significant change occurs in the interaction mode, time series analysis is used to determine it, and the normal interaction mode is established by analyzing the series trend, periodicity, seasonality, etc., and the real-time interaction data is compared with the normal interaction mode, and if there is a mismatch, it is determined to be an interaction abnormality, and when an interaction session is interrupted, a heartbeat detection mechanism is set up for key interaction signals to determine it, and if the heartbeat is a time attack, it is determined to be an interaction abnormality, and when an abnormal change occurs in the statistical characteristics of the interaction data, a statistical characteristic model is constructed to determine whether the real-time data deviates significantly from the established statistical distribution, and if there is a significant deviation, it is determined to be an interaction abnormality.

[0036] In S4.2, a snapshot of the interaction data is packaged and a global analysis request is sent to a cloud-based large-scale interaction detection model via the network.

[0037] In S4.3, the cloud-based large-scale interaction detection model receives the data, reconstructs this fault scenario in the digital twin environment, and simulates the propagation and impact of the interaction data to identify the fault source location.

[0038] In S4.4, the cloud-based large-scale interaction detection model comprehensively determines the fault source location, and performs experiments and simulations on the fault source location to estimate the possible fault causes.

[0039] Preferably, after the fault source location is identified, experiments and simulations should be conducted on the fault source location to identify the problem and objective to be solved, clarify the fault location and interaction process to be tested and simulated, design an appropriate experimental scheme according to the characteristics of the fault location and interaction process, simulate a specific testing environment, input conditions and other parameter settings according to the requirements of the fault location and interaction process, record and observe the experimental data, phenomena and changes, use sensors, recorders and other equipment to collect data generated during the experiment, and analyze and process the collected experimental data to find patterns, trends and correlations, observe and analyze the data to find the suspected causes of the failure or abnormal operation, and verify and confirm the suspected causes of the failure based on the experimental results.

[0040] In S4.5, the estimated fault source location and fault cause are fed back and transmitted to the edge-based small-scale interaction localization model.

[0041] In S5, the edge-based small-scale interaction localization model assists in local detailed judgment based on feedback to achieve accurate fault localization, and sends the localization results to the cloud-based large-scale interaction detection model to request correlation analysis.

[0042] Specifically, the process includes the following steps S5.1 to S5.4.

[0043] In S5.1, the edge-based small-scale interaction localization model obtains the information of the fault source location fed back from the cloud, and focuses on monitoring this fault location while collecting more real-time interaction data.

[0044] In S5.2, correlation analysis is used to identify the interaction signal that has the strongest correlation with the fault, and a time series model of the interaction signal with the strongest correlation is constructed to search for subtle abnormal change patterns.

[0045] Specifically, the process includes the following steps S5.2.1 to S5.2.5.

[0046] In S5.2.1, the interaction signal dataset of the fault source part fed back from the cloud is acquired.

[0047] In S5.2.2, for each interaction signal, the correlation coefficient between each interaction signal and a fault is calculated using the Pearson correlation coefficient method.

[0048] Specifically, the correlation coefficient ρ XY The specific calculation formula is as follows: JPEG2025538066000004.jpg28150where, ρ XY is the correlation coefficient, σ X and σ Y represents the standard deviation of variables X and Y, and cov represents the covariance of variables X and Y.

[0049] In addition, ρ XY The range of possible values ​​of is between -1 and 1, and ρ XY The closer the absolute value of ρ is to 1, the stronger the linear relationship between the two variables, and XY The closer the absolute value is to 0, the weaker the linear relationship between the two variables, and ρ XY > 0 indicates that the two variables are positively correlated, i.e., when one variable increases, the other also increases, and ρ XY < 0 indicates that the two variables are negatively correlated, i.e., when the value of one variable increases, the value of the other variable decreases, and ρ XY = 0 indicates that the two variables are nonlinearly correlated but otherwise correlated, and ρ XY =1 and ρ XY When =-1, it means that the two X and Y can be well described by a linear equation, and all the sample points lie on a straight line.

[0050] In S5.2.3, the correlation coefficients are sorted and the top five interacting signals from the sorted list are selected as the signals most strongly correlated with the fault.

