An intelligent fault diagnosis method for ac-dc equipment based on multi-source signal fusion and related equipment

Abstract

This application provides a method and related equipment for intelligent fault diagnosis of AC / DC equipment based on multi-source signal fusion, belonging to the field of intelligent fault diagnosis technology. The method includes: acquiring operating signals of AC or DC equipment in a power system; for AC equipment, based on the operating signals and main wiring topology diagram, associating remote signaling signals and telemetry signals with equipment ledgers to form a signal-to-equipment mapping relationship; classifying and configuring the signals associated with each equipment in the signal-to-equipment mapping relationship according to signal type to obtain signal type information; determining whether preset topology conditions are met; weightedly fusing the judgment results of basic electrical quantities and the collaborative judgment results of multi-source external data to output AC equipment fault diagnosis data; for DC equipment, monitoring serial event recording signals to trigger interlocking or restart judgment to obtain DC equipment fault diagnosis data. This application can improve the accuracy and efficiency of AC / DC equipment fault diagnosis.

Description

Technical Field

[0001] This application relates to the field of intelligent fault diagnosis technology, and in particular to an intelligent fault diagnosis method and related equipment for AC / DC equipment based on multi-source signal fusion. Background Technology

[0002] In related technologies, AC tripping analysis technology mainly relies on text content matching of SOE signals to identify tripping events. This involves pre-setting text keywords such as "tripping," "opening," and "protection action" to perform a full search and filtering of all SOE signals across the station, supplemented by manual verification of signal content against equipment records to complete the manual verification and analysis of tripping events. DC tripping analysis technology is primarily based on alarm text content recognition of SER signals, using keyword matching to filter trip-related alarm signals for analysis. These methods are highly dependent on the experience of maintenance personnel and are difficult to meet the needs of rapid analysis.

[0003] In summary, the technical problems existing in the relevant technologies need to be improved. Summary of the Invention

[0004] The main objective of this application is to propose an intelligent fault diagnosis method and related equipment for AC / DC equipment based on multi-source signal fusion. This method can perform weighted fusion diagnosis by combining basic electrical quantity criteria with multi-source external data collaborative criteria. At the same time, it adopts an adaptive polling strategy for DC equipment to quickly determine the blocking and restart status, thereby improving the accuracy and efficiency of AC / DC equipment fault diagnosis.

[0005] To achieve the above objectives, one aspect of this application proposes an intelligent fault diagnosis method for AC / DC equipment based on multi-source signal fusion, the method comprising the following steps: Acquire operating signals from AC or DC equipment in a power system; the operating signals include remote signaling signals and telemetry signals. For AC equipment, based on the operating signals and main wiring topology diagram, the remote signaling signals and the telemetry signals are associated with the equipment ledger to form a mapping relationship between signals and equipment; Based on the preset alarm setting specifications, the signals associated with each device in the signal-device mapping relationship are classified and configured according to signal type to obtain signal type information; In response to the detection of a switch tripping signal from the operating signal, the tripping judgment function is triggered, and the status of the equipment connected to the tripping switch is retrieved based on the mapping relationship between the signal and the equipment to determine whether the preset topology conditions are met. If the preset topology conditions are met, then based on the signal type information, a dual mechanism combining basic electrical quantity criteria and multi-source external data collaborative criteria is used for fault diagnosis to obtain basic electrical quantity judgment results and multi-source external data collaborative judgment results; the basic electrical quantity criteria are formed based on switch status signals and protection action signals. The basic electrical quantity judgment result and the multi-source external data collaborative judgment result are weighted and fused for output. When the basic electrical quantity judgment result and the multi-source external data collaborative judgment result are consistent, the confidence level is increased to a preset range. When they are inconsistent, a review prompt is generated to obtain AC equipment fault diagnosis data. For DC equipment, a DC voltage parameter mapping relationship under different operating states is constructed based on the operating signal. Serial event recording signals are monitored to trigger a lockout judgment or a restart judgment to obtain DC equipment fault diagnosis data. The lockout judgment includes dynamically querying trip records based on the operating signal and monitoring the disconnector status. The restart judgment includes using an adaptive polling strategy to query the DC voltage value and match it with the DC voltage parameter mapping relationship.

[0006] In some embodiments, the remote signaling signals include protection action signals, switch status signals, disconnector status signals, and serial event recording signals; the telemetry signals include voltage signals, current signals, or DC voltage value signals. The protection action signal is the action signal issued by the relay protection device when it detects a power system fault. It is used to combine with the switch status signal to form a basic electrical quantity criterion. The protection action signal includes line protection action signal, main transformer protection action signal, bus protection action signal, traveling wave protection signal, voltage change rate protection signal, DC undervoltage protection signal, and DC differential protection signal. The switch status signal includes the switch position status signal and the switch timing action signal. The serial event recording signal is obtained through the serial event recording interface; the serial event recording signal includes an event text description field, a site name field, and a system identifier field; the serial event recording signal is used to trigger a latch-up judgment or a restart judgment.

[0007] In some embodiments, the step of associating the remote signaling signals and the telemetry signals with the equipment ledger based on the operating signals and the main wiring topology diagram for AC equipment, forming a signal-to-equipment mapping relationship, includes: For AC equipment, based on the electrical connection relationship between devices in the preset main wiring topology diagram, the protection action signal, switch position status signal, voltage signal and current signal are associated with the nearest primary equipment ledger through electrical connection points to establish the correspondence between each signal and its corresponding equipment, and obtain the signal-to-equipment mapping relationship.

[0008] In some embodiments, based on preset alarm setting specifications, the signal type classification configuration is performed on the signals associated with each device in the signal-device mapping relationship to obtain signal type information, including: Based on the preset alarm setting specifications, the protection action signal associated with each device in the signal-to-device mapping relationship is configured as an accident signal type, and the switch status signal and disconnector status signal are configured as change signal types to obtain signal type information; the signal type information is used as the basis for signal retrieval and filtering.

[0009] In some embodiments, the step of triggering a tripping judgment function in response to detecting a switch tripping signal from the operating signal, and retrieving the status of the device connected to the tripping switch based on the mapping relationship between the signal and the device, and determining whether a preset topology condition is met, includes: When any switch changes from closed to open from the switch position status signal in the operation signal, the trip judgment function is triggered. Based on the mapping relationship between the signal and the equipment, all adjacent equipment that have an electrical connection with the trip switch and meet the preset conditions are retrieved from the main wiring topology diagram. Acquire the switch position status signal, disconnector position status signal, and grounding switch position status signal of the adjacent devices; Set topology conditions; the topology conditions include the switch being in the open position, the disconnect switch being in the closed position, and the grounding switch being in the open position. If all the adjacent devices retrieved meet the topology conditions, then the preset topology conditions are satisfied; if any of the adjacent devices does not meet the topology conditions, then the analysis is stopped and the device is determined to be a faultless trip.

