Intelligent analysis and diagnosis method and system for substation faults based on big data

Abstract

The present application provides a method and system for intelligent analysis and diagnosis of substation faults based on big data, wherein the method includes: obtaining temperature data streams on the surface and inside of each power equipment in the substation and the ambient temperature and humidity data stream of the substation in the substation operating environment; dynamically coupling the temperature data stream and the ambient temperature and humidity data stream across time scales to generate a heterogeneous temperature characteristic stream; dynamically adjusting the association rules between power equipment temperature anomalies and substation operating status based on the heterogeneous temperature characteristic stream; performing cross-device coupling analysis of the heterogeneous temperature characteristic stream through a deep belief network to generate a dynamic conduction topology structure; and combining the adjusted fault association rules with the dynamic conduction topology structure to generate substation fault diagnosis results. The present application improves the accuracy of early warning and diagnostic efficiency of substation faults.

Description

Technical Field

[0001] The present application relates to the technical field of intelligent operation and maintenance of power systems, and in particular to a method and system for intelligent analysis and diagnosis of substation faults based on big data. Background Art

[0002] As smart grid construction progresses, substation equipment scales up and operational complexity increases, making traditional fault diagnosis methods less effective and accurate. Modern substations require intelligent analysis technology that can integrate multi-source monitoring data, automatically identify abnormal relationships between equipment, and provide early warning of faults. This technology addresses the need for precise control of equipment status under high-load operating conditions.

[0003] Existing solutions utilize a temperature anomaly diagnosis model based on a support vector machine. This model collects device surface temperature data, combines it with preset temperature threshold rules for anomaly detection, and uses historical device failure data to train a classification model to identify common fault types. This data-driven approach improves diagnostic efficiency and reduces manual intervention.

[0004] This solution relies solely on a single data dimension, namely, the device surface temperature. Its static threshold rules are difficult to adapt to temperature variations under different operating conditions, resulting in a high false alarm rate. Furthermore, the model lacks the ability to analyze the thermal conductivity relationships between devices. Summary of the Invention

[0005] The present application provides a method and system for intelligent analysis and diagnosis of substation faults based on big data, which is used to solve the problems of low accuracy and poor diagnostic efficiency of early warning of substation faults in the prior art.

[0006] In a first aspect, the present application provides a method for intelligent analysis and diagnosis of substation faults based on big data, comprising:

[0007] In the substation operating environment, obtain the surface and internal temperature data streams of each power equipment in the substation, as well as the ambient temperature and humidity data streams of the substation;

[0008] Dynamically coupling the temperature data stream and the ambient temperature and humidity data stream across time scales to generate a heterogeneous temperature feature stream;

[0009] Based on the heterogeneous temperature characteristic flow, dynamically adjust the association rules between power equipment temperature anomalies and substation operating status;

[0010] Performing cross-device coupling analysis on the heterogeneous temperature characteristic flows through a deep belief network to generate a dynamic conduction topology structure;

[0011] The adjusted fault association rules are combined with the dynamic conduction topology to generate a substation fault diagnosis result.

[0012] Optionally, performing cross-device coupling analysis on the heterogeneous temperature characteristic flows through a deep belief network to generate a dynamic conduction topology structure includes:

[0013] Performing multi-level feature processing on the heterogeneous temperature feature stream through a deep belief network to construct an initial topological network;

[0014] Iteratively modifying the initial topology network based on the time persistence of the heterogeneous temperature characteristic flow to generate a target topology network;

[0015] The target topology network is spatially matched with the physical location coordinate data of each power device in the substation to generate a dynamic conduction topology structure.

[0016] Optionally, spatially matching the target topology network with physical location coordinate data of each power device in the substation to generate a dynamic conduction topology structure includes:

[0017] Obtaining physical location coordinate data of each power equipment in the substation;

[0018] Comparing the connection paths between nodes in the target topology network with the physical location coordinate data to screen out all initial valid conductive connection paths whose lengths are less than a preset spatial distance threshold;

[0019] weighting each of the initial effective conduction connection paths according to the electrical circuit relationship and heat dissipation area orientation of each power equipment in the substation; after the correction is completed, deleting redundant paths that do not meet the preset conduction conditions to generate a target effective conduction connection path;

[0020] The target effective conductive connection path is mapped to the actual layout orientation in the physical position coordinate data to generate a dynamic conductive topology structure.

[0021] Optionally, the weight correction of each of the initial effective conductive connection paths according to the electrical circuit relationship and heat dissipation area orientation of each power equipment in the substation includes:

[0022] Identifying device node pairs belonging to the same electrical connection group in all the initial valid conductive connection paths, and performing same-loop enhancement processing on the conductive weight values ​​of the device node pairs according to a preset ratio, wherein the electrical connection group is a set of device groups divided according to electrical loop relationships;

[0023] Calculating the conduction attenuation factor of the device node pair across the intervals based on the azimuth angle data of adjacent heat dissipation intervals in the heat dissipation area orientation;

[0024] Performing a product operation on the conduction weight value after the same-loop enhancement process and the conduction attenuation factor to generate a comprehensive modified weight value;

[0025] The weight of each of the initial effective conductive connection paths is corrected according to the comprehensive correction weight value.

[0026] Optionally, performing multi-level feature processing on the heterogeneous temperature feature stream by using a deep belief network to construct an initial topological network includes:

[0027] Using a first-level processing module of a deep belief network, the temperature fluctuation synchronization features corresponding to the devices within the group units are extracted from the heterogeneous temperature feature stream based on the physical location relationship of the power equipment in the substation;

[0028] The temperature fluctuation synchronization feature is superimposed on the substation environment temperature and humidity data stream by a second-level processing module to generate an intermediate feature set;

[0029] The temperature correlation features exceeding the preset conduction threshold are screened out from the intermediate feature set through the third-level processing module, and an initial topology network is constructed based on the conduction strength relationship between the various power equipment in the substation corresponding to the temperature correlation features. The temperature correlation features refer to the temperature correlation features between the various power equipment in the substation.

[0030] Optionally, dynamically coupling the temperature data stream and the ambient temperature and humidity data stream across time scales to generate a heterogeneous temperature feature stream includes:

[0031] Divide the temperature sampling period of each power device into a second-level time window, synchronously match the ambient temperature and humidity data with the temperature data stream within the second-level time window, and generate an instantaneous temperature data unit;

[0032] Divide the hourly time window according to the load change cycle of the substation, and accumulate the instantaneous temperature data units to extract the temperature trend change characteristics of each power equipment within the hourly time window;

[0033] Combine the substation's historical operating data to divide the monthly time window, associate and map the temperature trend change characteristics with the historical temperature fluctuation range of the same period under the same seasonal conditions, and generate a temperature feature set;

[0034] The instantaneous temperature data unit, the temperature trend change feature, and the temperature feature set are hierarchically integrated according to timestamps to form a heterogeneous temperature feature stream.

[0035] Optionally, combining the adjusted fault association rules with the dynamic conduction topology structure to generate a substation fault diagnosis result includes:

[0036] Marking a device node whose conduction strength in the dynamic conduction topology exceeds a preset alarm threshold as a suspected fault source node, and extracting a temperature anomaly characteristic pattern corresponding to the suspected fault source node;

[0037] Matching and verifying the temperature anomaly characteristic pattern with the adjusted fault association rules; when the matching verification result indicates that the suspected fault source node meets both the equipment aging gradient change law and the ambient temperature and humidity coupling conditions in the adjusted fault association rules, the suspected fault source node is confirmed as a confirmed fault source node;

[0038] generating risk warning information of associated devices according to the strength of the conduction paths between the confirmed fault source node and other device nodes in the dynamic conduction topology structure;

[0039] The location information of the confirmed fault source node is combined with the risk warning information of the associated equipment to generate a substation fault diagnosis result.