[0051] Specifically, obtain all the correlation coefficients calculated in S5.2.2, and record each interaction signal and its corresponding correlation coefficient in a list; sort the interaction signals in the list in descending order according to the magnitude of the correlation coefficient, where a higher correlation coefficient indicates a stronger correlation between the signal and the fault; select the top five interaction signals from the sorted list as the signals with the strongest correlation with the fault; and for the selected interaction signals, perform further analysis and research to determine their relationship with the fault.

[0052] In S5.2.4, an ARIMA model is established based on the interaction signal with the strongest correlation, and the correlation signal is predicted using the ARIMA model, and at the same time, the model is adjusted based on the prediction effect of the ARIMA model.

[0053] Specifically, the time series data of the signal with the strongest correlation is collected, and the stationarity of the time series is tested using ADF. If the time series is not stationary, a stationary series is obtained by differential processing, and a white noise test is performed on the stationary series. If the stationary series is white noise, the p and q orders of the ARIMA model are determined. The values ​​of p and q are first determined using ACK (autocorrelation) and PAC (partial correlation) plots. An ARIMA (p, d, q) model is constructed, and the model parameters are estimated using the maximum likelihood method. The optimal model is determined using the AIC and BIC criteria, and the p and q parameters are adjusted to obtain the model prediction value. Generate JPEG2025538066000005.jpg2316 and the prediction error e t Calculate the prediction error e t If does not match white noise, adjust the modeling parameters and rebuild the ARIMA model.

[0054] The test for stationarity of a time series using ADF is realized by three models. Here, a lag period is added to the lagged difference test model to remove serial correlation. The specific calculation formula is as follows: JPEG2025538066000006.jpg36150where t is the time variable, X i represents an arbitrary time series, and εt represents white noise.

[0055] An intercept term is added to the intercept test model to remove serial correlation, and the specific calculation formula is as follows: JPEG2025538066000007.jpg31150where t is the time variable, X i represents an arbitrary time series, εt represents white noise, and a represents a constant term.

[0056] A time trend is added to the trend test model to remove serial correlation, and the specific calculation formula is as follows: JPEG2025538066000008.jpg27150where t is the time variable, X i is an arbitrary time series, εt is white noise, a is a constant term, and βt is a trend term.

[0057] Furthermore, the actual test starts with the trend test model, and if the trend test model is a stationary series, the test stops; otherwise, continue testing the intercept test model; if the intercept test model is a stationary series, the test stops; otherwise, continue testing whether the lagged difference test model is stationary or not; various model tests also have corresponding critical value tables; as long as the test result of one of the three models can reject the null hypothesis, the time series is considered to be stationary; if none of the test results of the three models can reject the null hypothesis, the time series is considered to be non-stationary.

[0058] Note that the original assumption was H0:δ=0, i.e., the existence of one unit root.

[0059] In S5.2.5, the model prediction error e tThe present invention is based on the above method to detect minute abnormal change patterns, and analyze the detected abnormal change patterns to obtain more information about the occurrence of a fault.

[0060] Specifically, the prediction error e between the model predicted value and the actual observed value t Calculate the prediction error e t We perform a white noise test on the prediction error e t If is consistent with the white noise, the model has a good fitting effect; otherwise, it indicates an abnormal change, and the prediction error e t If it is detected that the abnormal change pattern does not match the white noise, the abnormal change pattern is recognized, and the position, time and duration of the abnormal change are determined by observing the residual time series diagram, and the degree of deviation of the residual is calculated to determine the degree of abnormality. The recognized abnormal change pattern is analyzed to determine whether it is a system fault signal.

[0061] In S5.3, the signal is traced to identify the signal source, and then filtered in combination with part information fed back from the cloud to eliminate unlikely components and accurately identify the fault location.

