Transformer fault classification early warning method, system and equipment based on heartbeat synchronization, and readable storage medium

Abstract

The invention relates to a transformer fault classification early warning method and system based on heartbeat synchronization, electronic equipment and a computer readable storage medium. The method comprises the following steps: acquiring partial discharge monitoring data of a transformer, and when the data indicates that a partial discharge event occurs, obtaining a fault classification result according to a partial discharge pulse signal; sending local event information including local detection time and local signal amplitude of the current partial discharge event to an adjacent monitoring node; receiving proximity event information, and obtaining the proximity event information corresponding to the current partial discharge event according to the proximity detection time and the local detection time; time synchronization is carried out among the monitoring nodes through heartbeat messages; performing statistics on each adjacent signal amplitude of the current partial discharge event to obtain an adjacent average amplitude; and when the difference value between the local signal amplitude and the adjacent average amplitude is greater than a first threshold value, fault classification early warning information is sent to an early warning platform. By adopting the method, the potential fault of the transformer can be reliably identified.

Description

Technical Field

[0001] This application relates to the field of power safety technology, and in particular to a method, system, device, electronic device, computer-readable storage medium, and computer program product for transformer fault classification and early warning based on heartbeat synchronization. Background Technology

[0002] Power transformers are core equipment in the power grid, and their insulation condition directly affects the safe and stable operation of the power system. Partial discharge (PD), as an important indicator of insulation degradation, is crucial for effective monitoring and early warning of potential transformer faults. Current technologies typically only provide simple alarms for the occurrence of partial discharge in transformers, which is insufficient for reliable identification of potential faults. Summary of the Invention

[0003] Therefore, it is necessary to provide a method, system, device, electronic device, computer-readable storage medium, and computer program product for transformer fault classification and early warning based on heartbeat synchronization to address the above-mentioned technical problems.

[0004] Firstly, this application provides a transformer fault classification and early warning method based on heartbeat synchronization, including:

[0005] Acquire partial discharge monitoring data of the transformer; when the partial discharge monitoring data indicates that a partial discharge event has occurred in the transformer, obtain the fault classification result of the transformer based on the partial discharge pulse signal of the current partial discharge event.

[0006] The local event information of the current partial discharge event is sent to each neighboring monitoring node; the local event information includes the local detection time of the current partial discharge event and the local signal amplitude of the partial discharge pulse signal.

[0007] The system receives neighboring event information from the neighboring monitoring nodes, and obtains neighboring event information corresponding to the current partial discharge event based on the neighboring detection time included in the neighboring event information and the local detection time; wherein, the monitoring nodes corresponding to the transformer synchronize their time through heartbeat messages.

[0008] The amplitude values ​​of the neighboring signals contained in the neighboring event information corresponding to the current partial discharge event are statistically analyzed to obtain the average amplitude of the neighboring signals.

[0009] If the difference between the local signal amplitude and the neighboring average amplitude is greater than a first threshold, a fault classification warning message containing the local detection time, the local signal amplitude, and the fault classification result is sent to the warning platform.

[0010] In one embodiment, the method further includes: upon receiving a heartbeat message from another monitoring node corresponding to the transformer, obtaining a clock offset based on the reception time of the heartbeat message, the reference time indicated by the heartbeat message, and the transmission delay; correcting the local logic clock based on the clock offset; obtaining the clock drift rate of the local logic clock based on the clock offset and the historical clock offset corresponding to the previous correction; and compensating the timekeeping of the local logic clock using the clock drift rate.

[0011] In one embodiment, the method further includes: when the local signal amplitude of the partial discharge pulse signal of the current partial discharge event is greater than a second threshold, sending a first heartbeat message containing the local detection time of the current partial discharge event to other monitoring nodes corresponding to the transformer.

[0012] In one embodiment, the method further includes: obtaining the average interval time of each first heartbeat message in a historical time period, and obtaining a candidate heartbeat synchronization period based on the average interval time; the first heartbeat message is sent by the monitoring node corresponding to the transformer when the local signal amplitude of the partial discharge pulse signal of the detected partial discharge event is greater than a second threshold; a heartbeat synchronization timer is started based on the minimum value of the candidate heartbeat synchronization period and the maximum heartbeat synchronization period; and a second heartbeat message is sent to other monitoring nodes corresponding to the transformer when triggered by the heartbeat synchronization timer.

[0013] In one embodiment, acquiring partial discharge monitoring data of the transformer includes: acquiring electric field monitoring data of the transformer using an ultra-high frequency sensor; acquiring mechanical wave monitoring data of the transformer using an ultrasonic partial discharge sensor; and obtaining partial discharge monitoring data of the transformer based on the electric field monitoring data and the mechanical wave monitoring data.

[0014] In one embodiment, obtaining the fault classification result of the transformer based on the partial discharge pulse signal of the current partial discharge event includes: extracting the waveform features of the partial discharge pulse signal; matching the waveform features with each feature template in the waveform fingerprint feature library to obtain the matching degree between the waveform features and each feature template; and obtaining the fault classification result of the transformer based on the fault type corresponding to the feature template with the highest matching degree.

[0015] Secondly, this application also provides a transformer fault classification and early warning system based on heartbeat synchronization. The system includes multiple monitoring nodes and an early warning platform, and the monitoring nodes synchronize their time through heartbeat messages.

[0016] The monitoring node is used to acquire partial discharge monitoring data of the transformer. When the partial discharge monitoring data indicates that a partial discharge event has occurred in the transformer, the fault classification result of the transformer is obtained based on the partial discharge pulse signal of the current partial discharge event.