[0040] In a second aspect, the present application provides a substation fault intelligent analysis and diagnosis system based on big data, comprising:

[0041] The acquisition module is used to obtain the surface and internal temperature data streams of each power equipment in the substation and the ambient temperature and humidity data stream of the substation in the substation operating environment;

[0042] A coupling module, configured to dynamically couple the temperature data stream and the ambient temperature and humidity data stream across time scales to generate a heterogeneous temperature feature stream;

[0043] An adjustment module, configured to dynamically adjust the association rules between power equipment temperature anomalies and substation operating states based on the heterogeneous temperature characteristic stream;

[0044] An analysis module, configured to perform cross-device coupling analysis on the heterogeneous temperature characteristic flows through a deep belief network to generate a dynamic conduction topology structure;

[0045] The generating module is used to combine the adjusted fault association rules with the dynamic conduction topology structure to generate a substation fault diagnosis result.

[0046] In a third aspect, the present application provides a computing device comprising a processor and a memory, wherein the memory stores a computer program, and the processor is configured to run the computer program to execute any of the big data-based intelligent analysis and diagnosis methods for substation faults described in the first aspect.

[0047] In a fourth aspect, the present application provides a computer storage medium having computer program instructions stored thereon, which, when executed by a processor, implements a big data-based intelligent analysis and diagnosis method for substation faults as described in any one of the first aspects.

[0048] In the present application, a method for intelligent analysis and diagnosis of substation faults based on big data is provided, which includes: in the substation operating environment, obtaining temperature data streams on the surface and inside of each power equipment in the substation and the ambient temperature and humidity data stream of the substation; dynamically coupling the temperature data stream and the ambient temperature and humidity data stream across time scales to generate a heterogeneous temperature feature stream; based on the heterogeneous temperature feature stream, dynamically adjusting the association rules between power equipment temperature anomalies and substation operating status; performing cross-device coupling analysis on the heterogeneous temperature feature stream through a deep belief network to generate a dynamic conduction topology structure; and generating a substation fault diagnosis result by combining the adjusted fault association rules with the dynamic conduction topology structure.

[0049] The technical solution provided by this application has the following beneficial effects:

[0050] This application establishes a multi-dimensional monitoring system covering the entire operating state of the equipment by synchronously collecting equipment surface and internal temperature data and environmental temperature and humidity parameters, significantly improving data integrity; realizes dynamic correlation modeling of environmental parameters and equipment temperature data, overcomes the traditional method's neglect of the impact of environmental factors, and enhances the reliability of temperature anomaly judgment; adaptively optimizes fault judgment rules based on real-time data, so that the diagnostic standards are automatically adjusted with operating load and environmental changes, greatly reducing misjudgments; reveals the implicit heat conduction relationship between equipment through a deep confidence network, and realizes the first visualization analysis of the topological conduction path of substation equipment faults; combines the dual verification mechanism of dynamic rules and conduction topology to significantly improve the diagnostic accuracy and early warning capabilities of complex faults.

[0051] Furthermore, the present application also constructs an initial network through multi-level feature processing, then iteratively optimizes the network structure based on time persistence, and finally generates a spatially matched dynamic conduction topology based on the physical location of the device.

[0052] In addition, it has achieved accurate modeling of the heat conduction relationship between equipment, effectively identifying chain failures caused by heat conduction, and bringing the diagnostic accuracy and warning timeliness of conduction-type failures to industry-leading levels.

[0053] These and other aspects of the present application will become more readily apparent from the description of the following embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0055] Figure 1 A flowchart of a method for intelligent analysis and diagnosis of substation faults based on big data provided in an embodiment of the present application;

[0056] Figure 2 A schematic diagram of the structure of a substation fault intelligent analysis and diagnosis system based on big data provided in an embodiment of the present application;

[0057] Figure 3 A schematic diagram of the structure of a computing device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0058] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application.

[0059] In some of the processes described in the specification and claims of this application and the above-mentioned figures, multiple operations that appear in a specific order are included, but it should be clearly understood that these operations may not be executed in the order in which they appear in this document or may be executed in parallel. The serial numbers of the operations, such as 101, 102, etc., are only used to distinguish between different operations, and the serial numbers themselves do not represent any order of execution. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel. It should be noted that the descriptions of "first", "second", etc. in this document are used to distinguish different messages, devices, modules, etc., and do not represent a sequential order, nor do they limit "first" and "second" to being different types.

[0060] Existing substation fault diagnosis solutions primarily rely on support vector machine models to perform static threshold analysis on equipment surface temperature data. These core flaws are: First, single-temperature-dimension monitoring fails to reflect the impact of ambient temperature and humidity on equipment operating conditions, resulting in anomaly determinations that are disconnected from actual operating conditions. Second, fixed threshold rules struggle to adapt to temperature variations under varying load conditions, and the model lacks the ability to model inter-device thermal conductivity, leading to a high misdiagnosis rate for conduction-type faults. The root of these issues lies in the fact that existing technologies lack the ability to analyze the dynamic correlations between multi-source data and the implicit device coupling relationships, making it difficult to meet the precise fault warning requirements of smart substations.

[0061] To address the above-mentioned shortcomings, the present invention proposes a big data-based intelligent analysis and diagnosis method for substation faults. Its core lies in generating heterogeneous temperature feature streams by dynamically coupling device surface / internal temperature with ambient temperature and humidity data, enabling multi-dimensional, real-time perception of operating status. Fault association rules are adaptively adjusted based on the feature streams, allowing diagnostic criteria to be dynamically optimized with operating conditions. Deep belief networks are further utilized to mine the conductive topological relationships between devices and establish a fault propagation path model. This method breaks through the limitations of traditional single-data analysis and static rules. Through the dual technical approaches of multi-source data fusion and implicit feature extraction, it not only solves the problem of misjudgment caused by environmental interference, but also achieves the precise location of conductive faults, significantly improving the reliability and timeliness of substation fault diagnosis.

[0062] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without making creative efforts are within the scope of protection of this application.

[0063] Figure 1 A flowchart of a method for intelligent analysis and diagnosis of substation faults based on big data provided in an embodiment of the present application is shown in FIG. Figure 1 As shown, the method includes:

[0064] Step 101: In the substation operating environment, obtain the surface and internal temperature data streams of each power equipment in the substation and the ambient temperature and humidity data stream of the substation.

[0065] In step 101, the temperature data stream represents the time series data of temperature continuously collected from the surface and interior of the power equipment, including temperature variation information of key parts of the equipment. The ambient temperature and humidity data stream represents the continuous monitoring data of air temperature and humidity collected from monitoring points within the substation.

[0066] In an embodiment of the present application, in the substation operating environment, distributed temperature sensors installed in key locations such as transformer bushings and circuit breaker contacts collect surface and internal temperature data of the equipment at a fixed sampling frequency to form a temperature data stream; at the same time, environmental monitoring devices collect temperature and humidity data in areas such as the main control room and switch room. Both types of data are synchronously cached based on timestamps to provide original input for subsequent processing.

[0067] For example, a 500kV substation deploys temperature sensors at 12 key locations, including transformer oil temperature probes and GIS (Gas Insulated Switchgear) equipment casings, to collect temperature data every second. Four temperature and humidity monitoring nodes are also deployed in the equipment area, collecting environmental data every five seconds. Both types of data are timestamped and stored in a real-time database.

[0068] Step 102: Dynamically couple the temperature data stream and the ambient temperature and humidity data stream across time scales to generate a heterogeneous temperature feature stream.

[0069] In step 102, dynamic coupling represents an environmental parameter compensation mechanism that automatically adjusts based on the device's operating status. The heterogeneous temperature feature stream represents a composite data stream that integrates features from multiple time dimensions, including second-level instantaneous fluctuations, hour-level trend changes, and monthly historical comparison features.

[0070] In an embodiment of the present application, first, the second-level time window is divided, the equipment temperature data is aligned with the ambient temperature and humidity according to the timestamp, and a data unit with an environmental compensation mark is generated; then the hourly window is divided according to the substation load change cycle, and the average temperature and fluctuation amplitude of each device are calculated as trend characteristics; finally, the historical data of the same period are correlated, and the current trend characteristics are compared with the data range of the same season and load period in previous years to generate a heterogeneous temperature feature stream containing three time dimensions.

[0071] For example, the aforementioned substation matches the oil temperature data of the last five minutes with the ambient temperature and humidity, marking it with the "high temperature and high humidity compensation" label; counts the characteristics of the oil temperature rise rate exceeding the historical same period in the last two hours; and combines the oil temperature fluctuation range during the same period last summer to generate composite feature data that currently includes environmental compensation, trend warning, and historical reference.