[0010] In some embodiments, if preset topology conditions are met, fault diagnosis is performed based on the signal type information using a dual mechanism combining basic electrical quantity criteria and multi-source external data collaborative criteria, resulting in basic electrical quantity judgment results and multi-source external data collaborative judgment results, including: Once the preset topology conditions are met, the protection action signals corresponding to the accident signal type and the switch status signals corresponding to the change signal type are filtered out based on the signal type information. The filtered protection action signals and switch status signals are analyzed using basic electrical quantity criteria. The corresponding switch timing action judgment logic or switch and protection action combination judgment logic is executed according to the equipment type to obtain the basic electrical quantity judgment result. The switch timing action judgment logic includes: based on the switch status signal, judging whether the switch has completed the open-close-open action or the open-close action within a preset time. If the open-close-open action is completed, it is judged as a permanent fault; if the open-close action is completed, it is judged as a momentary fault. The logic for determining the combination of switch and protection action includes: based on the switch status signal and the protection action signal, determining whether the switch remains in the open state after being opened and simultaneously receives the protection action signal; if so, it is determined to be a permanent fault. By accessing a multi-source external data system, data from lightning location systems, online monitoring devices, oil chromatography online monitoring data, and environmental monitoring systems are obtained. Trip events are then analyzed collaboratively to obtain collaborative judgment results from the multi-source external data. The collaborative analysis includes lightning location collaborative analysis, online monitoring collaborative analysis, oil chromatography collaborative analysis, and environmental monitoring collaborative analysis; The lightning location collaborative analysis includes: retrieving lightning strike records within a preset time period before and after the tripping time and within a preset distance range on both sides of the line corridor; if the spatial distance between the lightning strike point and the tripped line is less than a first preset threshold and the time difference is less than a second preset threshold, it is determined to be a high-confidence lightning strike fault. The online monitoring and collaborative analysis includes: acquiring monitoring data in conductor temperature, sag, insulator leakage current, capacitance value changes and unbalanced current; if the monitoring data before tripping shows an abnormal trend that meets preset conditions, it is marked as suspected equipment aging or internal fault. The oil chromatography synergistic analysis includes: obtaining the gas content of acetylene and hydrogen in the online monitoring data of transformer oil chromatography; if the gas content before tripping meets the abnormal increase in the preset conditions, it is used to help determine that it is an internal fault of the transformer. The environmental monitoring collaborative analysis includes: acquiring environmental monitoring data, and determining the correlation between the fault and environmental factors based on the environmental monitoring data; the environmental monitoring data includes temperature and humidity monitoring data, condensation monitoring data, and animal invasion monitoring data.

[0011] In some embodiments, the basic electrical quantity judgment result and the multi-source external data collaborative judgment result are weighted and fused for output. When the basic electrical quantity judgment result and the multi-source external data collaborative judgment result are consistent, the confidence level is increased to a preset range; when they are inconsistent, a review prompt is generated, and AC equipment fault diagnosis data is obtained, including: Configure a first weight value for the judgment result of the basic electrical quantity, and configure a second weight value for the collaborative judgment result of the multi-source external data; When the basic electrical quantity judgment result is consistent with the fault type indicated by the multi-source external data collaborative judgment result, the first weight value and the second weight value are added together and multiplied by a preset enhancement coefficient to obtain a confidence level within a preset high confidence range, and the fault type is output; the confidence level is used to characterize the reliability of the fault diagnosis result; When the basic electrical quantity judgment result is inconsistent with the fault type indicated by the multi-source external data collaborative judgment result, the first weight value and the second weight value are added together and multiplied by a preset weakening coefficient to obtain a preset low confidence range confidence level, generate a review prompt, and output judgment result difference information. The fault type or the difference in the judgment result is used as fault diagnosis data for AC equipment.

[0012] In some embodiments, for DC equipment, constructing a DC voltage parameter mapping relationship under different operating states based on the operating signal, monitoring serial event recording signals to trigger a lockout judgment or a restart judgment, and obtaining DC equipment fault diagnosis data includes: For DC equipment, based on the DC voltage value in the operating signal, a mapping relationship of DC voltage parameters under different operating states of the DC substation is constructed; the operating states include full voltage operating state, 80% step-down operating state, 70% step-down operating state, and lockout state; Listen to the event text description field and point group field in the serial event recording signal. When a phase-shifting blocking command or unlocking status keyword is detected, a blocking judgment is triggered. When the keywords "traveling wave protection", "voltage change rate protection", "DC undervoltage protection" or "DC differential protection" are detected, a restart judgment is triggered. The blocking determination includes: using a machine learning model to predict the best query time, dynamically querying the remote signaling historical event table to obtain trip records within the predicted time, and simultaneously collecting the dual position signals of the disconnect switches on both sides of the circuit breaker for consistency verification. When the disconnect switch is in the closed position, the DC blocking is determined to be established. The restart determination includes: adopting an adaptive polling strategy, first polling and waiting for a first preset time to query the DC voltage value, dynamically adjusting the subsequent polling interval according to the voltage recovery trend, matching the queried DC voltage value with the DC voltage parameter mapping relationship within a preset maximum number of polling times, and determining that the restart is successful if a full-voltage restart or buck-voltage restart state is matched; otherwise, the restart is determined to have failed. The result of the lockout judgment or the result of the restart judgment is output as fault diagnosis data for DC equipment.

[0013] To achieve the above objectives, another aspect of this application provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the method described above.

[0014] To achieve the above objectives, another aspect of the embodiments of this application proposes a computer-readable storage medium storing a computer program that, when executed by a processor, implements the methods described above.

[0015] To achieve the above objectives, another aspect of the embodiments of this application proposes a computer program product, including a computer program that, when executed by a processor, implements the aforementioned method.

[0016] The embodiments of this application include at least the following beneficial effects: This application provides an intelligent fault diagnosis method, electronic device, storage medium, and program product for AC / DC equipment based on multi-source signal fusion. The method includes: acquiring the operating signals of AC or DC equipment in a power system; for AC equipment, based on the operating signals and the main wiring topology diagram, associating remote signaling signals and telemetry signals with equipment ledgers to form a signal-to-equipment mapping relationship; classifying and configuring the signals associated with each equipment in the signal-to-equipment mapping relationship according to signal type to obtain signal type information; determining whether preset topology conditions are met; weightedly fusing the judgment results of basic electrical quantities and the collaborative judgment results of multi-source external data to output AC equipment fault diagnosis data; for DC equipment, monitoring serial event recording signals to trigger interlocking or restart judgment to obtain DC equipment fault diagnosis data. This application can improve the accuracy and efficiency of AC / DC equipment fault diagnosis. Attached Figure Description

[0017] Figure 1 This is a flowchart of an intelligent fault diagnosis method for AC / DC equipment based on multi-source signal fusion, provided in an embodiment of this application. Figure 2 This is a block diagram of the intelligent tripping analysis logic provided in the embodiments of this application. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of this application and are not intended to limit it. In the following description, when referring to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with those of this application; they are merely examples of apparatuses and methods consistent with some aspects of the embodiments of this application as detailed in the appended claims.

[0019] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0020] Before providing a detailed description of the embodiments of this application, some of the nouns and terms involved in the embodiments of this application will be explained first. The nouns and terms involved in the embodiments of this application are subject to the following interpretations.

[0021] 1) SOE (Sequence of Event) is a technology in power systems that records the sequence of changes in equipment status. It is used to record the occurrence time and sequence of events such as switch opening and closing, and protection actions. The time resolution is usually at the millisecond level, which is an important basis for fault analysis.

[0022] 2) SER (Serial Event Record) is event information recorded in a DC converter station through a serial interface. It includes fields such as event text description, point group, station name and system identifier, and is used to identify events such as protection actions and equipment status changes.

[0023] 3) UDL (DC voltage value) is a key telemetry parameter in DC transmission systems that characterizes the voltage of DC lines. It is used to determine the operating status of the DC system, including full-voltage operation, reduced-voltage operation (80%, 70%), and blocked status.