[0017] The monitoring node is also used to send local event information of the current partial discharge event to each neighboring monitoring node; the local event information includes the local detection time of the current partial discharge event and the local signal amplitude of the partial discharge pulse signal;

[0018] The monitoring node is also configured to receive neighboring event information from the neighboring monitoring nodes, and obtain neighboring event information corresponding to the current partial discharge event based on the neighboring detection time included in the neighboring event information and the local detection time; wherein, the monitoring nodes corresponding to the transformer synchronize their time through heartbeat messages;

[0019] The monitoring node is also used to statistically analyze the amplitude of the neighboring signal contained in the neighboring event information corresponding to the current partial discharge event, and obtain the neighboring average amplitude.

[0020] The monitoring node is also used to send fault classification warning information, including the local detection time, the local signal amplitude, and the fault classification result, to the early warning platform when the difference between the local signal amplitude and the neighboring average amplitude is greater than a first threshold.

[0021] The early warning platform is used to receive the fault classification early warning information and to visualize the fault classification early warning information.

[0022] The early warning platform is also used to generate an insulation status trend report of the transformer based on the fault classification early warning information.

[0023] Thirdly, this application also provides a transformer fault classification and early warning device based on heartbeat synchronization, comprising:

[0024] The fault classification module is used to acquire partial discharge monitoring data of the transformer. When the partial discharge monitoring data indicates that a partial discharge event has occurred in the transformer, the module obtains the fault classification result of the transformer based on the partial discharge pulse signal of the current partial discharge event.

[0025] The first transmitting module is used to transmit local event information of the current partial discharge event to each neighboring monitoring node; the local event information includes the local detection time of the current partial discharge event and the local signal amplitude of the partial discharge pulse signal;

[0026] The first receiving module is used to receive neighboring event information from the neighboring monitoring nodes, and obtain neighboring event information corresponding to the current partial discharge event based on the neighboring detection time included in the neighboring event information and the local detection time; wherein, the monitoring nodes corresponding to the transformer synchronize their time through heartbeat messages;

[0027] The amplitude statistics module is used to statistically analyze the amplitude values ​​of neighboring signals contained in the neighboring event information corresponding to the current partial discharge event, and obtain the neighboring average amplitude.

[0028] The second sending module is used to send fault classification warning information, which includes the local detection time, the local signal amplitude, and the fault classification result, to the warning platform when the difference between the local signal amplitude and the neighboring average amplitude is greater than a first threshold.

[0029] Fourthly, this application also provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0030] Acquire partial discharge monitoring data of the transformer; when the partial discharge monitoring data indicates that a partial discharge event has occurred in the transformer, obtain the fault classification result of the transformer based on the partial discharge pulse signal of the current partial discharge event.

[0031] The local event information of the current partial discharge event is sent to each neighboring monitoring node; the local event information includes the local detection time of the current partial discharge event and the local signal amplitude of the partial discharge pulse signal.

[0032] The system receives neighboring event information from the neighboring monitoring nodes, and obtains neighboring event information corresponding to the current partial discharge event based on the neighboring detection time included in the neighboring event information and the local detection time; wherein, the monitoring nodes corresponding to the transformer synchronize their time through heartbeat messages.

[0033] The amplitude values ​​of the neighboring signals contained in the neighboring event information corresponding to the current partial discharge event are statistically analyzed to obtain the average amplitude of the neighboring signals.

[0034] If the difference between the local signal amplitude and the neighboring average amplitude is greater than a first threshold, a fault classification warning message containing the local detection time, the local signal amplitude, and the fault classification result is sent to the warning platform.

[0035] Fifthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, performs the following steps:

[0036] Acquire partial discharge monitoring data of the transformer; when the partial discharge monitoring data indicates that a partial discharge event has occurred in the transformer, obtain the fault classification result of the transformer based on the partial discharge pulse signal of the current partial discharge event.

[0037] The local event information of the current partial discharge event is sent to each neighboring monitoring node; the local event information includes the local detection time of the current partial discharge event and the local signal amplitude of the partial discharge pulse signal.

[0038] The system receives neighboring event information from the neighboring monitoring nodes, and obtains neighboring event information corresponding to the current partial discharge event based on the neighboring detection time included in the neighboring event information and the local detection time; wherein, the monitoring nodes corresponding to the transformer synchronize their time through heartbeat messages.

[0039] The amplitude values ​​of the neighboring signals contained in the neighboring event information corresponding to the current partial discharge event are statistically analyzed to obtain the average amplitude of the neighboring signals.

[0040] If the difference between the local signal amplitude and the neighboring average amplitude is greater than a first threshold, a fault classification warning message containing the local detection time, the local signal amplitude, and the fault classification result is sent to the warning platform.

[0041] Sixthly, this application also provides a computer program product, including a computer program that, when executed by a processor, performs the following steps:

[0042] Acquire partial discharge monitoring data of the transformer; when the partial discharge monitoring data indicates that a partial discharge event has occurred in the transformer, obtain the fault classification result of the transformer based on the partial discharge pulse signal of the current partial discharge event.

[0043] The local event information of the current partial discharge event is sent to each neighboring monitoring node; the local event information includes the local detection time of the current partial discharge event and the local signal amplitude of the partial discharge pulse signal.

[0044] The system receives neighboring event information from the neighboring monitoring nodes, and obtains neighboring event information corresponding to the current partial discharge event based on the neighboring detection time included in the neighboring event information and the local detection time; wherein, the monitoring nodes corresponding to the transformer synchronize their time through heartbeat messages.

[0045] The amplitude values ​​of the neighboring signals contained in the neighboring event information corresponding to the current partial discharge event are statistically analyzed to obtain the average amplitude of the neighboring signals.