[0072] Step 103: Based on the heterogeneous temperature feature stream, dynamically adjust the association rules between power equipment temperature anomalies and substation operating status.

[0073] In step 103, dynamic adjustment means automatically updating the rule weight parameters according to the new feature data. The association rule represents the mapping relationship between the abnormal temperature of the equipment and the operating status parameters such as the load rate and the aging degree.

[0074] In an embodiment of the present application, temperature trend anomalies in heterogeneous feature flows are extracted, and operating parameters such as the current substation load rate and equipment operating years are matched to calculate the similarity between the anomaly pattern and historical fault cases. When a new anomaly pattern appears, the environmental temperature and humidity influencing factors are added to the original rule base, and the rule triggering threshold is updated. After each adjustment, the validity of the rule is verified through recent fault data.

[0075] For example, when it is found that the hourly rising trend of the main transformer oil temperature does not match the load, combined with the aging coefficient of the main transformer after 15 years of operation, an aging compensation item is added to the original "oil temperature-load" rule, and the warning threshold is changed from a fixed value to a dynamic value that increases with the operating years.

[0076] Step 104: Perform cross-device coupling analysis on the heterogeneous temperature characteristic flows through a deep belief network to generate a dynamic conduction topology.

[0077] In step 104, the deep belief network (DBN) is a hierarchical structure consisting of multiple feature processing layers. The bottom layer processes the temperature synchronization characteristics of physically adjacent equipment nodes, the middle layer analyzes the conduction delay characteristics of cross-regional equipment, and the upper layer integrates conduction strength and spatial constraints. This network extracts implicit heat conduction patterns between equipment groups through layer-by-layer nonlinear transformations. The inter-layer connection weights are pre-trained based on the temperature conduction characteristics of substation equipment, and the final output layer generates a conduction topology connection relationship that conforms to the actual spatial layout. Cross-device coupling analysis represents a method for identifying abnormal heat conduction paths between equipment groups. The dynamic conduction topology represents a directed graph with equipment as nodes and heat conduction strength as edge weights.

[0078] In an embodiment of the present application, the temperature characteristics of the device group are extracted layer by layer through a deep belief network. First, the temperature synchronization of physically adjacent devices is identified, and then the conduction delay characteristics of cross-regional devices are analyzed; the device pairs with conduction strength exceeding the threshold are constructed as topological edges, and the unreasonable spatial connections are eliminated based on the actual location of the devices, and finally a dynamic network reflecting the real heat conduction path is generated.

[0079] For example, analysis found that there was a 3-minute delay in temperature fluctuations between the No. 1 main transformer and the adjacent 201 switch, and the conduction strength reached the warning value. However, the actual distance between the two was within the allowable range, so a conduction edge was established in the topology. However, the equally strongly correlated No. 1 main transformer and the remote 202 switch were excluded because the distance exceeded the limit.

[0080] Step 105: combining the adjusted fault association rules with the dynamic conduction topology structure to generate a substation fault diagnosis result.

[0081] In step 105, the fault diagnosis result includes the location of the faulty device and warning information associated with the affected area.

[0082] In an embodiment of the present application, device nodes with sudden changes in conduction strength are located in a dynamic topology, and their temperature characteristics are checked to see whether they conform to the aging or environmental coupling patterns in the association rules. After confirming the source of the fault, the conduction path is traced along the topological edge to calculate the risk probability of the associated devices. The fault source characteristics and conduction path analysis results are integrated to generate a diagnostic report.

[0083] For example, after confirming the heating fault of the No. 1 main transformer bushing joint, it is found that the conduction risks of the 201 switch and the 101 knife switch decrease in sequence along the topology, and the hierarchical diagnostic conclusion "No. 1 main transformer bushing joint is abnormal (high risk), and the 201 switch (medium risk) and the 101 knife switch (low risk) need to be checked simultaneously" is output.

[0084] This method fuses multi-source data to construct a composite feature map encompassing environmental compensation, operational trends, and historical comparisons. Leveraging a dynamic rule base and collaborative analysis of conduction topology, it accurately locates substation equipment faults and predicts their associated impacts. Compared to traditional threshold alarms, this method can identify hidden conduction faults and differentiate risk levels, providing a complete chain of evidence for operational and maintenance decision-making. An example demonstrates its effectiveness in diagnosing heating in main transformer bushings and associated equipment, validating the method's engineering practicality.

[0085] To solve the problem of difficulty in identifying hidden heat conduction faults between substation devices, in some embodiments, step 104: performing cross-device coupling analysis on the heterogeneous temperature characteristic flows using a deep belief network to generate a dynamic conduction topology structure includes:

[0086] Step 201: Perform multi-level feature processing on the heterogeneous temperature feature stream through a deep belief network to construct an initial topology network.

[0087] In step 201, the initial topology network refers to a directed graph structure with power equipment as nodes and preliminary conduction relationships as edges, wherein the nodes include equipment temperature feature vectors and the edge weights represent conduction strengths that have not been spatially verified.

[0088] In an embodiment of the present application, the deep belief network first extracts the second-level temperature synchronization features in the heterogeneous feature stream to build the basic connection of the equipment, and then calculates the preliminary conduction weights based on the hourly trend similarity to form an initial network containing all potential conduction paths.

[0089] Step 202: Based on the time persistence of the heterogeneous temperature characteristic flow, the initial topology network is iteratively modified to generate a target topology network.

[0090] In step 202, the temporal persistence of the heterogeneous temperature feature stream is reflected by the temperature data variation patterns across different time dimensions within the feature stream. Specifically, these characteristics include the continuity of instantaneous temperature fluctuations within second-level time windows, the coherence of temperature trend changes within hour-level time windows, and the periodic recurrence of historical temperature features within monthly time windows. These three temporal data evolution characteristics together constitute the quantitative basis for temporal persistence. The target topology network is a conduction relationship network verified for temporal persistence. Its nodes represent power equipment, and its edges represent stable conduction paths confirmed through multiple rounds of iterative correction. Edge weights comprehensively reflect conduction strength and temporal stability, representing the optimization result of the initial network after eliminating artifacts and retaining true ones.

[0091] In an embodiment of the present application, multiple rounds of corrections are performed on the initial network: first, the conduction paths that appear repeatedly in three consecutive load cycles are screened out, then the weights are adjusted according to the trend of path strength changes, and finally, temporary connections generated by single abnormal fluctuations are eliminated to generate a target network that reflects a stable conduction pattern.

[0092] Step 203: spatially matching the target topology network with the physical location coordinate data of each power device in the substation to generate a dynamic conduction topology structure.

[0093] In step 203, the physical location coordinate data refers to the precise three-dimensional installation location of each power device within the substation. This data is derived from the equipment layout diagram in the substation design or actual coordinate data acquired through laser mapping. It includes the X / Y / Z coordinates and azimuth parameters of the device center point. Spatial matching is the process of verifying the physical plausibility of the network connection path with the actual device installation location.

[0094] In an embodiment of the present application, the actual distance between devices corresponding to each edge in the target network is calculated, the connection that meets the maximum effective heat transfer distance between substation equipment is retained, and the conduction direction weight is adjusted according to the arrangement orientation of the equipment, and finally a dynamic conduction topology that satisfies both data characteristics and spatial constraints is generated.

[0095] Here's a specific example:

[0096] In a 500kV substation implementation, the system first collects temperature data every second from temperature sensors deployed at 12 key locations, including transformer oil temperature probes and GIS equipment casings. Simultaneously, four environmental monitoring nodes collect temperature and humidity data every five seconds. Both data types are timestamped and stored in a real-time database. The system then matches the main transformer oil temperature data from the last five minutes with the ambient temperature and humidity data. When the ambient temperature exceeds 35°C and the humidity exceeds 80%, a high-temperature and high-humidity compensation flag is automatically issued. Statistics show that the oil temperature of main transformer No. 1 has risen at a rate of 2.5°C per hour over the past two hours. This value is calculated by dividing the difference between the current oil temperature and the temperature two hours ago by the time interval. This rate of increase is 39% higher than the average rate of 1.8°C per hour during the same load period during the same period last summer. Taking into account the aging factor of the main transformer, which has been in operation for 15 years, the warning threshold for the oil temperature-load association rule was adjusted from a fixed value of 85°C to a dynamic value of 85°C calculated using the formula "threshold = 80 + years of operation × 0.33." Deep confidence network analysis shows that there is a 3-minute delay in the temperature fluctuations between the No. 1 main transformer and the 201 switch. The conduction strength value, calculated by calculating the peak of the cross-correlation function of the temperature series of the two devices, is 0.78, exceeding the preset threshold of 0.7. The actual distance between the two is 3 meters, which is within the allowable range of 5 meters. Therefore, a conduction edge is established in the topology.