[0024] 4) Machine learning models are mathematical models trained on historical data that can learn patterns from data and make predictions. For example, LSTM (Recurrent Neural Network) is used in the following stages: training phase: inputting historical fault data, learning the mapping relationship between tripping type, fault type, historical time distribution, site characteristics and the best query time; prediction phase: inputting the features of the current tripping, outputting the optimal query time window (e.g., "8 seconds after tripping" instead of a fixed 10 seconds).

[0025] 5) Multi-source data fusion algorithms integrate information from different data sources (electrical quantities, lightning location, online monitoring, oil chromatography, environmental monitoring, etc.) to obtain more accurate and reliable judgment results than a single data source. Examples include weighted average method (multiplying each criterion result by a weight and then adding them together) and Bayesian network (based on the fusion of prior and posterior probabilities).

[0026] This application considers that equipment tripping in the power grid is a core abnormal event that directly affects the safe and stable operation of the power system. Once equipment such as lines or main transformers trips, it is necessary to complete the fault location of the tripping event within a short period of time to provide core decision-making basis for dispatch and emergency response. Currently, there are two main types of tripping event analysis technologies. The first is AC tripping analysis technology, which mainly relies on the text content matching of SOE signals to achieve tripping identification. That is, by preset text keywords such as "tripping," "opening," and "protection action," a full search and screening of SOE signals throughout the station is performed, supplemented by manual verification of signal content and equipment ledgers to complete the manual verification and analysis of tripping events. The second is DC tripping analysis technology, which is mainly based on the alarm text content recognition of SER signals. It completes the analysis by filtering tripping-related alarm signals through keyword matching.

[0027] The known shortcomings of the relevant technologies include: First, they do not deeply integrate topology information with signal data and equipment records, and the judgment logic still relies on signal text matching, resulting in a high false positive rate. Second, existing methods are heavily dependent on the experience of maintenance personnel, lack sufficient intelligence, and are unable to meet the needs of rapid judgment.

[0028] In view of this, this application provides an intelligent fault diagnosis method and related equipment for AC / DC equipment based on multi-source signal fusion. This scheme constructs a mapping relationship between remote signaling signals, telemetry signals and equipment ledgers in AC equipment and configures the signal types. It triggers trip judgment in response to switch opening signals. After meeting the topology conditions, it adopts a dual mechanism of weighted fusion diagnosis combining basic electrical quantity criteria and multi-source external data collaborative criteria. At the same time, it constructs a DC voltage parameter mapping relationship for DC equipment and triggers interlocking judgment or restart judgment by listening to serial event recording signals. The interlocking judgment uses machine learning to predict the query timing and monitor the disconnector status. The restart judgment uses an adaptive polling strategy to query the DC voltage value and match it with the mapping relationship, thereby realizing intelligent, accurate and rapid diagnosis of AC / DC equipment faults.

[0029] Figure 1 This is an optional flowchart of an intelligent fault diagnosis method for AC / DC equipment based on multi-source signal fusion provided in an embodiment of this application. Figure 1 The method may include, but is not limited to, steps S101 to S107.

[0030] Step S101: Obtain the operating signals of AC or DC equipment in the power system; the operating signals include remote signaling signals and telemetry signals; Step S102: For AC equipment, based on the operating signals and main wiring topology diagram, remote signaling signals and telemetry signals are associated with the equipment ledger to form a mapping relationship between signals and equipment. Step S103: Based on the preset alarm setting specifications, classify and configure the signal type of each device associated with the signal in the signal-device mapping relationship to obtain signal type information; Step S104: In response to the detection of a switch tripping signal from the operating signal, the tripping judgment function is triggered, and the status of the equipment connected to the tripping switch is retrieved based on the mapping relationship between the signal and the equipment to determine whether the preset topology conditions are met. Step S105: If the preset topology conditions are met, then based on the signal type information, a dual mechanism combining basic electrical quantity criteria and multi-source external data collaborative criteria is used for fault diagnosis to obtain the basic electrical quantity judgment result and the multi-source external data collaborative judgment result; the basic electrical quantity criteria are formed based on switch status signals and protection action signals. Step S106: The basic electrical quantity judgment result and the multi-source external data collaborative judgment result are weighted and fused for output. When the basic electrical quantity judgment result and the multi-source external data collaborative judgment result are consistent, the confidence level is increased to the preset range. When they are inconsistent, a review prompt is generated to obtain AC equipment fault diagnosis data. Step S107: For DC equipment, a DC voltage parameter mapping relationship under different operating states is constructed based on the operating signal. The serial event recording signal is monitored to trigger a lockout judgment or a restart judgment to obtain DC equipment fault diagnosis data. The lockout judgment includes dynamically querying trip records based on the operating signal and monitoring the status of disconnectors. The restart judgment includes using an adaptive polling strategy to query the DC voltage value and match it with the DC voltage parameter mapping relationship.

[0031] Steps S101 to S107, as illustrated in this embodiment, involve acquiring operating signals, mapping remote signaling signals, telemetry signals, and equipment ledgers based on the main wiring topology diagram, and classifying and configuring signals according to alarm setting specifications. This achieves precise association between signals and equipment and standardized classification of signal types, laying a data foundation for subsequent efficient retrieval and screening based on signal type and accurate fault diagnosis. It avoids blind matching of original signals and improves the standardization of signal processing and retrieval efficiency. The main wiring topology diagram is constructed during the construction of the dispatch automation master station. The dispatch automation master station is the core technology platform of the power system dispatch control center, used to realize real-time monitoring, operation control, fault handling, and dispatch management of the power grid. It is responsible for collecting, processing, analyzing, and displaying power grid operation data and supporting dispatchers in making decisions on power grid operation control and fault handling. This application's embodiments employ a dual mechanism combining basic electrical quantity criteria and multi-source external data collaborative criteria. It accurately distinguishes between transient and permanent faults through switching timing actions, and utilizes multi-source data such as lightning location, online monitoring, oil chromatography, and environmental monitoring to assist in verifying fault causes, significantly improving fault type identification capabilities. By weighted fusion of the two criteria results, it automatically outputs increased confidence when consistent and generates verification prompts when inconsistent, achieving an optimal balance between automatic diagnosis and manual intervention. Furthermore, by constructing voltage parameter mapping relationships for DC equipment, employing machine learning to predict query timing, and using an adaptive polling strategy, it achieves rapid and accurate judgment of DC blocking and restart states. This application's embodiments effectively solve the technical problems of traditional fault diagnosis methods relying on a single signal source, difficulty in distinguishing fault types, and low efficiency in judging DC equipment, providing high-accuracy, high-efficiency, and high-reliability technical support for intelligent operation and maintenance of power systems.

[0032] In some embodiments, the remote signaling signals in step S101 include protection action signals, switch status signals, disconnector status signals, and serial event recording signals; the telemetry signals include voltage signals, current signals, or DC voltage value signals. Protection action signals are the action signals issued by relay protection devices when they detect power system faults. They are used in combination with switch status signals to form basic electrical quantity criteria. Protection action signals include line protection action signals, main transformer protection action signals, bus protection action signals, traveling wave protection signals, voltage change rate protection signals, DC undervoltage protection signals, and DC differential protection signals. Switch status signals include switch position status signals and switch timing action signals. The serial event recording signal is obtained through the serial event recording interface; the serial event recording signal includes an event text description field, a site name field, and a system identifier field; the serial event recording signal is used to trigger a latch-up judgment or a restart judgment.