[0046] If the difference between the local signal amplitude and the neighboring average amplitude is greater than a first threshold, a fault classification warning message containing the local detection time, the local signal amplitude, and the fault classification result is sent to the warning platform.

[0047] The aforementioned transformer fault classification and early warning method, system, device, electronic equipment, computer-readable storage medium, and computer program product based on heartbeat synchronization first acquires partial discharge monitoring data of the transformer. When the partial discharge monitoring data indicates that a partial discharge event has occurred in the transformer, the fault classification result of the transformer is obtained based on the partial discharge pulse signal of the current partial discharge event. Subsequently, local event information of the current partial discharge event is sent to each neighboring monitoring node. This information includes the local detection time of the current partial discharge event and the local signal amplitude of the partial discharge pulse signal. At the same time, neighboring event information from neighboring monitoring nodes is received, and neighboring event information corresponding to the current partial discharge event is obtained based on the neighboring detection time and the local detection time contained in the neighboring event information. The monitoring nodes corresponding to the transformer synchronize their time through heartbeat messages. Then, the neighboring signal amplitudes contained in each neighboring event information corresponding to the current partial discharge event are statistically analyzed to obtain the neighboring average amplitude. When the difference between the local signal amplitude and the neighboring average amplitude is greater than a first threshold, fault classification early warning information containing the local detection time, local signal amplitude, and fault classification result is sent to the early warning platform.

[0048] This scheme utilizes multiple monitoring nodes to monitor the partial discharge status of the transformer and classifies partial discharge events upon detection, enabling preliminary intelligent identification of fault types at the edge. By exchanging event information among multiple monitoring nodes and using the time information contained in the event information to correlate partial discharge events, each monitoring node can collect signal amplitude data corresponding to the same partial discharge event from neighboring monitoring nodes. A group comparison mechanism can then determine whether the detected partial discharge event is a local anomaly occurring near the node, thus enabling self-diagnosis of the faulty node. Furthermore, using heartbeat messages to synchronize the time of multiple monitoring nodes ensures time consistency among distributed nodes. This allows monitoring nodes to accurately correlate partial discharge times using the time information in the event information, laying the foundation for subsequent insulation status trend analysis of the transformer. Finally, by sending fault classification warning information from the faulty node to the early warning platform, the platform can obtain more reliable information such as event occurrence time, partial discharge pulse signal amplitude, and fault classification results detected by monitoring nodes closer to the discharge point, enabling reliable identification and accurate early warning of potential transformer faults. Attached Figure Description

[0049] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0050] Figure 1 This is a schematic diagram of a transformer fault classification and early warning system based on heartbeat synchronization in one embodiment;

[0051] Figure 2 This is a schematic diagram showing the location of the monitoring nodes in one embodiment;

[0052] Figure 3 This is a flowchart illustrating a group consensus decision-making process in one embodiment;

[0053] Figure 4 This is a schematic diagram of the composite power supply module in one embodiment;

[0054] Figure 5 This is a schematic diagram of the waveform fingerprint comparison process in one embodiment;

[0055] Figure 6 This is a schematic diagram of a transformer fault classification and early warning system based on heartbeat synchronization in another embodiment;

[0056] Figure 7 This is a schematic diagram of an adaptive hybrid synchronization protocol in one embodiment;

[0057] Figure 8 This is a flowchart illustrating a transformer fault classification and early warning method based on heartbeat synchronization in one embodiment.

[0058] Figure 9 This is a structural block diagram of a transformer fault classification and early warning device based on heartbeat synchronization in one embodiment;

[0059] Figure 10 This is a diagram of the internal structure of an electronic device in one embodiment. Detailed Implementation

[0060] 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 and not intended to limit the scope of this application.

[0061] It should be noted that the terms "first," "second," etc., used in this application can be used to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish the first element from the second element. The terms "comprising" and "having," and any variations thereof, used in this application, are intended to cover non-exclusive inclusion. The term "multiple" used in this application refers to two or more. The term "and / or" used in this application refers to one of the embodiments, or any combination of multiple embodiments.

[0062] In one exemplary embodiment, such as Figure 1 As shown, a transformer fault classification and early warning system based on heartbeat synchronization is provided. The system includes multiple monitoring nodes and an early warning platform. The monitoring nodes synchronize their time through heartbeat messages.

[0063] Specifically, the system in this embodiment may include a monitoring sensor network composed of multiple monitoring nodes and an early warning platform. The monitoring nodes can communicate with each other, and each monitoring node can also communicate and connect to the early warning platform individually. For example, as shown... Figure 2 As shown, a monitoring sensor network consisting of multiple monitoring nodes can be deployed at different locations on the same transformer, and different monitoring nodes can synchronize their time through heartbeat messages.

[0064] The monitoring node is used to acquire partial discharge monitoring data of the transformer. When the partial discharge monitoring data indicates that a partial discharge event has occurred in the transformer, the fault classification result of the transformer is obtained based on the partial discharge pulse signal of the current partial discharge event. For example, the probe of the monitoring node can be installed inside the transformer, or it can be screwed into or attached to the dielectric window of the transformer.

[0065] Each monitoring node can integrate one or more sensors to continuously monitor the electric field environment around the transformer to obtain partial discharge monitoring data. For example, the partial discharge monitoring data may include mechanical wave monitoring data and electric field monitoring data. When a partial discharge pulse signal is captured in the partial discharge monitoring data, a suspected partial discharge event can be identified in the transformer. This event can be taken as the current partial discharge event, and a preliminary fault classification can be performed on the transformer based on the partial discharge pulse signal to obtain the corresponding fault classification result. For example, the fault classification result corresponding to the current partial discharge event can be determined by waveform fingerprint comparison.