[0097] In the embodiment of the present application, the dynamic conduction topology constructed through multi-level feature extraction and dual spatiotemporal verification can accurately identify the actual heat conduction path between devices, avoid misjudgment caused by environmental interference or measurement errors, and provide a reliable conduction relationship evidence chain for fault tracing.

[0098] To improve the accuracy of conduction path analysis in substation fault diagnosis, in some embodiments, step 203: spatially matching the target topology network with the physical location coordinate data of each power device in the substation to generate a dynamic conduction topology structure includes:

[0099] Step 301: Obtain physical location coordinate data of each power equipment in the substation.

[0100] In an embodiment of the present application, the planar layout coordinates of major equipment such as transformers and circuit breakers are extracted from the substation design drawings, and combined with the height data collected by the on-site laser rangefinder to construct a three-dimensional coordinate data set containing the spatial position and orientation angle of the equipment.

[0101] Step 302: Compare the connection paths between nodes in the target topology network with the physical location coordinate data to screen out all initial valid conductive connection paths whose lengths are less than a preset spatial distance threshold.

[0102] In step 302, the comparison process specifically involves calculating the theoretical conduction path length between each pair of device nodes in the topological network and comparing the difference with the actual installation distance measured in the physical location coordinates. If the absolute value of the difference is less than a preset tolerance threshold, the path is retained. For example, if circuit breaker A and disconnector B in a 220kV substation appear to have a strong conductive connection in the topological network, but the actual installation distance measured in the coordinates is 8 meters (exceeding the preset 5-meter threshold), the conductive path is deemed invalid. An initial valid conductive connection path refers to a device-to-device connection path in the target topological network that simultaneously meets the following two conditions: first, the physical distance between nodes is less than a preset spatial distance threshold, and second, the conduction weight value remains above a preset conduction strength threshold after orientation correction. "Valid" refers to a conductive path that simultaneously meets three requirements: first, the topological network conduction strength meets the required standard, second, the physical spatial distance matches the actual installation location between the devices, and third, the weight value, after correction for electrical circuit relationships and heat dissipation orientation, exceeds a preset threshold. These three conditions together constitute the validity determination criteria.

[0103] In an embodiment of the present application, the theoretical conduction distance between each device node in the target topology network is calculated and compared with the actual installation distance in the physical coordinate data. The connection path whose actual distance is less than the preset maximum effective heat transfer distance is retained, and the remaining paths are marked as pending verification.

[0104] Step 303: weight correction is performed on each of the initial effective conduction connection paths according to the electrical circuit relationship and heat dissipation area orientation of each power equipment in the substation. After the correction is completed, redundant paths that do not meet the preset conduction conditions are deleted to generate a target effective conduction connection path.

[0105] In step 303, the electrical circuit relationship refers to the combination of devices within the substation that form a current path through electrical connections such as cables and busbars. This relationship is derived from the main electrical wiring diagram and device connection methods clearly marked in the substation design drawings. The heat dissipation zone orientation refers to the directional positional relationship of temperature-affected zones within the substation, which are divided based on the equipment's heat dissipation characteristics and spatial layout. This is derived from the equipment heat dissipation zoning data defined in the substation thermal design specifications. The preset conduction condition is a composite standard used to determine the effectiveness of a conduction path. It includes three dimensions: a conduction strength threshold, a lower limit for the electrical circuit enhancement coefficient, and an upper limit for the heat dissipation zone attenuation coefficient. Only paths that simultaneously meet the conduction strength threshold after electrical enhancement and heat dissipation attenuation correction, and whose physical distance and orientation meet the actual project requirements, will be retained. The target effective conduction connection path refers to the set of true and valid inter-device heat conduction relationships that have been finally confirmed after spatial distance screening, electrical circuit weight enhancement, heat dissipation zone attenuation correction, and orientation verification. Each path meets the preset conduction condition and accurately reflects the actual heat conduction characteristics between substation equipment.

[0106] In an embodiment of the present application, first, device pairs belonging to the same electrical circuit in the initial path are identified, and their conduction weights are enhanced according to the tightness of the circuit connection; then, based on the location of the heat dissipation area where the device is located, the weight of the cross-region path is attenuated according to the thermal resistance characteristics between regions; finally, redundant paths whose weights are lower than the threshold after comprehensive correction are deleted.

[0107] Step 304: Mapping the target effective conductive connection path with the actual layout orientation in the physical position coordinate data to generate a dynamic conductive topology structure.

[0108] In step 304, the actual layout orientation is the specific installation location coordinate of the equipment in the substation, and the heat dissipation area orientation is the functional area divided based on the actual layout orientation and the heat dissipation characteristics of the equipment. The two are the relationship between the basic physical location and the derived functional partition.

[0109] In an embodiment of the present application, the corrected effective path is calibrated with the actual coordinates of the device to ensure that the conduction direction is consistent with the physical orientation of the device, and connections with excessive direction deviations are eliminated, ultimately generating a conduction topology that conforms to the real space layout.

[0110] Here's a specific example:

[0111] In a specific implementation case at a 500kV substation, the system acquired the physical location coordinates of equipment such as Main Transformer 1, Switchgear 201, and Switchgear 202. The measured distance between Main Transformer 1 and Switchgear 201 was 3 meters, and the distance between Main Transformer 1 and Switchgear 202 was 8 meters. The initial conductive connection path between Main Transformer 1 and Switchgear 201 in the target topology network was compared with this coordinate data. The path, 3 meters long, was retained as the initial valid conductive connection path, as it was less than the preset spatial distance threshold of 5 meters. However, the path between Main Transformer 1 and Switchgear 202 was eliminated as it exceeded the 8-meter threshold. Based on the main electrical wiring diagram, Main Transformer 1 and Switchgear 201 were confirmed to belong to the same 220kV I bus circuit. An electrical circuit enhancement factor of 1.2 was assigned. Furthermore, since they were located in different heat dissipation zones, the heat dissipation attenuation factor was calculated as cos60° = 0.5, based on a 60-degree angle between the zones. Multiplying the initial conduction strength of 0.78 by the enhancement factor of 1.2 and then by the attenuation factor of 0.5 yields a comprehensive correction value of 0.468, which is lower than the preset conduction condition threshold of 0.7. However, given the persistence of this connection in historical data, the system activates an exception handling mechanism, temporarily lowering the threshold to 0.45 while retaining the path. Finally, the corrected valid path is matched to the actual equipment orientation. After confirming that the orientation of the No. 1 main transformer bushing aligns with the orientation of the 201 switch connector, the conduction edge is established in the dynamic conduction topology for subsequent diagnostic analysis.

[0112] In the embodiment of the present application, the dynamic conduction topology generated by spatial matching not only retains the data-driven conduction relationship, but also complies with the physical layout constraints of the substation, effectively solving the problem of false conduction paths in traditional methods and providing a more reliable topological basis for fault location.

[0113] To further improve the accuracy of the conduction path weight correction, in some embodiments, step 303: correcting the weight of each of the initial effective conduction connection paths based on the electrical circuit relationship and heat dissipation area orientation of each power equipment in the substation includes:

[0114] Step 401: Identify all device node pairs belonging to the same electrical connection group in the initial valid conductive connection paths, and perform same-loop enhancement processing on the conductive weight values ​​of the device node pairs according to a preset ratio. The electrical connection group is a collection of device groups divided according to electrical loop relationships.