[0062] Specifically, based on the detected fault signal and the sensor location, a preliminary determination is made of the possible source of the abnormal signal. According to the amplitude size and frequency characteristics of the abnormal signal fluctuation, the possible source of the abnormal signal is further narrowed down by referring to the operating characteristics of different components. Temporary sensors are installed in specific locations to collect more correlation signal data. The behavior of the abnormal signal from these sensors is observed to identify the exact source of the abnormal signal. The identified source of the fault signal is compared with the records of historical fault locations in the cloud platform database to filter out matching fault locations. This is combined with the part failure probability analysis results fed back from the expert experience system to exclude unlikely fault locations determined by the system. The consistency of the integrated multiple model results is compared to determine the final fault location. The final fault part is inspected on site, the actual fault situation is confirmed to verify the accuracy of the fault location, the relationship between the fault location and the fault signal source is labeled, and this is fed back to the cloud platform. The database is continuously improved to improve subsequent location identification efficiency.

[0063] In S5.4, the location determination result and the interaction signal dataset are fed back and transmitted to the cloud-based large-scale interaction detection model, and the cloud-based large-scale interaction detection model is requested to perform correlation analysis based on global data to determine the fault type and root cause.

[0064] In S6, the cloud-based large-scale interaction detection model gives the fault type and root cause based on the localization result, and the edge-based small-scale interaction localization model verifies the root cause to complete the overall fault localization.

[0065] Specifically, the process includes the following steps S6.1 to S6.4.

[0066] In S6.1, the cloud-based large-scale interaction detection model receives the accurate fault location data fed back from the edge-based small-scale interaction localization model and constructs the relevant topology subgraph of this location on the digital twin platform.

[0067] S6.2: Comprehensively analyze and determine the most likely failure pattern by simulating the propagation effect under different failure conditions, and determine the appropriate failure type based on the failure pattern.

[0068] Specifically, a simulation is performed to inject different types of faults into the subgraph, and the fault propagation effect is observed. Under the condition that the fault location is given, the propagation effect of different fault patterns is analyzed, the fault pattern that is most likely to match the actual situation is determined, and the appropriate fault type is determined based on the fault pattern database.

[0069] S6.3: Use complex system analysis techniques to determine the most likely failure occurrence chain, analyze the failure occurrence chain to obtain the root cause of the failure, and feed back the estimated failure type and root cause to the edge-based small-scale interaction localization model.

[0070] Preferably, the critical locations of the fault locations in the topology diagram are analyzed to construct possible fault occurrence chains, the most likely fault occurrence chains are determined by simulating them based on fault propagation, and degradation analysis is performed along the fault occurrence chains to estimate the most likely root causes of the faults.

[0071] S6.4: After the edge-based small-scale interaction localization model receives the cause, it checks whether the real-time monitoring data supports the cause.

[0072] Specifically, if it is found that the real-time data does not match the cause, the problem to be analyzed is provided again, and the cloud-edge collaborative analysis is rerun. If the data verification passes, the provided fault localization results are confirmed, and the overall fault localization is completed.

[0073] In S7, based on the fault results output from multiple models, the repair knowledge base is consulted to determine the optimal solution, and a control command is issued to automate the repair process, and the system makes a secondary judgment to ensure the fault is resolved.

[0074] Furthermore, a repair knowledge base is queried based on the failure results output from the model to find a standard repair solution corresponding to the location result. The knowledge base compiles repair strategies for various types of failures based on historical cases, and simulations are performed based on real-time state data and the digital twin model to determine the feasibility of the standard solution, and individual optimization and adjustment are performed to form an optimal repair solution. The optimized repair solution is decomposed into sequential control commands and issued to the edge controller. The control commands include switching, restarting, and reconfiguring equipment modules. After receiving the commands, the edge controller checks the rationality and safety of the commands to avoid new failures. The edge controller then executes the repair control commands and completes the automatic repair process.

[0075] Furthermore, after the repair is completed, new condition detection and diagnosis is performed to check whether the fault has been resolved, and if the fault has not been resolved, localization analysis is performed again until the problem is resolved, and the current repair events and processes are recorded and fed back into the knowledge base for continuous iterative optimization.