[0066] The monitoring node is also used to send local event information of the current partial discharge event to each neighboring monitoring node; the local event information includes the local detection time of the current partial discharge event and the local signal amplitude of the partial discharge pulse signal.

[0067] In this system, each monitoring node that detects a partial discharge event can broadcast local event information of the current partial discharge event to other nearby monitoring nodes via a wireless network. This information may include the local detection time of the current partial discharge event and the local signal amplitude of the partial discharge pulse signal. The local detection time can be used to indicate the time when the monitoring node detected the current partial discharge event locally, and can be represented by a local timestamp T. local The local signal amplitude can be the signal amplitude of the partial discharge pulse signal corresponding to the current partial discharge event, which is captured locally by the monitoring node. The neighboring monitoring nodes corresponding to each monitoring node can be pre-set based on the deployment location of each monitoring node on the transformer.

[0068] The monitoring node is also used to receive neighboring event information from neighboring monitoring nodes, and obtain neighboring event information corresponding to the current partial discharge event based on the neighboring detection time and the local detection time contained in the neighboring event information.

[0069] In a scenario where a neighboring monitoring node detects a partial discharge event, that monitoring node can receive neighboring event information from the neighboring monitoring node. It can be understood that for the neighboring monitoring node that sent this neighboring event information, this information is its own local event information.

[0070] Upon receiving proximity event information from neighboring monitoring nodes, the monitoring node can determine the proximity detection time and the proximity signal amplitude based on this information. The proximity detection time can indicate the time it takes for a neighboring monitoring node to detect a partial discharge event locally, and the proximity signal amplitude can be the signal amplitude of the partial discharge pulse signal captured locally by the neighboring monitoring node.

[0071] Since the physical time difference between each node detecting the same event is usually extremely short, and the monitoring nodes maintain time synchronization through heartbeat signals, the monitoring node can determine whether the neighboring event information corresponds to the current partial discharge event based on the local detection time corresponding to the current partial discharge event and the neighboring detection time extracted from the neighboring event information. For example, when the time difference between the neighboring detection time and the local detection time is less than a preset duration, the neighboring event information can be determined to correspond to the current partial discharge event.

[0072] The monitoring node is also used to statistically analyze the amplitude of neighboring signals contained in the information of each neighboring event corresponding to the current partial discharge event, and obtain the average amplitude of the neighboring events; when the difference between the local signal amplitude and the average amplitude of the neighboring events is greater than a first threshold, it sends fault classification warning information containing the local detection time, local signal amplitude and fault classification result to the warning platform.

[0073] Among them, the monitoring nodes in the system can determine whether to trigger the early warning reporting process through group consensus decision-making.

[0074] For example, the process of group consensus decision-making can be as follows: Figure 3 As shown in the diagram, after determining the information of each neighboring event corresponding to the current partial discharge event, the monitoring node can obtain the amplitude of the neighboring signals contained in these neighboring event information, and calculate the average of these neighboring signal amplitudes to obtain the monitoring node's neighboring average amplitude. Subsequently, the monitoring node can compare its own local signal amplitude obtained from checking the current partial discharge event with the neighboring average amplitude to determine whether its own local signal amplitude is significantly higher than the neighboring average amplitude.

[0075] Specifically, when the local signal amplitude is greater than the neighboring average amplitude, and the difference between the local signal amplitude and the neighboring average amplitude is greater than a first threshold, it can be determined that the local signal amplitude is significantly higher than the neighboring average amplitude. In this case, the monitoring node can determine that the current partial discharge event is a local anomaly, indicating that the discharge source is located near the node, with a high degree of confidence. Therefore, the monitoring node can trigger an early warning reporting process, sending a fault classification early warning message to the early warning platform, which includes the local detection time, local signal amplitude, and fault classification result. For example, the fault classification early warning message may also include the monitoring node's node identifier and confidence level. This confidence level can be used to indicate the reliability of the fault classification result. For example, the larger the difference between the monitoring node's local signal amplitude and the neighboring average amplitude, the higher the corresponding confidence level.

[0076] When the local signal amplitude is not significantly higher than the neighboring average amplitude, the monitoring node can determine that the detected current partial discharge event is a global interference with low confidence, and thus can ignore and discard the event.

[0077] The early warning platform is used to receive fault classification early warning information, visualize the fault classification early warning information, and generate a transformer insulation status trend report based on the fault classification early warning information.

[0078] The early warning platform can receive fault classification early warning information from monitoring nodes and display it visually. For example, it can visualize the occurrence time of a partial discharge event (i.e., the local detection time), the node identifier of the monitoring node that reported the event, the amplitude of the partial discharge pulse signal, and the fault classification results.

[0079] The early warning platform can also aggregate fault classification early warning information corresponding to multiple partial discharge events of a transformer over a period of time to generate an insulation status trend report for the transformer. This insulation status trend report can be used to display the changing trend of the transformer's insulation status over time.