[0115] In step 401, the electrical connection group is a subset of the power equipment in the substation that is directly connected by cables or busbars. It is a functional combination of the power equipment divided according to the electrical connection relationship. The two are in a part-to-whole relationship. Same-loop enhancement processing refers to the operation of weighting the conduction path between devices in the same electrical circuit. A device node pair refers to a combination of two power equipment nodes with a potential heat conduction relationship in a dynamic conduction topology. Each node corresponds to a specific power equipment, and the node pair indicates that there may be a heat transfer path between the two devices. The conduction weight value represents a quantitative indicator of the heat conduction intensity between the two devices in the node pair. It comes from the analysis results of the deep belief network on the heterogeneous temperature feature flow. The initial conduction intensity estimate is obtained by calculating the cross-correlation and trend synchronization of the temperature change series of the two devices.

[0116] In an embodiment of the present application, the connection relationship of equipment at each voltage level is first extracted from the main electrical connection diagram of the substation, and the directly electrically connected equipment is divided into connection groups. Then, the conduction path weights between the equipment in the group are multiplied by a preset enhancement coefficient to reflect the promoting effect of the electrical circuit on heat conduction.

[0117] Step 402: Calculate the conduction attenuation factor of the device node pair across the intervals based on the azimuth angle data of adjacent heat dissipation intervals in the heat dissipation area orientation.

[0118] In step 402, adjacent heat dissipation zones refer to physically adjacent sub-zones within the heat dissipation region where heat transfer is possible. The heat dissipation region is divided into several zones based on the device's heat dissipation characteristics and spatial location. Adjacent zones share common boundaries or overlap. Azimuth angle data refers to the angle between the line connecting the center points of two adjacent heat dissipation zones and the normal to the device's main heat dissipation surface. This angle is derived from the device's azimuth coordinates as shown in the substation design drawings and is calculated through spatial geometry. A cross-zone refers to a device node pair in which the two devices are located in different heat dissipation zones, and these two zones are adjacent, requiring heat transfer across the zone boundary. Cross-zone device node pairs and device node pairs in the same electrical connection group are two separate categorical concepts: the former is based on the thermodynamic division of the heat dissipation region, while the latter is based on the topological structure of the electrical connections. A pair of device nodes may possess both attributes (e.g., located in different heat dissipation zones but belonging to the same electrical circuit) or only one of the attributes (e.g., located in different zones but belonging to the same circuit). The two are cross-related through the device's physical location and electrical connection relationships, but they are not the same concept. The conduction attenuation factor refers to a parameter that reflects the degree of intensity attenuation when the heat conduction path crosses different heat dissipation areas, and is inversely proportional to the azimuth angle between the areas.

[0119] In an embodiment of the present application, the heat dissipation area is divided according to the equipment layout diagram, and the angle between the line connecting the center points of the area and the equipment orientation is calculated. The larger the angle, the smaller the attenuation factor, reflecting the hindering effect of differences in heat dissipation conditions on conduction.

[0120] Step 403: performing a product operation on the conduction weight value after the same-loop enhancement process and the conduction attenuation factor to generate a comprehensive corrected weight value.

[0121] In step 403, the comprehensive correction weight value refers to the final conduction strength evaluation value after simultaneously considering the electrical circuit enhancement and heat dissipation attenuation.

[0122] In an embodiment of the present application, the path weight that has undergone the same-loop enhancement process is multiplied by the corresponding attenuation factor to obtain a modified weight that reflects the electrical connection characteristics and conforms to the laws of thermodynamics.

[0123] Step 404: performing weight correction on each of the initial effective conductive connection paths according to the comprehensive correction weight value.

[0124] In step 404, weight modification is a process of adjusting the conduction path strength according to the comprehensive evaluation result.

[0125] In an embodiment of the present application, the corrected weight value is compared with a preset threshold, the valid path is retained and its conduction strength value is updated, and the substandard path is deleted to complete the network optimization.

[0126] Here's a specific example:

[0127] In a specific implementation case at a 500kV substation, the system first identified that the No. 1 main transformer and the 201 switch in the initial effective conduction connection path belonged to the same electrical connection group, which was determined based on the 220kV I bus main electrical wiring diagram. The system then enhanced the initial conduction strength of 0.78 using the standard enhancement factor of 1.2 for equipment at this voltage level, resulting in an intermediate value of 0.936. Next, based on the heat dissipation area layout diagram, the No. 1 main transformer was located in the main transformer heat dissipation zone, and the 201 switch was located in the switch heat dissipation zone. The angle between the centers of the two zones and the normal to the equipment heat dissipation surface was 60 degrees. The conduction attenuation factor was calculated using the formula cos60°=0.5. The enhanced conduction strength of 0.936 was multiplied by the attenuation factor of 0.5 to obtain a comprehensive correction weight of 0.468.

[0128] In the embodiment of the present application, the conduction weight is corrected by taking into account the dual factors of the electrical circuit and the heat dissipation area, which not only reflects the unique electrical connection characteristics of the substation, but also takes into account the influence of actual heat dissipation conditions, so that the generated conduction topology is more consistent with the actual heat conduction law between devices, providing a more reliable path basis for fault diagnosis.

[0129] In order to more accurately construct an initial conduction topology network between substation devices, in some embodiments, step 201: performing multi-level feature processing on the heterogeneous temperature feature stream using a deep belief network to construct an initial topology network includes:

[0130] Step 501: Using the first-level processing module of the deep belief network, based on the physical location relationship of each power equipment in the substation, the group units are divided, and the temperature fluctuation synchronization characteristics corresponding to the internal equipment of the group unit are extracted from the heterogeneous temperature feature stream.

[0131] In step 501, the physical position relationship refers to the relative position and connection relationship between devices, and the physical position coordinate data refers to the specific coordinate position data of the device within the substation. The former is used to establish the group unit, and the latter is used to verify the physical rationality of the conduction path. The group unit is an analysis unit composed of multiple electrical devices that are physically adjacent and have the possibility of heat conduction. It is in the form of a combination of devices divided according to the device installation location and electrical connection relationship, and is used to analyze local temperature conduction characteristics. Internal devices specifically refer to the subset of devices analyzed within the group unit. They are some devices with positional correlation among the various electrical devices in the substation. The two are in a containment and being contained relationship. The temperature fluctuation synchronization feature refers to the correlation characteristics of the temperature changes of devices within the group in terms of time and amplitude.

[0132] In an embodiment of the present application, devices with adjacent installation locations are first divided into groups according to the spatial layout of the substation equipment, and then the time domain correlation of the device temperature data in each group is analyzed to extract device combinations with synchronous rising / falling trends and their quantitative synchronization degree values.

[0133] Step 502: The temperature fluctuation synchronization feature is superimposed on the substation environment temperature and humidity data stream through the second-level processing module to generate an intermediate feature set.

[0134] In step 502, superposition processing refers to the process of fusing and calculating the device temperature characteristics with the environmental parameters. The intermediate feature set is a transitional data set containing the temperature conduction characteristics after environmental compensation.

[0135] In an embodiment of the present application, the extracted temperature synchronization features are aligned with the ambient temperature and humidity data in time, the temperature fluctuation amplitude is compensated and corrected according to the environmental parameters, and a temperature conduction feature set including environmental influencing factors is generated.

[0136] Step 503: The temperature-related features exceeding the preset conduction threshold are screened out from the intermediate feature set through the third-level processing module, and an initial topology network is constructed based on the conduction strength relationship between the power equipment in the substation corresponding to the temperature-related features. The temperature-related features refer to the temperature-related features between the power equipment in the substation.

[0137] In step 503, the conduction threshold is the minimum characteristic strength standard for determining whether there is a significant heat conduction relationship between devices. The temperature correlation feature refers to a characteristic indicator that reflects the possibility of heat transfer between devices. The equipment room includes both power equipment within the group unit and between different group units, including both internal equipment conduction and cross-group equipment conduction. The conduction strength relationship is a quantitative indicator of the heat transfer capability between devices. It is derived from the analysis results of the temperature feature flow by the deep belief network. It is obtained by calculating the time domain correlation, trend synchronization and amplitude matching of the device temperature series, and reflects the actual strength of heat conduction between devices.

[0138] In an embodiment of the present application, strongly associated device pairs exceeding a threshold are screened out from the intermediate feature set, network edge weights are constructed based on their association strength values, and the devices are used as nodes and the conductive relationships as edges to form an initial topological network.