[0033] This application's embodiments enable real-time acquisition of the basic data required for fault diagnosis, providing a signal source for subsequent analysis, allowing diagnosis to be independent of a single signal, and avoiding misjudgments due to missing data. For example: When a line trip occurs in a 220kV substation, the following operating signals are acquired in real time: Remote signaling signals: line protection action signal (issued by the protection device), circuit breaker tripping signal (the switch changes from the closed position to the open position), disconnector position signal (the disconnector is in the closed position); Telemetry signals: The line voltage dropped sharply from 220kV to 0kV, and the line current dropped from 500A to 0A.

[0034] In some embodiments, step S102 may include, but is not limited to, step S201: Step S201: For AC equipment, based on the electrical connection relationship between equipment in the preset main wiring topology diagram, the protection action signal, switch position status signal, voltage signal and current signal are associated with the nearest primary equipment ledger through electrical connection points to establish the correspondence between each signal and its corresponding equipment, and obtain the signal-equipment mapping relationship.

[0035] In step S201 of some embodiments, the correspondence between each signal and its corresponding device is established, thereby solving the problem of unclear signal attribution in traditional methods and providing a basis for subsequent fault diagnosis and topology retrieval based on device dimensions. For example, based on the topology information that "circuit breaker QF101 and line L101 have an electrical connection" in the main wiring topology diagram, the following signals are associated with the device ledger: the protection action signal "line protection action" is associated with line L101; the switch tripping signal "QF101 tripping" is associated with circuit breaker QF101; and the voltage and current signals are associated with line L101.

[0036] In some embodiments, step S103 may include, but is not limited to, step S301: Step S301: Based on the preset alarm setting specifications, configure the protection action signal associated with each device in the signal-to-device mapping relationship as an accident signal type, and configure the switch status signal and disconnector status signal as a change signal type to obtain signal type information; the signal type information is used as the basis for signal retrieval and filtering.

[0037] In step S301 of some embodiments, standardized signal classification is achieved by configuring signal types. Subsequent fault diagnosis can then directly perform rapid retrieval and filtering based on signal type (fault signal, position change signal), avoiding the inefficiency and inaccuracy of fuzzy matching based on signal names in traditional methods. For example, based on preset alarm setting specifications, the associated signals are classified and configured as follows: the "line protection action" signal is configured as a "fault signal" type; the "QF101 trip" signal is configured as a "position change signal" type; and the "disconnector position" signal is configured as a "position change signal" type.

[0038] In some embodiments, step S104 may include, but is not limited to, steps S401 to S404: Step S401: When any switch changes from closed to open from the switch position status signal in the operation signal, the trip judgment function is triggered. Based on the mapping relationship between the signal and the equipment, all adjacent equipment that have an electrical connection with the trip switch and meet the preset conditions are retrieved from the main wiring topology diagram. Step S402: Obtain the switch position status signal, disconnector position status signal, and grounding switch position status signal of adjacent devices; Step S403: Set topology conditions; topology conditions include the switch being in the open position, the disconnect switch being in the closed position, and the grounding switch being in the open position. Step S404: If all the adjacent devices retrieved meet the topology conditions, then it is determined that the preset topology conditions are met; if any adjacent device does not meet the topology conditions, then the analysis is stopped and it is determined to be a faultless trip.

[0039] In steps S401 to S404 of some embodiments, non-faulty tripping (such as normal switch opening by operators or tripping operations during maintenance) is effectively filtered through topology condition verification, avoiding misdiagnosis of non-faulty events, reducing false alarm rate, and saving maintenance personnel unnecessary on-site confirmation time. For example: when the system detects that circuit breaker QF101 changes from closed to open, it triggers the tripping judgment function: first, it searches for adjacent devices connected to QF101 based on the mapping relationship: line L101, disconnector G101, and grounding switch E101; then it obtains the status: circuit breaker QF101 is in the open position, disconnector G101 is in the closed position, and grounding switch E101 is in the open position; knowing that all adjacent devices meet the topology condition of "switch open, disconnector closed, and grounding switch open", it determines that the condition is met and continues the diagnosis. If disconnector G101 is in the open position (does not meet the topology condition), the analysis stops and it is determined to be a faultless tripping, avoiding misdiagnosis of normal switch opening and closing by operators.

[0040] In some embodiments, step S105 may include, but is not limited to, steps S501 to S502: Step S501: When it is determined that the preset topology conditions are met, the protection action signal corresponding to the fault signal type and the switch status signal corresponding to the change signal type are filtered out based on the signal type information. The basic electrical quantity criteria are used to analyze the filtered protection action signal and switch status signal. According to the equipment type, the corresponding switch timing action judgment logic or switch and protection action combination judgment logic is executed to obtain the basic electrical quantity judgment result. The switch timing action judgment logic includes: based on the switch status signal, judging whether the switch has completed the open-close-open action or the open-close action within a preset time. If the open-close-open action is completed, it is judged as a permanent fault; if the open-close action is completed, it is judged as a momentary fault. The logic for judging the combination of switch and protection action includes: based on the switch status signal and the protection action signal, judging whether the switch remains in the open state after being opened and simultaneously receives the protection action signal; if so, it is judged as a permanent fault. Step S502: Connect to the multi-source external data system to obtain data from the lightning location system, online monitoring device, oil chromatography online monitoring, and environmental monitoring system; perform collaborative analysis on the tripping event to obtain the collaborative judgment result of the multi-source external data. Collaborative analysis includes lightning location collaborative analysis, online monitoring collaborative analysis, oil chromatography collaborative analysis, and environmental monitoring collaborative analysis; The lightning location collaborative analysis includes: retrieving lightning strike records within a preset time period before and after the tripping time and within a preset distance range on both sides of the line corridor. If the spatial distance between the lightning strike point and the tripped line is less than the first preset threshold and the time difference is less than the second preset threshold, it is determined to be a high-confidence lightning strike fault. Online monitoring and collaborative analysis includes: acquiring monitoring data on conductor temperature, sag, insulator leakage current, capacitance value changes, and unbalanced current. If the monitoring data before tripping shows an abnormal trend that meets preset conditions, it is marked as suspected equipment aging or internal fault. Oil chromatography synergistic analysis includes: obtaining the gas content of acetylene and hydrogen in the online monitoring data of transformer oil chromatography; if the gas content before tripping meets the abnormal increase in the preset conditions, it can be used to help determine that it is an internal fault of the transformer. Collaborative analysis of environmental monitoring includes: acquiring environmental monitoring data and determining the correlation between faults and environmental factors based on the environmental monitoring data; environmental monitoring data includes temperature and humidity monitoring data, condensation monitoring data, and animal invasion monitoring data.

[0041] In steps S501 to S502 of some embodiments, the complementary integration of single electrical quantity criteria and multi-source external data collaborative criteria is achieved through basic electrical quantity criteria and multi-source external data collaborative criteria. This not only distinguishes between transient and permanent faults through switch timing actions, but also assists in verifying the cause of the fault through multi-source data such as lightning location, oil chromatography, and online monitoring, thereby improving the accuracy of fault diagnosis and the ability to identify fault types. This effectively enhances the identification accuracy of complex faults such as lightning strike faults, equipment aging faults, and transformer internal faults. For example, based on the basic electrical quantity criteria, the switch timing action is judged as follows: if the circuit breaker QF101 completes the "open-close-open" action within 15 seconds, it is determined to be a permanent fault; the combination of switch and protection actions is judged as follows: if the circuit breaker QF101 remains in the open state after opening and receives a line protection action signal, it is determined to be a permanent fault. Based on multi-source external data collaborative judgment criteria, lightning location collaborative analysis: retrieval of lightning strike records within 5 minutes before and after the tripping time and within a 5-kilometer radius on both sides of the line corridor; if the lightning strike point is 300 meters from the line and the time difference is 20 seconds, it is judged as a high-confidence lightning strike fault; online monitoring collaborative analysis: if the conductor temperature rises by 30% continuously within 3 days before the tripping time, it is marked as a suspected equipment aging fault; oil chromatography collaborative analysis: if the acetylene content in the transformer rises from 0.5 μL / L to 15 μL / L, it is used to assist in judging an internal transformer fault.