[0080] In the aforementioned transformer fault classification and early warning system based on heartbeat synchronization, multiple monitoring nodes monitor the partial discharge status of the transformer and classify it as a fault upon detection of a partial discharge event. This enables preliminary intelligent identification of the fault type of the partial discharge event at the edge. By exchanging event information among multiple monitoring nodes and using the time information contained in the event information to correlate partial discharge events, each monitoring node can collect signal amplitude data corresponding to the same partial discharge event from neighboring monitoring nodes. A group comparison mechanism can then be used to determine whether the detected partial discharge event is a local anomaly occurring near the node, thus enabling self-diagnosis of the faulty node. Furthermore, using heartbeat messages to synchronize the time of multiple monitoring nodes ensures time consistency among distributed nodes. This allows monitoring nodes to accurately correlate partial discharge times using the time information in the event information, laying the foundation for subsequent insulation status trend analysis of the transformer. In this way, by sending fault classification warning information to the early warning platform from the fault node, the early warning platform can obtain information with higher reliability, such as the event occurrence time, partial discharge pulse signal amplitude, and fault classification results, which are detected by monitoring nodes closer to the discharge point, thus realizing reliable identification and accurate early warning of potential transformer faults.

[0081] In one exemplary embodiment, the monitoring node includes a composite power supply module; the composite power supply module is used to power the circuits within the monitoring node using thermal and mechanical energy collected from the operating environment of the transformer.

[0082] Specifically, each monitoring node in the system can achieve self-powering through a composite power supply module. This module utilizes composite energy harvesting technology to collect thermal and mechanical energy from the transformer's operating environment and convert it into electrical energy to power the circuits within the monitoring node.

[0083] For example, such as Figure 4As shown, the composite energy supply module may include a thermoelectric generator (TEG), a vibration energy harvesting unit, a power management IC (PMIC), and an energy storage unit.

[0084] The hot end of the thermoelectric generator unit can be tightly attached to the transformer tank wall through a high thermal conductivity insulating ceramic sheet, and the cold end of the thermoelectric generator unit can be connected to the aluminum alloy heat sink fins of the transformer. The thermoelectric module of the thermoelectric generator unit can output electrical energy to the power management circuit when there is a temperature difference between the hot end and the cold end.

[0085] The vibration energy harvesting unit may include a vibration energy harvester and an AC / DC energy conversion circuit. The vibration energy harvester may employ a piezoelectric cantilever beam structure, which generates electrical energy when vibration occurs in the transformer's operating environment and inputs it to the AC / DC energy conversion circuit. For example, the natural frequency of the vibration energy harvester can be tuned to the 100Hz magnetostrictive vibration frequency of the transformer body. The AC / DC energy conversion circuit converts the AC power from the vibration energy harvester into DC power and outputs it to the power management circuit.

[0086] The power management circuit may include a maximum power point tracking (MPPT) controller and a charge / discharge management circuit. The MPPT controller may have dual-input MPPT functionality, enabling it to monitor the output status of the thermoelectric generator and vibration energy harvesting unit in real time and automatically adjust the input impedance. The charge / discharge management circuit can input the electrical energy output from the MPPT controller to the energy storage unit to charge its capacitors. The circuit can also control the energy storage unit to discharge, stably releasing the stored energy to the load, thus powering the circuitry within the monitoring node.

[0087] In this embodiment, by utilizing composite energy harvesting technology to achieve self-powering of the monitoring nodes, the long-term maintenance-free operation requirements of the nodes can be met, and the reliability of transformer fault early warning can be improved.

[0088] In an exemplary embodiment, the monitoring node is further configured to: upon receiving a heartbeat message from another monitoring node corresponding to the transformer, obtain a clock offset based on the reception time of the heartbeat message, the reference time indicated by the heartbeat message, and the transmission delay; correct the local logic clock based on the clock offset; obtain the clock drift rate of the local logic clock based on the clock offset and the historical clock offset corresponding to the previous correction; and compensate for the timekeeping of the local logic clock using the clock drift rate.

[0089] In this system, monitoring nodes receive heartbeat messages via a wireless communication network. These heartbeat messages can be sent by one of the other monitoring nodes corresponding to the same transformer, and all other nodes can receive them. For example, the node sending the heartbeat message can be a designated master node in the system, which can send heartbeat messages to other nodes when triggered by a heartbeat synchronization timer. For example, the node sending the heartbeat message can also be a node that detects a strong partial discharge event (hereinafter referred to as a "strong PD event") in the system, and this node can send heartbeat messages to other nodes upon detecting a strong PD event. For example, a strong PD event can be a partial discharge event where the amplitude of the partial discharge pulse signal is greater than a second threshold.

[0090] The heartbeat message may include reference time information, which can indicate the reference time used for this calibration. For example, the reference time information may be the local timestamp T at which the node sending the heartbeat message was triggered to send the message. local The node that receives the heartbeat message can record the message's arrival time T. receive And utilize the receiving time T receive Reference time T for heartbeat message indication local and transmission delay Δt x (Based on the estimated fixed wireless transmission delay), calculate the clock offset. For example, the clock offset can be expressed as: Δoffset = T receive -Δt x -T local Subsequently, the monitoring nodes can calibrate their local logic clock using the clock offset. Thus, the monitoring nodes in the system can achieve time synchronization via heartbeat signals. After all monitoring nodes have completed calibration, the system can reset the heartbeat synchronization timer.

[0091] After completing calibration based on the heartbeat signal, the monitoring node can also obtain the historical clock offset from the previous calibration and calculate the clock drift rate of the local logic clock based on the clock offset and the historical clock offset corresponding to the previous calibration. Subsequently, during subsequent timekeeping processes, the monitoring node can use this drift rate to dynamically compensate for the timekeeping of the local logic clock, thereby significantly suppressing the accumulation of offset between two synchronizations and keeping the time base stable over a long period of time.

[0092] In this embodiment, by using heartbeat messages to synchronize the time between nodes and by calculating the clock drift rate to dynamically compensate the local clock, the time synchronization between the monitoring nodes in the system can be improved, which is beneficial for more accurate fault monitoring and early warning of transformers.