[0139] Here's a specific example:

[0140] In a specific implementation case at a 500kV substation, the system first grouped devices within 3 meters of each other, such as the No. 1 main transformer, switch 201, and switch 101, into a single unit based on the equipment layout. Using a deep belief network, the system analyzed the temperature data for the last 24 hours for each device in the group. The correlation coefficient between the temperature series of the No. 1 main transformer and switch 201 was calculated to be 0.75. This value was obtained by calculating the maximum correlation coefficient of the two device temperature series at different time shifts. The 0.75 synchronization feature was then overlaid with the ambient temperature and humidity data. When the ambient temperature was 35°C and the humidity was 85%, the corrected eigenvalue of 0.82 was calculated using the compensation formula: synchronization eigenvalue = original value × [1 + 0.1 × (ambient temperature - 30°C) / 5] × [1 + 0.05 × (humidity - 80°C) / 5]. The system screened all group features for temperature-related features exceeding a preset threshold of 0.7. The value of 0.82 for the No. 1 main transformer-201 switch met the standard, while the value of 0.68 for the No. 1 main transformer-101 switch was eliminated. Based on the qualified feature values, an initial topology network was constructed with the No. 1 main transformer and the No. 201 switch as nodes and an edge weight of 0.82, which directly adopted the compensated feature values.

[0141] In the embodiment of the present application, the initial topology network constructed through multi-level feature processing not only takes into account the group characteristics of the physical location of the equipment, but also integrates environmental influencing factors. It can effectively identify the potential heat conduction relationship between substation equipment and lay the foundation for subsequent accurate diagnosis.

[0142] In order to more accurately construct a temperature characteristic model of substation equipment, in some embodiments, step 102: dynamically coupling the temperature data stream and the ambient temperature and humidity data stream across time scales to generate a heterogeneous temperature characteristic stream includes:

[0143] Step 601: Divide the temperature sampling period of each power device into a second-level time window, synchronously match the ambient temperature and humidity data with the temperature data stream within the second-level time window, and generate an instantaneous temperature data unit.

[0144] In step 601, the second-level time window refers to the minimum analysis period synchronized with the temperature sensor sampling period. The instantaneous temperature data unit is a data structure containing the instantaneous value of the device temperature after environmental compensation and its rate of change.

[0145] In an embodiment of the present application, the time window is divided based on the sampling interval of the temperature sensor. In each window, the device temperature reading is accurately matched with the ambient temperature and humidity monitoring value according to the acquisition time. The temperature value is compensated in real time according to the environmental parameters to generate a temperature data unit with an environmental tag.

[0146] Step 602: Divide the hourly time window according to the load change cycle of the substation, and perform accumulation processing on the instantaneous temperature data units to extract the temperature trend change characteristics of each power equipment within the hourly time window.

[0147] In step 602, the hourly time window refers to the analysis period that reflects the typical load fluctuation cycle of the substation. The temperature trend change characteristics include indicators such as the equipment temperature mean, extreme value, and change rate.

[0148] In an embodiment of the present application, the windows are divided according to the daily load curve of the substation, the distribution characteristics of the equipment temperature data in each window are counted, the temperature change trend slope is calculated, the abnormal fluctuation period is identified, and a trend feature set reflecting the change of the equipment operating status is formed.

[0149] Step 603: Divide the monthly time window in combination with the historical operation data of the substation, associate and map the temperature trend change characteristics with the historical temperature fluctuation range of the same period under the same seasonal conditions, and generate a temperature feature set.

[0150] In step 603, the monthly time window refers to the analysis period covering the seasonal operation cycle of the equipment. The same season means that the current analysis period has the same seasonal characteristics as the historical data (such as the high temperature period in summer / low temperature period in winter). The historical period specifically refers to the historical time period that is close to the current analysis period in calendar date (such as the same month last year ±15 days). The two together define the time range for historical data comparison. The temperature feature set refers to a composite feature data set that integrates the comparative analysis results of the current equipment temperature trend characteristics and the historical data of the same period. It contains dimensional information such as the relative position of the current temperature characteristics in the historical period, the degree of deviation, and the seasonal variation pattern. It is used to reflect the abnormality and seasonal adaptability of the equipment temperature changes in long-term operation.

[0151] In an embodiment of the present application, historical data is organized on a monthly basis, the current trend characteristics are compared with the historical data for the same period, the percentile of the current value in the historical distribution is calculated, and a temperature feature set including seasonal adaptability is generated.

[0152] Step 604: hierarchically integrate the instantaneous temperature data unit, the temperature trend change feature, and the temperature feature set according to timestamps to form a heterogeneous temperature feature stream.

[0153] In step 604, hierarchical integration refers to the process of organically combining features of different time dimensions according to a unified time axis. The hierarchical integration process is specifically as follows: the second-level instantaneous temperature data unit corresponding to the same timestamp is used as the base layer, the temperature trend change feature of the corresponding hour-level window is superimposed as the middle layer, and the historical temperature feature set of the monthly window is associated as the reference layer to form a three-dimensional nested data structure. For example, the monitoring data of a transformer at 10:30:00 on a certain day of a certain month of a certain year is integrated into: the base layer (the environmentally corrected temperature value of 42.3°C at that second), the middle layer (the temperature rise rate of 0.5°C / min during the period of 10:00-11:00), and the reference layer (the temperature reference range of 38-45°C for the same working conditions during the same period of the same month last year). The three are associated as a complete feature unit through a unified timestamp.

[0154] In an embodiment of the present application, based on the device operation timeline, second-level instantaneous values, hour-level trend features, and monthly-level historical comparison features are vertically correlated to construct a composite feature stream containing complete time dimension information.

[0155] Here's a specific example:

[0156] In a case study at a 500kV substation, the system first divided the time window into one-second intervals. The oil temperature reading of 85.3°C collected by the No. 1 main transformer's oil temperature probe was matched with the 35°C temperature and 80% humidity data recorded simultaneously by the environmental monitoring node. Using the compensation formula ("compensated temperature = original temperature + 0.2 × (ambient temperature - 30°C) - 0.1 × (humidity - 75°C)"), the compensated temperature value of 86.1°C was calculated, generating an instantaneous temperature data unit with an environmental compensation tag. The system then divided the time window into two-hour load cycles and observed the oil temperature of No. 1 main transformer rising from 82.5°C to 87.3°C within that window. The calculated rate of increase was 2.4°C per hour. This value was calculated by subtracting the temperature at the beginning of the window from the temperature at the end of the window, divided by the time interval. The system retrieved historical data from the same load period last summer and calculated the average oil temperature increase rate to be 1.9°C per hour, with a standard deviation of 0.3°C. The system determined that the current rate deviated by 1.67 standard deviations from the historical mean, generating a risk signature that included the degree of trend deviation. Finally, the features of the three dimensions, namely, the second-level compensation value of 86.1 degrees, the hour-level rising rate of 2.4 degrees / hour, and the historical deviation degree of 1.67σ, are integrated according to a unified timestamp to form a complete heterogeneous temperature feature stream.

[0157] In the embodiment of the present application, the heterogeneous temperature feature stream constructed by multi-time scale feature fusion not only captures the instantaneous fluctuations of equipment temperature, but also reflects the long-term operating rules, providing a comprehensive and three-dimensional temperature feature representation for fault diagnosis, and significantly improving the coverage of the spatiotemporal dimensions of the analysis.

[0158] In order to more accurately locate the source of the substation fault and assess the associated risk, in some embodiments, step 105: combining the adjusted fault association rules with the dynamic conduction topology to generate a substation fault diagnosis result includes:

[0159] Step 701: Mark the device nodes in the dynamic conduction topology structure whose conduction strength exceeds a preset alarm threshold as suspected fault source nodes, and extract the temperature anomaly characteristic patterns corresponding to the suspected fault source nodes.

[0160] In step 701, a suspected fault source node is a device node in the conduction topology that has abnormal conduction strength but has not yet been verified. The temperature anomaly characteristic pattern includes the temperature deviation characteristics of the node at the second, hour, and month time scales.

[0161] In an embodiment of the present application, the weight values ​​of all edges in the dynamic conduction topology are first scanned to screen out device nodes that exceed the preset warning value, and then the temperature anomaly characteristics of the node in each time dimension are extracted from the heterogeneous temperature feature stream to form a set of fault modes to be verified.