[0076] Furthermore, this embodiment further provides a fault location system for intelligent GIS equipment based on cloud-edge collaboration, including: a digital twin construction module, which is used to construct a digital twin model in the cloud, including all modules of the GIS equipment and their interaction relationships, and simulate changes in the interaction relationships under different fault situations, thereby generating a large amount of fault simulation data; an interaction anomaly detection module, which is used to request a cloud-based large-scale interaction detection model to perform a global analysis when an interaction anomaly is detected in the edge-based small-scale interaction localization model, so that the cloud-based large-scale interaction detection model comprehensively determines the fault source location and the fault cause, and feeds the result back to the edge-based small-scale interaction localization model; an overall fault localization module, which is used to assign a fault type and a root cause based on the localization result, verify the root cause through the edge-based small-scale interaction localization model, and complete overall fault localization; and a secondary judgment module, which is used to perform new condition detection and diagnosis after repair is completed, to check whether the fault has been resolved, and if the fault has not been resolved, to perform localization analysis again until the problem is resolved.

[0077] This embodiment further provides a computer device applicable to the fault location detection method for intelligent GIS equipment, the computer device comprising a memory and a processor, the memory being used to store computer-executable commands, and the processor being used to execute the computer-executable commands, thereby realizing the fault location detection method for intelligent GIS equipment provided in the above embodiment.

[0078] The computer device may be a terminal, and includes a processor, memory, a communication interface, a display, and an input device connected via a system bus. The processor of the computer device is used to provide calculation and control functions. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and computer programs. The internal memory provides an operating environment for the operating system and computer programs stored in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal via a wired or wireless method, which may be achieved by Wi-Fi, an operator network, NFC (near field communication), or other technologies. The display of the computer device may be a liquid crystal display or an electronic ink display, and the input device of the computer device may be a touch layer covered by the display, buttons, a trackball, or a touchpad on the housing of the computer device, or an external keyboard, touchpad, or mouse.

[0079] This embodiment further provides a storage medium having a computer program stored thereon, which, when executed by a processor, realizes the fault location detection method for intelligent GIS equipment presented in the above embodiment.

[0080] In summary, this invention utilizes the idea of ​​cloud-edge collaboration and combines a digital twin model with anomaly change pattern recognition techniques to achieve comprehensive equipment monitoring, fault prediction, and accurate location. Furthermore, the application of a digital twin model to simulate the interaction relationships between each module of GIS equipment is a first in the field of GIS equipment fault location, providing new ideas and methods for solving complex equipment fault problems. Furthermore, a cloud-based large-scale interaction detection model built using a graph neural network algorithm can monitor and analyze GIS equipment interaction data in real time, improving the precision and recall of interaction anomaly detection. Furthermore, the use of intelligent algorithms and automatic repair techniques allows equipment faults to be automatically repaired after fault location, thereby improving the reliability and stability of the equipment.

[0081] Example 2 Referring to Figures 1 and 2, a second embodiment of the present invention provides a fault location detection method for intelligent GIS equipment based on cloud-edge collaboration, and scientific demonstration was carried out through economic effect calculations and simulation experiments to verify the beneficial effects of the present invention.

[0082] Specifically, GIS design drawings were collected, and a simulation model of the drawings was constructed as a digital twin. A large amount of failure simulation data was generated by simulating the changes in the interaction relationships under different failure conditions, from which 100 pairs of failure data and normal data were collected, some of which are shown in Tables 1 and 2.

[0083] Table 1. Normal data table for some GIS equipment JPEG2025538066000009.jpg41150 Table 2. Failure data table for some GIS equipment JPEG2025538066000010.jpg40150Furthermore, using the data in Tables 1 and 2, we extract sensor output and actuator input as features and build a cloud-based large-scale interaction detection model based on a graph neural network.Sensors are installed on the edge of the equipment to collect normal operating data in real time, and an edge-based small-scale interaction location model is built.If the sensor output exceeds a threshold, it is determined to be an abnormality.

[0084] The edge-based small-scale interaction localization model then detects that Sensor A has an abnormal output of 11.8V and sends the data to the cloud platform. The cloud-based large-scale interaction detection model then analyzes it using the digital twin and determines that Sensor A is the source of the fault. The edge-based small-scale interaction localization model collects 100 sets of Sensor A data and finds that the output fluctuates between 11 and 13V. It then uses ADF to test the stationarity of the time series and, in combination with the ADF distribution critical value table, determines that the time series is stable and that the stable series is white noise. An ARIMA model is then constructed to detect that the error does not conform to white noise. Then, abnormal change pattern recognition is performed to confirm that Sensor A is faulty. The ADF distribution critical value table is shown in Table 3.