[0042] In some embodiments, step S106 may include, but is not limited to, steps S601 to S603: Step S601: Configure the first weight value for the basic electrical quantity judgment result and configure the second weight value for the multi-source external data collaborative judgment result; Step S602: When the basic electrical quantity judgment result is consistent with the fault type indicated by the multi-source external data collaborative judgment result, the first weight value and the second weight value are added together and multiplied by a preset enhancement coefficient to obtain the confidence level of the preset high confidence range, and the fault type is output; the confidence level is used to characterize the reliability of the fault diagnosis result. Step S603: When the fault type indicated by the judgment result of the basic electrical quantity is inconsistent with the judgment result of the multi-source external data collaboration, the first weight value and the second weight value are added together and multiplied by the preset weakening coefficient to obtain the confidence level of the preset low confidence level range, a review prompt is generated, and the judgment result difference information is output. Step S604: Use the fault type or judgment result difference information as fault diagnosis data for AC equipment.

[0043] In steps S601 to S604 of some embodiments, a quantitative confidence assessment and intelligent review mechanism for the diagnostic results are performed. When the results of the two criteria are consistent, a high-confidence diagnostic result is automatically output, reducing manual intervention. When the results are inconsistent, a review prompt is generated and the difference information is output, guiding maintenance personnel to focus on the points of disagreement. This not only improves diagnostic efficiency but also avoids the risk of misjudgment based on a single criterion, achieving the optimal balance between automatic diagnosis and manual review. For example: If the basic electrical quantity judgment result is a permanent fault (weight 0.6), and the multi-source collaborative judgment result is a lightning strike fault (weight 0.4), both point to a permanent fault, the judgments are consistent, and the fusion confidence level = (0.6 + 0.4) × enhancement coefficient 1.2 = 100%, the system will automatically output "Permanent fault, high confidence lightning strike fault, confidence level 100%"; If the basic electrical quantity judgment result is a permanent fault (weight 0.6), and the multi-source collaborative judgment result is no abnormality (weight 0.4), the judgments are inconsistent, and the fusion confidence level = (0.6 + 0.4) × weakening coefficient 0.5 = 50%, then a review prompt will be generated: "The basic electrical quantity is judged as a permanent fault, and the multi-source collaborative judgment is judged as no abnormality. Please review."

[0044] In some embodiments, step S107 may include, but is not limited to, steps S701 to S704: Step S701: For DC equipment, based on the DC voltage value in the operating signal, construct the DC voltage parameter mapping relationship of the DC substation under different operating states; the operating states include full voltage operating state, 80% voltage reduction operating state, 70% voltage reduction operating state, and lockout state; Step S702: Listen to the event text description field and point group field in the serial event recording signal. When a phase-shifting blocking command or unlocking status keyword is detected, a blocking judgment is triggered. Step S703: When the keywords of traveling wave protection, voltage change rate protection, DC undervoltage protection or DC differential protection are detected, a restart judgment is triggered. The blocking determination includes: using a machine learning model to predict the best query time, dynamically querying the remote signaling historical event table to obtain trip records within the predicted time, and simultaneously collecting the dual position signals of the disconnectors on both sides of the circuit breaker for consistency verification. When the disconnectors are in the closed position, the DC blocking is determined to be established. The restart determination includes: adopting an adaptive polling strategy, first polling waits for a first preset time to query the DC voltage value, dynamically adjusting the subsequent polling interval according to the voltage recovery trend, matching the queried DC voltage value with the DC voltage parameter mapping relationship within the preset maximum number of polling times, and determining that the restart is successful if a full-voltage restart or buck-voltage restart state is matched; otherwise, the restart is determined to have failed. Step S704: Output the result of the lockout judgment or the restart judgment as fault diagnosis data for DC equipment.

[0045] In steps S701 to S704 of some embodiments, rapid intelligent judgment of the DC equipment's lockout and restart status is performed. Machine learning optimizes the query timing instead of a fixed waiting time, improving query efficiency. An adaptive polling strategy dynamically adjusts the query interval based on voltage recovery trends, terminating polling early when the voltage continues to drop, reducing invalid queries. Dual-position signal consistency verification avoids misjudgments caused by signal anomalies, thereby achieving rapid, accurate, and intelligent fault diagnosis of DC equipment. For example, the lockout judgment includes: SER signal monitoring detects the keyword "phase-shift lockout command appears," triggering the lockout judgment; machine learning predicts the query timing: the model outputs the optimal query window as 8 seconds after tripping (replacing a fixed 10-second wait); dynamic query of tripping records: querying the tripping record of the converter transformer's AC incoming circuit breaker; disconnector status monitoring: collecting dual-position signals of the disconnectors on both sides of the circuit breaker (closed signal is 1, open signal is 0), and if the consistency verification passes and the disconnector is in the closed position, the judgment result is: DC lockout is established, and pole 1 is locked out. The restart judgment includes: the SER signal listener detects the keyword "WFPDL" (traveling wave protection), triggering the restart judgment; the adaptive polling strategy includes: the first polling (waiting for 2 seconds): the UDL value is 50kV (500kV before the fault, a drop of 90%), the voltage recovery trend: the UDL value increases by 35% compared to the fault time (exceeding the 30% threshold), and the polling interval is shortened to 2 seconds; the second polling: the UDL value rises to 400kV (matching the 80% step-down state), and the judgment result: the step-down restart is successful.

[0046] refer to Figure 2 The intelligent tripping analysis logic in this application includes AC tripping logic and DC tripping logic.

[0047] The AC tripping logic in this embodiment includes: Step 1, Data Modeling: Step 1.1, Associate signals and devices: Based on the main wiring topology diagram, remote signaling signals such as protection actions and switch positions, as well as telemetry signals such as voltage and current, are associated with the nearest equipment ledger through electrical connections to form a mapping relationship.

[0048] Step 1.2, Configure signal type: According to the preset alarm setting specifications, the signals associated with each device are classified and configured by signal type (the protection action signal type is the fault signal, and the switch position status signal type is the change signal). The signal retrieval and filtering are mainly based on the signal type of each device.

[0049] Step 2, Diagnosing AC equipment tripping faults: Step 2.1, Topological Search: When the system receives a switch tripping signal, it will automatically trigger the tripping judgment function. It will automatically search the status of all devices connected to the tripping switch on the main wiring topology diagram in step 1.1. If the conditions of switch tripping, disconnector closing, and grounding disconnector opening are met, then step 2.2 will be executed. Otherwise, the judgment will stop and the system will be considered to have no fault.

[0050] Step 2.2, Fault Type Determination: A dual mechanism of basic electrical quantity judgment and multi-source data collaborative auxiliary judgment is adopted. After the basic judgment conditions are met, auxiliary verification is performed through external system data. If the 220kV line (busbar double-section connection) trips, step 2.3 is executed; if the 500kV line (3 / 2 connection method) trips, step 2.4 is executed; if the transformer fault judgment logic is executed, step 2.5 is executed; if the busbar fault trips, step 2.6 is executed; if the capacitor reactor (filter) trips, step 2.7 is executed.