[0093] In an exemplary embodiment, the monitoring node is further configured to: send a first heartbeat message containing the local detection time of the current partial discharge event to other monitoring nodes corresponding to the transformer when the local signal amplitude of the partial discharge pulse signal of the current partial discharge event is greater than a second threshold.

[0094] Specifically, when the local signal amplitude of the partial discharge pulse signal corresponding to the current partial discharge event captured by any monitoring node in the system is greater than the second threshold, the current partial discharge event can be determined to be a strong PD event. Therefore, the node can act as a temporary master node, and the local timestamp T corresponding to the local detection time of the strong PD event can be used. local The message is encapsulated into a first heartbeat message and broadcast. This allows the remaining nodes in the system to receive the first heartbeat message and use it for clock calibration.

[0095] In an exemplary embodiment, the system further includes a heartbeat synchronization timer, which can be used to: obtain the average interval time of each first heartbeat message in a historical period, and obtain a candidate heartbeat synchronization period based on the average interval time; the first heartbeat message is sent by the monitoring node corresponding to the transformer when the local signal amplitude of the partial discharge pulse signal of the detected partial discharge event is greater than a second threshold; the heartbeat synchronization timer is started based on the minimum value of the candidate heartbeat synchronization period and the maximum heartbeat synchronization period; and the second heartbeat message is sent to other monitoring nodes corresponding to the transformer when triggered by the heartbeat synchronization timer.

[0096] The system may include a heartbeat synchronization timer, which can be deployed on a designated master node in the system or deployed independently and communicate with the master node.

[0097] Among them, the heartbeat synchronization timer can obtain the average interval T of the first heartbeat message triggered by a strong PD event (i.e., a partial discharge event where the signal amplitude of the partial discharge pulse signal is greater than the second threshold) in the historical period. avg (i.e., the average interval between historical strong PD events), and calculate the candidate heartbeat synchronization period based on the scaling factor K (usually K>1). Subsequently, the candidate heartbeat synchronization period and the maximum heartbeat synchronization period T can be used as a basis for determining the synchronization period. max The minimum value is used to obtain the dynamic heartbeat synchronization period T. heartbeat For example, the dynamic heartbeat synchronization period can be represented as: The heartbeat synchronization timer can start timing based on the calculated dynamic heartbeat synchronization period, and trigger the master node to update the current local timestamp T when the timing ends. local The message is encapsulated into a second heartbeat message and broadcast. This allows the remaining nodes in the system to receive the second heartbeat message and use it for clock calibration.

[0098] After synchronizing the time of all monitoring nodes in the system with the first or second heartbeat message, the heartbeat synchronization timer can be reset. This timer can then recalculate the new dynamic heartbeat synchronization cycle and start a new round of timing.

[0099] In this embodiment, an adaptive hybrid synchronization protocol is used to dynamically adjust the heartbeat synchronization cycle length based on the average interval of historical strong PD events. This combines the efficiency of event-triggered synchronization with the reliability of heartbeat synchronization, thereby improving the accuracy of time synchronization between nodes.

[0100] In an exemplary embodiment, the monitoring node is further configured to: acquire electric field monitoring data of the transformer using an ultra-high frequency sensor; acquire mechanical wave monitoring data of the transformer using an ultrasonic partial discharge sensor; and obtain partial discharge monitoring data of the transformer based on the electric field monitoring data and the mechanical wave monitoring data.

[0101] Specifically, the monitoring node can integrate a dual-sensor unit, including an ultra-high frequency (UHF) sensor and an ultrasonic partial discharge (PD) sensor. The UHF sensor detects electromagnetic waves in the UHF band, thereby capturing the partial discharge signal of the transformer to obtain electric field monitoring data. The ultrasonic PD sensor senses mechanical waves to obtain mechanical wave monitoring data. Therefore, by utilizing the electric field monitoring data and mechanical wave monitoring data obtained from the dual-sensor unit, the partial discharge monitoring data of the transformer can be obtained.

[0102] In this embodiment, by employing a dual magnetic sensing architecture for monitoring data acquisition, which takes into account both high-frequency response and mechanical wave measurement, data reliability can be ensured.

[0103] In an exemplary embodiment, the monitoring node is further configured to: extract waveform features of the partial discharge pulse signal; match the waveform features with each feature template in the waveform fingerprint feature library to obtain the matching degree between the waveform features and each feature template; and obtain the fault classification result of the transformer based on the fault type corresponding to the feature template with the highest matching degree.

[0104] Specifically, such as Figure 5 As shown, the monitoring node can perform preliminary fault classification of the current partial discharge event by comparing waveform fingerprints.

[0105] The monitoring node may include a storage module, which can pre-store a waveform fingerprint feature library. This library may include multiple feature templates, each associated with a specific fault type. Before system deployment, raw discharge pulse signals of partial discharge events corresponding to various fault types can be collected. These signals are then preprocessed, denoised, and normalized before feature extraction to obtain waveform features corresponding to each raw discharge pulse signal. These waveform features can then be clustered to obtain feature templates corresponding to various typical fault types. Subsequently, the waveform fingerprint feature library can be constructed using these feature templates and their corresponding fault types and stored in the monitoring node's storage module.

[0106] The monitoring node can preprocess and extract features from the partially discharged pulse signal after monitoring and capturing it to obtain the waveform features. For example, the waveform features of the partially discharged pulse signal may include, but are not limited to, the rise time, fall time, pulse width, oscillation frequency, amplitude, and amplitude asymmetry. Subsequently, the Mahalanobis distance between the waveform features and each feature template in the waveform fingerprint feature library can be calculated to obtain the matching degree between the waveform features and each feature template. The feature template with the smallest Mahalanobis distance to the waveform feature is the feature template with the highest matching degree. Then, based on the fault type corresponding to the feature template with the highest matching degree, the fault classification result of the transformer can be obtained.