[0085] Table 3 ADF distribution critical value table JPEG2025538066000011.jpg63150Furthermore, the cloud-based large-scale interactive detection model receives the location result, simulates the fault propagation in the digital twin, determines that the fault type is deviation drift of sensor A, queries the repair knowledge base according to the deviation drift, and finds a standard repair solution corresponding to the deviation drift. At the same time, it performs simulation based on the real-time condition data and the digital twin model to verify the feasibility of the standard solution, and individually optimizes and adjusts it to form an optimal repair solution. The optimal repair solution is decomposed into a sequencing control command and issued to adjust the calibration parameters of sensor A. After the repair is completed, a secondary confirmation is performed to confirm that the fault has been resolved.

[0086] Furthermore, a comparison of the detection and location of the present invention with conventional GIS equipment is shown in Table 4.

[0087] Table 4 Comparison of detection and location of the present invention with conventional GIS equipment JPEG2025538066000012.jpg41150 Preferably, as can be seen from the results in Table 4, the present invention has significant advantages in many aspects, including fault detection precision, localization precision, response time, repair time, and maintenance costs. The cloud-edge collaboration-based detection and localization idea adopted in this invention improves analysis and judgment capabilities through digital twin and multi-source data fusion, resulting in an 11% and 10% improvement in fault detection precision and localization precision, respectively, compared to conventional methods.

[0088] The response time and repair time of the present invention are more advantageous than those of conventional methods, which significantly reduces the downtime of faulty equipment and significantly improves system availability. In terms of maintenance costs, the present invention can reduce them by 23%, while conventional methods do not achieve any significant effect.

[0089] It should be noted that the above-described embodiments are only for illustrating the technical solutions of the present invention, and are not intended to limit the same. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art may make modifications or equivalent substitutions to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and these modifications or substitutions should be included in the scope of the claims of the present invention.

Claims

1. A digital twin model including all modules of the GIS equipment and their interactions is constructed in the cloud, and a large amount of failure simulation data is generated by simulating changes in the interactions under different failure conditions. Extracting interaction data between components, building a cloud-based large-scale interaction detection model based on a graph neural network, optimizing the model hyperparameters, and simultaneously deploying the optimized large-scale interaction detection model on a cloud platform; A sensor is placed locally on the GIS device to collect interaction data of the components, and an edge-based small-scale interaction localization model is constructed to analyze the interaction data; When the edge-based small-scale interaction localization model detects an interaction abnormality, it requests a cloud-based large-scale interaction detection model to perform a global analysis, and the cloud-based large-scale interaction detection model comprehensively determines the fault source location and the fault cause, and feeds back the result to the edge-based small-scale interaction localization model; The edge-based small-scale interaction localization model supports local detailed judgment based on feedback to achieve accurate fault localization, and sends the localization result to a cloud-based large-scale interaction detection model to request correlation analysis; Giving a fault type and a root cause based on the localization result by a cloud-based large-scale interaction detection model, and verifying the root cause by an edge-based small-scale interaction localization model to complete the overall fault localization; A fault location detection method for intelligent GIS equipment based on cloud-edge collaboration, comprising: querying a repair knowledge base to determine an optimal solution based on the fault results output from multiple models, and issuing a control command to automate the repair process, and the system makes a secondary judgment to ensure the fault is resolved.