[0051] Step 2.3.1, Basic Electrical Quantity Judgment Logic: Switch timing action judgment logic: If a switch connected to the line completes the open-close-open action within 15 seconds, it is judged as a permanent fault; if a switch connected to the line completes the open-close action within 15 seconds, it is judged as a momentary fault (reclosing successful).

[0052] Switch + Protection Action Judgment Logic: If the switch opens once and remains in the open state, and the line protection activates, it is judged as a permanent fault. (This logic can be used to determine if the line is not configured for reclosing).

[0053] Step 2.3.2, Cross-system collaborative auxiliary judgment logic: Meteorological data collaboration: By accessing lightning location system data, when a line trips, the system searches for lightning strike records within 5 minutes before and after the trip, and within a 5-kilometer radius on both sides of the line corridor. If a lightning strike record is found, it is marked as a "suspected lightning strike fault," helping to distinguish between transient lightning strike faults and permanent faults.

[0054] Collaborative equipment status monitoring: Integrates data from online monitoring devices, including conductor temperature, sag, and insulator leakage current. If abnormal trends are observed in the monitoring data before tripping, it is marked as "suspected equipment aging fault".

[0055] Intelligent fusion judgment: The judgment results of basic electrical quantities are weighted and fused with the judgment results of external systems. When the two judgments are consistent, the confidence level is increased; when the two judgments are inconsistent, a manual review prompt is generated.

[0056] Step 2.4, Fault Trip Judgment Logic for 500kV Lines (3 / 2 Connection): Step 2.4.1, Basic Electrical Quantity Judgment Logic: Switch timing action judgment logic: If two switches connected to the line complete the open-close-open action simultaneously within 15 seconds, it is judged as a permanent fault; if two switches connected to the line complete the open-close action simultaneously within 15 seconds, it is judged as a momentary fault (successful reclosing); if two switches connected to the line open once simultaneously and remain in the open state, it is judged as a permanent fault.

[0057] Switch + Protection Action Judgment Logic: If a single switch connected to the line trips once and remains in the tripped state, and the line protection activates, it is judged as a permanent fault.

[0058] Step 2.4.2, Cross-system collaborative auxiliary judgment logic: Meteorological data collaboration: By accessing lightning location system data and meteorological early warning information, the following specific methods are used to assist in judgment when a line trips: Retrieve lightning strike location data within 5 minutes before and after the tripping time, and within a 5-kilometer radius on both sides of the line corridor; If the spatial distance between the lightning strike point and the tripped line is less than 500 meters and the time difference is less than 30 seconds, it is judged as a high-confidence lightning strike fault. If severe convective weather is present in the tripped area, the weighting of lightning strike fault assessment should be increased by combining meteorological radar data.

[0059] Equipment status monitoring collaboration: Connect to the online monitoring system for transmission lines, including data from micro-meteorological monitoring, conductor galloping monitoring, and insulator pollution monitoring, to help determine the cause of faults.

[0060] Intelligent fusion judgment: The algorithm uses multi-source data fusion to comprehensively evaluate electrical quantity judgment, meteorological data and equipment status data, and outputs the fault type and confidence level.

[0061] Step 2.5, Transformer fault trip judgment logic: Step 2.5.1, Basic Electrical Quantity Judgment Logic: Simultaneous tripping of transformer switches (number of switches greater than or equal to 2) + protection action signal: When two or more switches connected to the transformer trip once at the same time and remain in the tripped state, they simultaneously receive the main transformer protection action signal, indicating a transformer fault trip.

[0062] Step 2.5.2, Cross-system collaborative auxiliary judgment logic: Collaborative oil chromatography monitoring: By accessing online oil chromatography monitoring data of transformers, if there is an abnormal increase in the content of characteristic gases such as acetylene and hydrogen before the trip, it can help determine that the fault is internal to the transformer.

[0063] Temperature monitoring collaboration: By accessing the top oil temperature and winding temperature monitoring data of the transformer, if there is an abnormal upward trend in temperature before tripping, it can help determine the type of fault.

[0064] Intelligent fusion judgment: Correlate and analyze electrical protection action signals with status monitoring data such as oil chromatography and temperature to improve the accuracy of fault diagnosis.

[0065] Step 2.6, Busbar Fault Trip Judgment Logic: Step 2.6.1, Basic Electrical Quantity Judgment Logic: Busbar connected switches trip (number of switches greater than or equal to 2) + protection action signal + voltage judgment logic: all switches connected to the busbar trip once at the same time and remain in the tripped state while receiving relevant busbar protection, and the busbar voltage is greater than 0 before the switches trip.

[0066] Step 2.6.2, Cross-system collaborative auxiliary judgment logic: Environmental factor coordination: Integrate with the substation environmental monitoring system, including data such as temperature, humidity, condensation, and small animal intrusion monitoring, to help determine whether busbar faults are related to environmental factors.

[0067] Intelligent integrated judgment: Combines electrical quantity judgment with environmental monitoring data to output fault type and possible causes.

[0068] Step 2.7, Fault Trip Judgment Logic for Capacitor-Reactor (Filter): Step 2.7.1, Basic Electrical Quantity Judgment Logic: If the switch connected to the capacitor reactor only opens once and remains in the open state, plus the protection action signal associated with the capacitor reactor, and the bus voltage connected to the switch is greater than 0 before the switch opens, it is determined that the capacitor has tripped due to a fault.

[0069] Step 2.7.2, Cross-system collaborative auxiliary judgment logic: Collaborative equipment status monitoring: Access online monitoring data of capacitor banks, including capacitance changes, unbalanced current, temperature, etc., to help determine the type of fault.

[0070] Intelligent fusion judgment: Correlate and analyze protection action signals with equipment status monitoring data to improve the accuracy of fault diagnosis.

[0071] The DC tripping logic in this application embodiment includes: Step 1, Data Modeling: Create data table a: store information such as the name of the AC incoming circuit breaker corresponding to pole 1 and pole 2 of each DC substation.

[0072] Create data table b: Store the data range of each site's UDL under full voltage, 80% voltage reduction, 70% voltage reduction, and lockout states.

[0073] Step 2: Determine if DC is blocked: Step 2.1, Listen for the SER signal: Listen to the EventText field. If it contains "Phase shift blocking command appeared" or "STATUS OF OPERATION DEBLOCKED", then trigger step 2.2. At the same time, obtain the tripped site name based on the strStName field of the SER signal and obtain the SER signal generation time based on the activeTime field.

[0074] Step 2.2: Query trip records: Dynamic query timing calculation: Instead of a fixed 10-second wait, a machine learning model is used to predict the optimal query timing. Model input features include: Trip type (AC side / DC side); Fault type (short circuit / overload / protection action); The time distribution of tripping records for similar historical faults; Site equipment characteristic parameters.

[0075] The model output is the optimal query time window. The system dynamically adjusts the query timing based on the prediction results. Within the prediction time window, it queries the remote signaling historical event table to obtain all tripping records of AC incoming circuit breakers of converter transformers at the current site and the opposite site. If a tripping record is found, it is determined that DC blocking has occurred. Based on the name of the tripped circuit breaker, it refers to data table a to determine which pole or multiple poles were blocked.

[0076] Step 2.3: Monitor the status of the disconnect switch: Disconnector status monitoring method: Collect the dual position signals (closed / open) of the disconnectors on both sides of the circuit breaker.