[0107] In this embodiment, by performing feature matching based on a waveform fingerprint feature library, preliminary intelligent identification of fault types can be achieved at the edge side, thereby improving the fault identification efficiency of the transformer.

[0108] In one exemplary embodiment, a transformer fault classification and early warning system based on heartbeat synchronization is provided.

[0109] Please refer to the following: Figure 6This is a schematic diagram of the architecture of the transformer fault classification and early warning system based on heartbeat synchronization in this embodiment. The system can include distributed monitoring nodes, an in-station aggregation unit, and a background early warning platform. The monitoring nodes can be distributed and installed outside the transformer tank. Multiple monitoring nodes form a monitoring sensor network deployed together on a single transformer. Different monitoring nodes can perform adaptive heartbeat synchronization using heartbeat messages. The monitoring nodes can communicate with each other through a wireless communication network and utilize comprehensive information to achieve collective intelligent collaborative work. The deployment location of each monitoring node on the transformer can be determined based on the monitoring node's performance and the transformer's monitoring requirements. The in-station aggregation unit can act as a gateway, communicating with the monitoring nodes through a wireless communication network and with the background early warning platform through a fiber optic network. For example, when a monitoring node triggers the early warning reporting process, it can send fault classification and early warning information to the in-station aggregation unit through the wireless communication network, and then the in-station aggregation unit will send the information to the background early warning platform through the fiber optic network.

[0110] Among them, such as Figure 6 As shown, each monitoring node in the system can integrate a storage module, a processing and control module, a composite power supply module, a core sensing module, and a wireless communication module. The structure of the composite power supply module can be as follows: Figure 4 As shown, it can utilize thermal and mechanical energy collected from the transformer's operating environment to power the circuitry within the monitoring node. The core sensing module may include a dual-sensor unit, which can include an ultrasonic partial discharge sensor for sensing mechanical waves and an ultra-high frequency sensor for measuring ultra-high frequency electromagnetic waves. The wireless communication module can support low-power wireless technologies such as Zigbee or LoRa, and can be used for data interaction between nodes and between nodes and the station's aggregation platform. The storage module can store a pre-set database of typical discharge waveform fingerprint features and the monitoring node's operating data. The signal processing and control module can incorporate a low-power microprocessor (MCU), which can be used to control the sensors and perform related operations such as signal feature extraction, fingerprint matching, time synchronization using synchronization protocols, and group decision-making.

[0111] Among them, such as Figure 6 As shown, the background early warning platform (hereinafter referred to as the "early warning platform") in the system can provide functions such as cloud server, data aggregation, visualization, and trend analysis.

[0112] Based on such Figure 6 The system shown can perform fault classification and early warning for transformers through the following steps:

[0113] Step S1, Signal Sensing and Feature Extraction. Each monitoring node in the system can continuously monitor the electric field environment around the transformer using the core sensing module; when a suspected partial discharge pulse signal is captured, the signal processing and control module can extract the key feature vector of the pulse waveform, such as, but not limited to, rise time, fall time, pulse width, oscillation frequency, amplitude, amplitude asymmetry, polarity, bandwidth, center frequency, etc.

[0114] Step S2: Waveform fingerprint comparison and preliminary classification. The signal processing and control module of the monitoring node can calculate the similarity between the key feature vector obtained in S1 and the feature templates in the waveform fingerprint feature library pre-stored in the storage module. By calculating the Mahalanobis distance, the matching degree between the key feature vector and each feature template is obtained. The template type with the highest matching degree is the fault classification result for this partial discharge event.

[0115] Step S3, group consensus decision. Monitoring nodes can reach each other through mechanisms such as... Figure 3 The process described involves group consensus decision-making. Each monitoring node can broadcast local event information (including fault classification results, local signal amplitude, and local detection time) to neighboring nodes via a wireless communication module, and simultaneously receive neighboring event information from all neighboring nodes. The monitoring node can use its signal processing and control module to determine whether the local event information corresponds to the same partial discharge event (i.e., the current partial discharge event) based on the local detection time and the neighboring event information. After determining the neighboring event information corresponding to the current partial discharge event, the average amplitude of all neighboring signals corresponding to the current partial discharge event can be calculated, and then the local signal amplitude is compared with the average amplitude of the neighbors. If the abnormal signal strength detected by the node is significantly higher than that of most surrounding nodes (i.e., the local signal amplitude is significantly higher than the neighboring average amplitude), it is determined to be a local anomaly, indicating that the discharge source is located near the node, with high confidence, and the early warning reporting process is triggered. If the signal levels of all nodes are similar, it is determined to be global interference, with low confidence, and the event is ignored and discarded.

[0116] Step S4: Early Warning Reporting. For locally anomalous events confirmed as having high confidence, monitoring nodes can upload fault classification early warning information, including node ID, precise logical timestamp (local detection time), fault classification result, local signal amplitude, and confidence level, to the backend early warning platform via the on-site aggregation unit. The platform can aggregate all information, provide visual display, and generate an insulation status trend report.

[0117] Among them, each monitoring node in the system can adopt such as Figure 7The adaptive hybrid synchronization protocol shown maintains the time logic consistency of all nodes. This protocol is a bimodal system that combines the efficiency of event-triggered synchronization with the reliability of heartbeat synchronization. Its core process is as follows:

[0118] Step 1 Protocol Initialization: After the system is powered on, each monitoring node uses its own local clock and enters a continuous listening state.