2. Requesting global analysis from the above-mentioned cloud-based large-scale interaction detection model includes the following steps: The edge-based small-scale interaction localization model continuously detects interaction data between components, and if the interaction data meets a judgment criterion, the edge-based small-scale interaction localization model determines that an interaction anomaly has occurred, and obtains a snapshot of the interaction data at the moment the anomaly occurs; The edge-based small-scale interaction localization model packages a snapshot of the interaction data and requests a global analysis from a cloud-based large-scale interaction detection model via a network; a cloud-based large-scale interaction detection model receiving the data, reconstructing the fault scenario in a digital twin environment, and simulating the propagation and impact of the interaction data to identify the fault source location; The cloud-based large-scale interaction detection model comprehensively judges the fault source location, and performs experiments and simulations on the fault source location to estimate possible fault causes; The fault location detection method for intelligent GIS equipment based on cloud-edge collaboration as claimed in claim 1, characterized in that the cloud-based large-scale interaction detection model includes a step of feeding back and transmitting the estimated fault source location and fault cause to the edge-based small-scale interaction location identification model.

3. Determining that the above-described interaction anomaly has occurred includes: Interaction data deviation exceeds μ+3σ, interaction response time is 2T ｅ determining the criterion as exceeding the threshold, a significant change in the interaction mode, an interruption of the interaction session, or an abnormal change in the statistical characteristics of the interaction data; Regarding the interaction data deviation abnormality, calculating the mean value μ and standard deviation σ of the normal data and determining the interaction data deviation threshold as μ + 3σ, and determining that the interaction data is abnormal if the real-time data exceeds the interaction data deviation threshold μ + 3σ; For interaction response time anomalies, the average response time T ｅ is calculated and determined, and the interaction response time is 2T ｅ If the difference exceeds 1, it is determined that the interaction is abnormal. When a significant change occurs in the interaction mode, use time series analysis to determine the normal interaction mode by analyzing the series trend, periodicity and seasonal characteristics, and compare the real-time interaction data with the normal interaction mode. If there is a discrepancy, it is determined to be an interaction abnormality; A heartbeat detection mechanism is set for a key interaction signal to determine whether an interaction session is interrupted, and if the heartbeat is timed out, it is determined that an interaction abnormality has occurred; 3. The method for detecting the location of a fault in an intelligent GIS device based on cloud-edge collaboration according to claim 2, characterized in that it comprises: constructing a statistical feature model to judge whether an abnormal change occurs in the statistical features of the interaction data; detecting whether the real-time data deviates significantly from the established statistical distribution; and determining that an interaction abnormality has occurred if there is a significant deviation.

4. The above-mentioned edge-based small-scale interaction localization model supports local detailed judgment based on feedback to achieve accurate fault localization by the following steps: The edge-based small-scale interaction localization model obtains information of the fault source location fed back from the cloud, and focuses on monitoring the fault location, while collecting more real-time interaction data; Using correlation analysis to identify the interaction signal that has the strongest correlation with the fault, and building a time series model of the interaction signal that has the strongest correlation to search for subtle abnormal change patterns; A step of tracing the signal to identify the signal source, and filtering the signal in combination with the part information fed back from the cloud to eliminate unlikely components and accurately identify the fault location; Feedback and transmit the location determination result and the interaction signal dataset to a cloud-based large-scale interaction detection model, and request the cloud-based large-scale interaction detection model to perform correlation analysis based on global data to determine the fault type and root cause; Building a time series model of the interaction signal with the strongest correlation mentioned above and searching for subtle abnormal change patterns can be achieved by the following steps: Obtaining an interaction signal dataset of the fault source site fed back from the cloud; calculating, for each interaction signal, a correlation coefficient between each interaction signal and a fault using the Pearson correlation coefficient method; sorting the correlation coefficients and selecting the top five interacting signals from the sorted list as the signals most strongly correlated with the fault; Establishing an ARIMA model based on the most strongly correlated interaction signals, and predicting the correlation signals using the ARIMA model, and simultaneously adjusting the model based on the prediction effect of the ARIMA model; Model prediction error e ｔ and determining whether the fault location of an intelligent GIS device is based on the cloud-edge collaboration of claim 1 , further comprising: determining whether the fault location of an intelligent GIS device is based on the cloud-edge collaboration of claim 1 , and then detecting a subtle abnormal change pattern using the determined abnormal change pattern; and analyzing the detected abnormal change pattern to obtain more information about the occurrence of the fault therefrom.