[0077] Mechanism for handling inconsistencies between dual-position signals: 1. When the combined signal and the separated signal are both 1 or both 0, it is determined to be a signal abnormality; 2. Start the signal verification program, continuously acquire signals 3 times, and take the majority consistent result; 3. If the three data collections are still inconsistent, an alarm will be triggered and the most recent valid status will be used as the basis for judgment; 4. Record abnormal statuses and report them to the operation and maintenance system.

[0078] If the disconnect switch is in the closed position, then the DC blocking is considered to be successful.

[0079] Step 3: Determine the tripped and restarted status: Step 3.1, Listen for the SER signal: Listen to the EventText or PointGroup field. If it contains "WFPDL, 27du / dt, 27DCL, 87DCLL", then trigger step 3.2. Simultaneously, obtain the tripped site based on the strStName field of the SER signal, and determine whether the tripped line is pole 1 or pole 2 based on the SystemID field. (EventText: Event text description, containing protection action type information; PointGroup: Point group, identifying the protection function category; strStName: Site name field; SystemID: System identifier, distinguishing pole 1 / pole 2) Keyword matching rules: A combination of fuzzy matching and exact matching is used; Keyword list: "WFPDL" (traveling wave protection), "27du / dt" (voltage change rate protection), "27DCL" (DC undervoltage protection), "87DCLL" (DC differential protection); Matching priority: exact match > containment match > regular expression match; If the EventText field or PointGroup field contains any of the keywords mentioned above, then step 3.2 is triggered.

[0080] At the same time, the tripping site is obtained based on the strStName field of the SER signal, and the tripping line is determined as pole 1 or pole 2 based on the SystemID field.

[0081] Step 3.2: Determine the reboot status using UDL: Adaptive polling strategy: Initial polling: The first UDL query will be performed after a 2-second wait. Dynamic interval adjustment: 1. If the initial query shows a voltage recovery trend (UDL value increases by ≥30% compared to the fault time), the subsequent polling interval is shortened to 2 seconds; 2. If the first query does not show obvious signs of voltage recovery (UDL value change <10%), the subsequent polling interval will be extended to 4 seconds; 3. If the UDL value continues to decrease after two consecutive queries, the restart is deemed to have failed, and the polling is terminated prematurely.

[0082] Maximum number of polls: 5; Termination conditions: 1. If the system detects a full-pressure restart or a reduced-pressure restart, stop the process and determine that the restart was successful; All 2.5 calls were judged as restart failures, and the final result was determined to be a restart failure. 3. Stop immediately when the early termination condition is triggered.

[0083] Based on UDL reference data table b, determine whether the line restarted at full voltage, restarted at reduced voltage, or failed to restart.

[0084] Deduplication mechanism: After the line protection of the same line triggers step 3.2, step 2 will not be triggered again within 2 seconds.

[0085] The advantages of this application's embodiments include: Firstly, regarding the tripping of AC equipment, this application does not simply match the content in signal retrieval, but filters based on the pre-marked signal type attributes, combined with logical criteria such as "switch position + protection action", which effectively improves the accuracy of judgment.

[0086] Secondly, in response to DC equipment tripping, this application enables rapid judgment of DC blocking by presetting key parameter ranges under different operating conditions and monitoring SER signals and real-time UDL values.

[0087] This application also provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the above-described method.

[0088] It is understood that the content of the above method embodiments is applicable to this device embodiment. The specific functions implemented by this device embodiment are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0089] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method.

[0090] It is understood that the content of the above method embodiments is applicable to this storage medium embodiment. The specific functions implemented in this storage medium embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments.

[0091] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method.

[0092] It is understood that the content of the above method embodiments is applicable to the embodiments of this program product. The specific functions implemented by the embodiments of this program product are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0093] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0094] This application provides an intelligent fault diagnosis method, electronic device, storage medium, and program product for AC / DC equipment based on multi-source signal fusion. It constructs a mapping relationship between remote signaling signals, telemetry signals, and equipment ledgers in AC equipment and configures the signal types. In response to a switch tripping signal, it triggers a tripping judgment. After meeting topology conditions, it employs a dual mechanism combining basic electrical quantity criteria and multi-source external data collaborative criteria for weighted fusion diagnosis. Simultaneously, it constructs a DC voltage parameter mapping relationship for DC equipment and triggers a blocking or restart judgment by listening to serial event recording signals. The blocking judgment uses machine learning to predict query timing and monitor the disconnector status, while the restart judgment uses an adaptive polling strategy to query the DC voltage value and match it with the mapping relationship. This achieves intelligent, accurate, and rapid fault diagnosis for AC / DC equipment.

[0095] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided by the embodiments of this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems.

[0096] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this application, and may include more or fewer steps than shown, or combine certain steps, or different steps.

[0097] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0098] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.

[0099] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0100] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0101] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of the units described above is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0102] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0103] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0104] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0105] The preferred embodiments of the present application have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present application. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and substance of the embodiments of the present application shall be within the scope of the claims of the present application.

Claims

1. A method for intelligent fault diagnosis of AC / DC equipment based on multi-source signal fusion, characterized in that, The method includes the following steps: Acquire operating signals from AC or DC equipment in a power system; the operating signals include remote signaling signals and telemetry signals. For AC equipment, based on the operating signals and main wiring topology diagram, the remote signaling signals and the telemetry signals are associated with the equipment ledger to form a mapping relationship between signals and equipment; Based on the preset alarm setting specifications, the signals associated with each device in the signal-device mapping relationship are classified and configured according to signal type to obtain signal type information; In response to the detection of a switch tripping signal from the operating signal, the tripping judgment function is triggered, and the status of the equipment connected to the tripping switch is retrieved based on the mapping relationship between the signal and the equipment to determine whether the preset topology conditions are met. If the preset topology conditions are met, then based on the signal type information, a dual mechanism combining basic electrical quantity criteria and multi-source external data collaborative criteria is used for fault diagnosis to obtain basic electrical quantity judgment results and multi-source external data collaborative judgment results; the basic electrical quantity criteria are formed based on switch status signals and protection action signals. The basic electrical quantity judgment result and the multi-source external data collaborative judgment result are weighted and fused for output. When the basic electrical quantity judgment result and the multi-source external data collaborative judgment result are consistent, the confidence level is increased to a preset range. When they are inconsistent, a review prompt is generated to obtain AC equipment fault diagnosis data. For DC equipment, a DC voltage parameter mapping relationship under different operating states is constructed based on the operating signal. Serial event recording signals are monitored to trigger a lockout judgment or a restart judgment to obtain DC equipment fault diagnosis data. The lockout judgment includes dynamically querying trip records based on the operating signal and monitoring the disconnector status. The restart judgment includes using an adaptive polling strategy to query the DC voltage value and match it with the DC voltage parameter mapping relationship.

2. The method according to claim 1, characterized in that, The remote signaling signals include protection action signals, switch status signals, disconnector status signals, and serial event recording signals; the telemetry signals include voltage signals, current signals, or DC voltage value signals. The protection action signal is the action signal issued by the relay protection device when it detects a power system fault. It is used to combine with the switch status signal to form a basic electrical quantity criterion. The protection action signal includes line protection action signal, main transformer protection action signal, bus protection action signal, traveling wave protection signal, voltage change rate protection signal, DC undervoltage protection signal, and DC differential protection signal. The switch status signal includes the switch position status signal and the switch timing action signal. The serial event recording signal is obtained through the serial event recording interface; the serial event recording signal includes an event text description field, a site name field, and a system identifier field; the serial event recording signal is used to trigger a latch-up judgment or a restart judgment.