[0162] Step 702: Match and verify the temperature anomaly characteristic pattern with the adjusted fault association rules. When the matching verification result indicates that the suspected fault source node meets both the equipment aging gradient change law and the ambient temperature and humidity coupling conditions in the adjusted fault association rules, the suspected fault source node is confirmed as a confirmed fault source node.

[0163] In step 702, the matching verification process specifically involves performing a dual judgment by comparing key indicators in the temperature anomaly characteristic pattern with preset thresholds in the fault association rules. Conformity with the equipment aging gradient change pattern is demonstrated by an exponential growth trend in temperature anomaly values ​​with operating time (e.g., a circuit breaker's temperature rise rate exceeds 0.8°C / month after three years of operation). Conformity with the ambient temperature and humidity coupling condition is demonstrated by a lagged correlation between temperature fluctuations and environmental parameter changes (e.g., when the humidity is >70%, the equipment temperature exceeds the baseline value by 15% for two consecutive hours). The temperature anomaly characteristic pattern is considered to meet the conditions when both of these characteristics are present. The equipment aging gradient change pattern refers to the typical trend of temperature anomalies with increasing equipment operating age. The ambient temperature and humidity coupling condition refers to the correlation between temperature anomalies and environmental parameters. A confirmed fault source node is a suspected node that has been dually verified by the fault association rules and whose temperature anomaly characteristics meet both the equipment aging gradient change pattern and the ambient temperature and humidity coupling condition, confirming it as a truly faulty power equipment node. This serves as the core location target in the fault diagnosis results.

[0164] In an embodiment of the present application, the temperature characteristics of the suspicious node are compared with the aging pattern in the fault rule library to check whether its change trend conforms to the equipment aging curve, and at the same time verify whether the temperature anomaly is synchronized with the change of environmental parameters. Only the node that passes the double verification is confirmed as the real fault source.

[0165] Step 703: Generate risk warning information of associated devices according to the conduction path strength between the confirmed fault source node and other device nodes in the dynamic conduction topology structure.

[0166] In step 703, other devices refer to all power equipment nodes in the dynamic conduction topology structure except the confirmed fault source node, and specifically refer to power equipment in the substation that has a heat conduction correlation relationship with the fault source. The conduction path strength refers to a quantitative indicator of the heat conduction relationship between the confirmed fault source node and other nodes. It is derived from the weight value of the edge in the dynamic conduction topology structure and is obtained by analyzing the temperature correlation characteristics between devices through a deep confidence network. It reflects the attenuation degree of the fault heat conducted along the path. The conduction path strength specifically refers to the quantitative value of the conduction relationship between two specific device nodes in the dynamic conduction topology structure. The conduction strength is the specific manifestation of the conduction path strength on a single path. The two are the relationship between the overall topology attribute and the single path attribute, and are not the same concept. The associated equipment risk warning information refers to the equipment that may be affected by the fault source along the conduction path and its risk level assessment.

[0167] In an embodiment of the present application, starting from the confirmed fault source, adjacent nodes are traversed along the conductive topology edge, risk levels are divided according to the degree of conductive strength attenuation, and a multi-level warning list including directly associated devices and indirectly associated devices is generated.

[0168] Step 704: combining the location information of the confirmed fault source node with the associated equipment risk warning information to generate a substation fault diagnosis result.

[0169] In an embodiment of the present application, the location information of the confirmed fault source, the fault type and the risk list of associated equipment are combined to form a complete diagnostic report including processing priority recommendations.

[0170] Here's a specific example:

[0171] In a case study at a 500kV substation, the system first detected a conduction strength of 0.85 at the No. 1 main transformer node in the dynamic conduction topology, exceeding the preset alarm threshold of 0.8. It was then flagged as a suspected fault source. Temperature characteristics of this node were extracted, revealing that the oil temperature in the second-level window, 87°C, was reduced to 88.2°C after environmental compensation (calculated as the original temperature plus the ambient temperature compensation value of 0.2 × (35-30) minus the humidity compensation value of 0.1 × (80-75)). The hourly window oil temperature rise rate was 2.8°C / hour (calculated as the current temperature minus the temperature two hours ago, divided by the time difference). This deviation was 1.67 standard deviations from the historical mean of 1.9°C / hour, calculated using historical data statistics. The system matched these characteristics with the adjusted fault association rules and confirmed that they matched the gradient curve for bushing aging after 15 years of operation, y=0.1x+1.2, where x is the number of years in operation and changes synchronously with ambient high temperature and humidity conditions. Therefore, the No. 1 main transformer node was identified as the confirmed fault source. Based on the conduction topology analysis, the conduction path strength from the No. 1 main transformer to the 201 switch was 0.62, and to the 101 switch was 0.45. These values ​​were directly derived from the edge weights in the topological network. The final system combination generated the following diagnostic results: the No. 1 main transformer bushing joint abnormality required immediate attention, the 201 switch connector required inspection within 24 hours, and the 101 switch required observation within 72 hours.

[0172] In the embodiment of the present application, through the dual verification mechanism of conduction topology and fault rules, the accuracy of fault location is guaranteed and the quantitative assessment of associated risks is achieved, providing a complete decision-making basis for substation operation and maintenance from the source of the fault to the scope of impact, and significantly improving the systematicness and predictability of fault handling.

[0173] Figure 2 This is a schematic diagram of the structure of a substation fault intelligent analysis and diagnosis system based on big data provided by an embodiment of the present application, such as Figure 2 As shown, the system includes:

[0174] The acquisition module 21 is used to acquire the surface and internal temperature data streams of each power equipment in the substation and the ambient temperature and humidity data stream of the substation in the substation operating environment.

[0175] The coupling module 22 is used to dynamically couple the temperature data stream and the ambient temperature and humidity data stream across time scales to generate a heterogeneous temperature feature stream.

[0176] The adjustment module 23 is configured to dynamically adjust the association rules between power equipment temperature anomalies and substation operating states based on the heterogeneous temperature feature stream.

[0177] The analysis module 24 is configured to perform cross-device coupling analysis on the heterogeneous temperature characteristic flows through a deep belief network to generate a dynamic conduction topology structure.

[0178] The generating module 25 is configured to generate a substation fault diagnosis result by combining the adjusted fault association rules with the dynamic conduction topology structure.

[0179] Figure 2 The intelligent analysis and diagnosis system for substation faults based on big data can be performed Figure 1 The implementation principles and technical effects of the big data-based intelligent analysis and diagnosis method for substation faults described in the illustrated embodiment are not further elaborated. The specific manner in which each module and unit performs operations in the big data-based intelligent analysis and diagnosis system for substation faults in the above embodiment has been described in detail in the relevant embodiments of the method and will not be elaborated on here.

[0180] In one possible design, Figure 2 The embodiment shown is a big data-based intelligent analysis and diagnosis system for substation faults that can be implemented as a computing device, such as Figure 3 As shown, the computing device may include a storage component 31 and a processing component 32;

[0181] The storage component 31 stores one or more computer instructions, wherein the one or more computer instructions are called and executed by the processing component 32 .

[0182] The processing component 32 is as follows Figure 1 The embodiment provides a method for intelligent analysis and diagnosis of substation faults based on big data.

[0183] The processing component 32 may include one or more processors to execute computer instructions to complete all or part of the steps in the above method. Of course, the processing component may also be implemented as one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the above method.

[0184] The storage component 31 is configured to store various types of data to support operations on the terminal. The storage component can be implemented by any type of volatile or non-volatile memory device, or a combination thereof, such as static random-access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk, or optical disk.

[0185] Of course, a computing device may also include other components, such as input / output interfaces, display components, communication components, etc.

[0186] The input / output interface provides an interface between the processing component and the peripheral interface module, which can be an output device, an input device, etc.

[0187] The communication component is configured to facilitate, among other things, wired or wireless communications between the computing device and other devices.

[0188] Among them, the computing device can be a physical device or an elastic computing host provided by a cloud computing platform, etc. In this case, the computing device can refer to a cloud server, and the above-mentioned processing components, storage components, etc. can be basic server resources rented or purchased from the cloud computing platform.

[0189] The present application also provides a computer storage medium storing a computer program, wherein the computer program can achieve the above-mentioned Figure 1 The illustrated embodiment provides a method for intelligent analysis and diagnosis of substation faults based on big data.