5. The correlation coefficient ρ ＸＹ The specific calculation formula is as follows: where ρ ＸＹ is the correlation coefficient, σ Ｘ and σ Ｙ represents the standard deviation of variables X and Y, and cov represents the covariance of variables X and Y. ρ ＸＹ The range of possible values ​​of is between -1 and 1, and ρ ＸＹ The closer the absolute value of ρ is to 1, the stronger the linear relationship between the two variables, and ＸＹ The closer the absolute value of is to 0, the weaker the linear relationship between the two variables. ρ ＸＹ > 0 indicates that the two variables are positively correlated, i.e., when one variable increases, the other variable also increases; ρ ＸＹ < 0 indicates that the two variables are negatively correlated, i.e., when the value of one variable increases, the value of the other variable decreases; ρ ＸＹ = 0 indicates that the two variables are nonlinearly correlated but otherwise correlated; ρ ＸＹ = 1 and ρ ＸＹ The fault location detection method for intelligent GIS equipment based on cloud-edge cooperation according to claim 4, characterized in that when = -1, it means that the sample points of the two variables X and Y are both on a single straight line.

6. The ARIMA model includes a lagged difference test model, an intercept test model, and a trend test model. The trend test model removes serial correlation by adding a time trend. The specific calculation formula is as follows: where t is the time variable, X ｉ is an arbitrary time series, εt is white noise, a is a constant term, and βt is a trend term. The actual test starts with the trend test model, and if the trend test model is a stationary series, the test stops; otherwise, the test continues with the intercept test model; If the intercept test model is stationary, stop the test; otherwise, continue to test whether the lagged difference test model is stationary; As long as the test result of one of the three models can reject the null hypothesis, the time series is considered to be stationary. The fault location detection method for intelligent GIS equipment based on cloud-edge collaboration according to claim 4, characterized in that if none of the test results of the three models can reject the null hypothesis, the time series is considered to be non-stationary.

7. The above-mentioned secondary judgment includes the following steps: querying a repair knowledge base based on the failure results output from the model to find a standard repair solution corresponding to the location results; Conducting simulations based on real-time condition data and digital twin models to determine the feasibility of standard solutions, and then individually optimizing and adjusting them to form optimal repair solutions; decomposing the optimized repair solution as ordered control commands and issuing the control commands to an edge controller; After receiving the command, the edge controller checks the rationality and safety of the command to avoid new failures; the edge controller executes a repair control command to complete an automatic repair process; After the repair is completed, a new status detection and diagnosis is performed to check whether the fault has been resolved; If the fault has not been resolved, performing the localization analysis again until the problem is resolved; and recording the current repair event and process and feeding it back to the knowledge base for continuous iterative optimization.

8. A fault location detection system for intelligent GIS equipment based on cloud-edge collaboration, the system being based on the fault location detection method for intelligent GIS equipment based on cloud-edge collaboration according to claim 1, a digital twin construction module that is used to generate a large amount of failure simulation data by constructing a digital twin model in the cloud, including all modules of the GIS equipment and their interaction relationships, and simulating changes in the interaction relationships under different failure conditions; and an interaction anomaly detection module, which is used to request a global analysis from a cloud-based large-scale interaction detection model when an interaction anomaly is detected in the edge-based small-scale interaction localization model, and the cloud-based large-scale interaction detection model comprehensively determines the fault source location and the fault cause, and feeds back the result to the edge-based small-scale interaction localization model; a global fault localization module, which is used to give a fault type and a root cause based on the localization result, and verify the root cause through an edge-based small-scale interaction localization model to complete the global fault localization; and a secondary judgment module that is used to perform new status detection and diagnosis after repair is completed, to check whether the fault has been resolved, and if the fault has not been resolved, to perform location analysis again until the problem is resolved.

9. 1. A computer device comprising: a processor having a computer program stored therein; and the computer device, when the processor executes the computer program, realizing the steps of the method for detecting a fault location of an intelligent GIS device based on cloud-edge cooperation according to claim 1.

10. 1. A computer-readable storage medium having a computer program stored thereon, the computer program being characterized in that, when executed by a processor, the computer program implements the steps of the method for detecting a fault location of an intelligent GIS device based on cloud-edge cooperation as set forth in claim 1.

Images