3. The method according to claim 1, characterized in that, For AC equipment, based on the operating signals and main wiring topology diagram, the remote signaling signals and telemetry signals are associated with the equipment ledger to form a signal-to-equipment mapping relationship, including: For AC equipment, based on the electrical connection relationship between devices in the preset main wiring topology diagram, the protection action signal, switch position status signal, voltage signal and current signal are associated with the nearest primary equipment ledger through electrical connection points to establish the correspondence between each signal and its corresponding equipment, and obtain the signal-to-equipment mapping relationship.

4. The method according to claim 1, characterized in that, Based on the preset alarm setting specifications, the signals associated with each device in the signal-to-device mapping relationship are classified and configured according to signal type to obtain signal type information, including: Based on the preset alarm setting specifications, the protection action signal associated with each device in the signal-to-device mapping relationship is configured as an accident signal type, and the switch status signal and disconnector status signal are configured as change signal types to obtain signal type information; the signal type information is used as the basis for signal retrieval and filtering.

5. The method according to claim 1, characterized in that, The response to detecting a switch tripping signal from the operating signal triggers a tripping judgment function, and based on the mapping relationship between the signal and the equipment, retrieves the status of the equipment connected to the tripping switch, and determines whether the preset topology conditions are met, including: When any switch changes from closed to open from the switch position status signal in the operation signal, the trip judgment function is triggered. Based on the mapping relationship between the signal and the equipment, all adjacent equipment that have an electrical connection with the trip switch and meet the preset conditions are retrieved from the main wiring topology diagram. Acquire the switch position status signal, disconnector position status signal, and grounding switch position status signal of the adjacent devices; Set topology conditions; the topology conditions include the switch being in the open position, the disconnect switch being in the closed position, and the grounding switch being in the open position. If all the adjacent devices retrieved meet the topology conditions, then the preset topology conditions are satisfied; if any of the adjacent devices does not meet the topology conditions, then the analysis is stopped and the device is determined to be a faultless trip.

6. The method according to claim 1, characterized in that, If the preset topology conditions are met, then based on the signal type information, a dual mechanism combining basic electrical quantity criteria and multi-source external data collaborative criteria is used for fault diagnosis to obtain the basic electrical quantity judgment result and the multi-source external data collaborative judgment result, including: Once the preset topology conditions are met, the protection action signals corresponding to the accident signal type and the switch status signals corresponding to the change signal type are filtered out based on the signal type information. The filtered protection action signals and switch status signals are analyzed using basic electrical quantity criteria. The corresponding switch timing action judgment logic or switch and protection action combination judgment logic is executed according to the equipment type to obtain the basic electrical quantity judgment result. The switch timing action judgment logic includes: based on the switch status signal, judging whether the switch has completed the open-close-open action or the open-close action within a preset time. If the open-close-open action is completed, it is judged as a permanent fault; if the open-close action is completed, it is judged as a momentary fault. The logic for determining the combination of switch and protection action includes: based on the switch status signal and the protection action signal, determining whether the switch remains in the open state after being opened and simultaneously receives the protection action signal; if so, it is determined to be a permanent fault. By accessing a multi-source external data system, data from lightning location systems, online monitoring devices, oil chromatography online monitoring data, and environmental monitoring systems are obtained. Trip events are then analyzed collaboratively to obtain collaborative judgment results from the multi-source external data. The collaborative analysis includes lightning location collaborative analysis, online monitoring collaborative analysis, oil chromatography collaborative analysis, and environmental monitoring collaborative analysis; The lightning location collaborative analysis includes: retrieving lightning strike records within a preset time period before and after the tripping time and within a preset distance range on both sides of the line corridor; if the spatial distance between the lightning strike point and the tripped line is less than a first preset threshold and the time difference is less than a second preset threshold, it is determined to be a high-confidence lightning strike fault. The online monitoring and collaborative analysis includes: acquiring monitoring data in conductor temperature, sag, insulator leakage current, capacitance value changes and unbalanced current; if the monitoring data before tripping shows an abnormal trend that meets preset conditions, it is marked as suspected equipment aging or internal fault. The oil chromatography synergistic analysis includes: obtaining the gas content of acetylene and hydrogen in the online monitoring data of transformer oil chromatography; if the gas content before tripping meets the abnormal increase in the preset conditions, it is used to help determine that it is an internal fault of the transformer. The environmental monitoring collaborative analysis includes: acquiring environmental monitoring data, and determining the correlation between the fault and environmental factors based on the environmental monitoring data; the environmental monitoring data includes temperature and humidity monitoring data, condensation monitoring data, and animal invasion monitoring data.

7. The method according to claim 1, characterized in that, The basic electrical quantity judgment result and the multi-source external data collaborative judgment result are weighted and fused for output. When the basic electrical quantity judgment result and the multi-source external data collaborative judgment result are consistent, the confidence level is increased to a preset range. When they are inconsistent, a review prompt is generated, and AC equipment fault diagnosis data is obtained, including: Configure a first weight value for the judgment result of the basic electrical quantity, and configure a second weight value for the collaborative judgment result of the multi-source external data; When the basic electrical quantity judgment result is consistent with the fault type indicated by the multi-source external data collaborative judgment result, the first weight value and the second weight value are added together and multiplied by a preset enhancement coefficient to obtain a confidence level within a preset high confidence range, and the fault type is output; the confidence level is used to characterize the reliability of the fault diagnosis result; When the basic electrical quantity judgment result is inconsistent with the fault type indicated by the multi-source external data collaborative judgment result, the first weight value and the second weight value are added together and multiplied by a preset weakening coefficient to obtain a preset low confidence range confidence level, generate a review prompt, and output judgment result difference information. The fault type or the difference in the judgment result is used as fault diagnosis data for AC equipment.

8. The method according to claim 1, characterized in that, For DC equipment, a DC voltage parameter mapping relationship is constructed based on the operating signal under different operating states. Serial event recording signals are monitored to trigger a lockout or restart judgment, resulting in DC equipment fault diagnosis data, including: For DC equipment, based on the DC voltage value in the operating signal, a mapping relationship of DC voltage parameters under different operating states of the DC substation is constructed; the operating states include full voltage operating state, 80% step-down operating state, 70% step-down operating state, and lockout state; Listen to the event text description field and point group field in the serial event recording signal. When a phase-shifting blocking command or unlocking status keyword is detected, a blocking judgment is triggered. When the keywords "traveling wave protection", "voltage change rate protection", "DC undervoltage protection" or "DC differential protection" are detected, a restart judgment is triggered. The blocking determination includes: using a machine learning model to predict the best query time, dynamically querying the remote signaling historical event table to obtain trip records within the predicted time, and simultaneously collecting the dual position signals of the disconnect switches on both sides of the circuit breaker for consistency verification. When the disconnect switch is in the closed position, the DC blocking is determined to be established. The restart determination includes: adopting an adaptive polling strategy, first polling and waiting for a first preset time to query the DC voltage value, dynamically adjusting the subsequent polling interval according to the voltage recovery trend, matching the queried DC voltage value with the DC voltage parameter mapping relationship within a preset maximum number of polling times, and determining that the restart is successful if a full-voltage restart or buck-voltage restart state is matched; otherwise, the restart is determined to have failed. The result of the lockout judgment or the result of the restart judgment is output as fault diagnosis data for DC equipment.

9. An electronic device, characterized in that, The electronic device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the method according to any one of claims 1 to 8.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the method of any one of claims 1 to 8.

Images