[0190] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0191] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they may be located in one location or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of the present embodiment. Persons of ordinary skill in the art will be able to understand and implement the present invention without inventive effort.

[0192] Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented using software plus a necessary general-purpose hardware platform, or of course, hardware. Based on this understanding, the essence of the above technical solution, or the portion that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, or an optical disk, and includes a number of instructions for causing a computer device (such as a personal computer, server, or network device) to execute the methods described in each embodiment or certain portions of the embodiments.

[0193] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for intelligent analysis and diagnosis of substation faults based on big data, characterized in that: include: In the substation operating environment, obtain the surface and internal temperature data streams of each power equipment in the substation, as well as the ambient temperature and humidity data streams of the substation; Dynamically coupling the temperature data stream and the ambient temperature and humidity data stream across time scales to generate a heterogeneous temperature feature stream; Based on the heterogeneous temperature characteristic flow, dynamically adjust the association rules between power equipment temperature anomalies and substation operating status; Performing cross-device coupling analysis on the heterogeneous temperature characteristic flows through a deep belief network to generate a dynamic conduction topology structure; Combining the adjusted fault association rules with the dynamic conduction topology structure to generate a substation fault diagnosis result; Combining the adjusted fault association rules with the dynamic conduction topology structure to generate a substation fault diagnosis result includes: Marking a device node whose conduction strength in the dynamic conduction topology exceeds a preset alarm threshold as a suspected fault source node, and extracting a temperature anomaly characteristic pattern corresponding to the suspected fault source node; Matching and verifying the temperature anomaly characteristic pattern with the adjusted fault association rules; when the matching verification result indicates that the suspected fault source node meets both the equipment aging gradient change law and the ambient temperature and humidity coupling conditions in the adjusted fault association rules, the suspected fault source node is confirmed as a confirmed fault source node; generating risk warning information of associated devices according to the strength of the conduction paths between the confirmed fault source node and other device nodes in the dynamic conduction topology structure; Combining the location information of the confirmed fault source node with the associated equipment risk warning information to generate a substation fault diagnosis result; The cross-device coupling analysis of the heterogeneous temperature characteristic flow is performed by a deep belief network to generate a dynamic conduction topology structure, including: Performing multi-level feature processing on the heterogeneous temperature feature stream through a deep belief network to construct an initial topological network; Iteratively modifying the initial topology network based on the time persistence of the heterogeneous temperature characteristic flow to generate a target topology network; The target topology network is spatially matched with the physical location coordinate data of each power device in the substation to generate a dynamic conduction topology structure.

2. The method according to claim 1, characterized in that The step of spatially matching the target topology network with the physical location coordinate data of each power device in the substation to generate a dynamic conduction topology structure includes: Obtaining physical location coordinate data of each power equipment in the substation; Comparing the connection paths between nodes in the target topology network with the physical location coordinate data to screen out all initial valid conductive connection paths whose lengths are less than a preset spatial distance threshold; weighting each of the initial effective conduction connection paths according to the electrical circuit relationship and heat dissipation area orientation of each power equipment in the substation; after the correction is completed, deleting redundant paths that do not meet the preset conduction conditions to generate a target effective conduction connection path; The target effective conductive connection path is mapped to the actual layout orientation in the physical position coordinate data to generate a dynamic conductive topology structure.

3. The method according to claim 2, characterized in that The weight correction of each of the initial effective conductive connection paths according to the electrical circuit relationship and heat dissipation area orientation of each power equipment in the substation includes: Identifying device node pairs belonging to the same electrical connection group in all the initial valid conductive connection paths, and performing same-loop enhancement processing on the conductive weight values ​​of the device node pairs according to a preset ratio, wherein the electrical connection group is a set of device groups divided according to electrical loop relationships; Calculating the conduction attenuation factor of the device node pair across the intervals based on the azimuth angle data of adjacent heat dissipation intervals in the heat dissipation area orientation; Performing a product operation on the conduction weight value after the same-loop enhancement process and the conduction attenuation factor to generate a comprehensive modified weight value; The weight of each of the initial effective conductive connection paths is corrected according to the comprehensive correction weight value.

4. The method according to claim 1, wherein The multi-level feature processing of the heterogeneous temperature feature stream by the deep belief network to construct an initial topological network includes: Using a first-level processing module of a deep belief network, the temperature fluctuation synchronization features corresponding to the devices within the group units are extracted from the heterogeneous temperature feature stream based on the physical location relationship of the power equipment in the substation; The temperature fluctuation synchronization feature is superimposed on the substation environment temperature and humidity data stream by a second-level processing module to generate an intermediate feature set; The temperature correlation features exceeding the preset conduction threshold are screened out from the intermediate feature set through the third-level processing module, and an initial topology network is constructed based on the conduction strength relationship between the various power equipment in the substation corresponding to the temperature correlation features. The temperature correlation features refer to the temperature correlation features between the various power equipment in the substation.

5. The method according to claim 1, wherein The step of dynamically coupling the temperature data stream and the ambient temperature and humidity data stream across time scales to generate a heterogeneous temperature feature stream includes: Divide the temperature sampling period of each power device into a second-level time window, synchronously match the ambient temperature and humidity data with the temperature data stream within the second-level time window, and generate an instantaneous temperature data unit; Divide the hourly time window according to the load change cycle of the substation, and accumulate the instantaneous temperature data units to extract the temperature trend change characteristics of each power equipment within the hourly time window; Combine the substation's historical operating data to divide the monthly time window, associate and map the temperature trend change characteristics with the historical temperature fluctuation range of the same period under the same seasonal conditions, and generate a temperature feature set; The instantaneous temperature data unit, the temperature trend change feature, and the temperature feature set are hierarchically integrated according to timestamps to form a heterogeneous temperature feature stream.

6. A big data-based intelligent analysis and diagnosis system for substation faults, characterized in that: include: The acquisition module is used to obtain the surface and internal temperature data streams of each power equipment in the substation and the ambient temperature and humidity data stream of the substation in the substation operating environment; A coupling module, configured to dynamically couple the temperature data stream and the ambient temperature and humidity data stream across time scales to generate a heterogeneous temperature feature stream; An adjustment module, configured to dynamically adjust the association rules between power equipment temperature anomalies and substation operating states based on the heterogeneous temperature characteristic stream; An analysis module, configured to perform cross-device coupling analysis on the heterogeneous temperature characteristic flows through a deep belief network to generate a dynamic conduction topology structure; A generating module, configured to generate a substation fault diagnosis result by combining the adjusted fault association rules with the dynamic conduction topology structure; Combining the adjusted fault association rules with the dynamic conduction topology structure to generate a substation fault diagnosis result includes: Marking a device node whose conduction strength in the dynamic conduction topology exceeds a preset alarm threshold as a suspected fault source node, and extracting a temperature anomaly characteristic pattern corresponding to the suspected fault source node; Matching and verifying the temperature anomaly characteristic pattern with the adjusted fault association rules; when the matching verification result indicates that the suspected fault source node meets both the equipment aging gradient change law and the ambient temperature and humidity coupling conditions in the adjusted fault association rules, the suspected fault source node is confirmed as a confirmed fault source node; generating risk warning information of associated devices according to the strength of the conduction paths between the confirmed fault source node and other device nodes in the dynamic conduction topology structure; Combining the location information of the confirmed fault source node with the associated equipment risk warning information to generate a substation fault diagnosis result; The cross-device coupling analysis of the heterogeneous temperature characteristic flow is performed by a deep belief network to generate a dynamic conduction topology structure, including: Performing multi-level feature processing on the heterogeneous temperature feature stream through a deep belief network to construct an initial topological network; Iteratively modifying the initial topology network based on the time persistence of the heterogeneous temperature characteristic flow to generate a target topology network; The target topology network is spatially matched with the physical location coordinate data of each power device in the substation to generate a dynamic conduction topology structure.

7. A computing device, characterized in that It includes a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement a big data-based intelligent analysis and diagnosis method for substation faults as described in any one of claims 1 to 5.

8. A computer storage medium, characterized in that A computer program is stored, and when the computer program is executed by a computer, the method for intelligent analysis and diagnosis of substation faults based on big data as described in any one of claims 1 to 5 is implemented.

Images