[0119] Step 2 Event-Triggered Synchronization: This mode is triggered immediately when any node detects a strong PD event (partial discharge pulse signal amplitude exceeding a threshold). This node can act as a temporary master node, recording its local timestamp T for detecting the event. local The message is encapsulated into a first heartbeat message and broadcast. Upon receiving the message, the remaining nodes record the arrival time T. receive Because electromagnetic waves propagate much faster than wireless signals, the physical time difference between nodes detecting the same event is extremely short (on the order of microseconds). The receiving node can then use the formula... Calculate its own clock offset (where The system estimates a fixed wireless transmission delay and immediately applies this offset to calibrate its local logical clock. After completing all actions for event-triggered synchronization (recording timestamps, broadcasting messages, and calibrating the clock), the system resets the "Adaptive Heartbeat Timer".

[0120] Step 3: Adaptive Heartbeat Synchronization: Set a dynamic heartbeat synchronization period. The calculation formula is as follows: ,in The average interval between historical strong PD events is denoted by K, which is a scaling factor (usually K>1). The maximum heartbeat synchronization period is set (e.g., 7 days). After system initialization, the adaptive heartbeat timer can begin counting down, and the period of this timer is... It is dynamically calculated. As soon as this timer expires, regardless of whether a partial discharge event occurs, the system will initiate synchronization via the designated master node. The master node can broadcast a second heartbeat message, which includes the master node's current local timestamp T. local The receiving node also calculates the offset and calibrates the clock.

[0121] Step 4: Clock Drift Rate Compensation: To prevent excessive drift within the heartbeat interval, the protocol introduces a clock drift rate estimation and compensation mechanism. During each synchronization, a node not only calculates its current offset Δoffset but also compares it with the data from the previous synchronization to estimate its own clock drift rate. In subsequent timekeeping processes, the node uses this drift rate to dynamically compensate for the timekeeping of its logical clock, thereby significantly suppressing the accumulation of offsets between synchronizations and maintaining long-term stability of the time base.

[0122] Step 5: Historical Data Update: After each successful synchronization, the system will update the average interval of historical strong PD events. This is used to adjust the next heartbeat cycle so that the protocol behavior always matches the actual operating state of the transformer.

[0123] This embodiment provides an early classification and warning system for insulation faults based on distributed sensing, adaptive time synchronization, waveform fingerprint comparison, and group consensus decision-making, and capable of self-powering. It achieves complete self-powering by utilizing composite energy harvesting technology, meeting the requirements for long-term maintenance-free operation; it employs a dual-magnetic sensing architecture, balancing high-frequency response and mechanical wave measurement to ensure data reliability; based on a waveform fingerprint feature library, it achieves preliminary intelligent identification of discharge types at the edge; by adopting an adaptive hybrid synchronization protocol, it ensures long-term, reliable time consistency among distributed nodes, laying the foundation for trend analysis; and through a group consensus comparison mechanism, it reduces the false alarm rate and enables self-diagnosis of faulty nodes, ultimately forming a distributed, intelligent, highly reliable, and low-communication-requirement early warning solution for transformer insulation conditions.

[0124] Based on the same inventive concept, this application also provides a transformer fault classification and early warning method based on heartbeat synchronization. The solution provided by this method is similar to the solution described in the above system embodiments. Therefore, the specific limitations of one or more embodiments of the transformer fault classification and early warning method based on heartbeat synchronization provided below can be found in the limitations of the transformer fault classification and early warning system based on heartbeat synchronization described above, and will not be repeated here.

[0125] In one exemplary embodiment, such as Figure 8 As shown, a transformer fault classification and early warning method based on heartbeat synchronization is provided, and this method is applied to... Figure 1 Taking the monitoring nodes in the example, the following steps are included:

[0126] Step S801: Obtain partial discharge monitoring data of the transformer. If the partial discharge monitoring data indicates that a partial discharge event has occurred in the transformer, obtain the fault classification result of the transformer based on the partial discharge pulse signal of the current partial discharge event.

[0127] Step S802: Send local event information of the current partial discharge event to each neighboring monitoring node; the local event information includes the local detection time of the current partial discharge event and the local signal amplitude of the partial discharge pulse signal.

[0128] Step S803: Receive neighboring event information from neighboring monitoring nodes, and obtain neighboring event information corresponding to the current partial discharge event based on the neighboring detection time and local detection time contained in the neighboring event information; wherein, the monitoring nodes corresponding to the transformer synchronize their time through heartbeat messages.

[0129] Step S804: Statistically analyze the amplitude values ​​of neighboring signals contained in the information of each neighboring event corresponding to the current partial discharge event to obtain the average amplitude of the neighboring events.

[0130] Step S805: If the difference between the local signal amplitude and the neighboring average amplitude is greater than the first threshold, send fault classification warning information containing the local detection time, local signal amplitude and fault classification result to the warning platform.

[0131] In an exemplary embodiment, the method may further include: upon receiving a heartbeat message from another monitoring node corresponding to the transformer, obtaining a clock offset based on the reception time of the heartbeat message, the reference time indicated by the heartbeat message, and the transmission delay; correcting the local logic clock based on the clock offset; obtaining the clock drift rate of the local logic clock based on the clock offset and the historical clock offset corresponding to the previous correction; and compensating for the timekeeping of the local logic clock using the clock drift rate.

[0132] In an exemplary embodiment, the method may further include: when the local signal amplitude of the partial discharge pulse signal of the current partial discharge event is greater than a second threshold, sending a first heartbeat message containing the local detection time of the current partial discharge event to other monitoring nodes corresponding to the transformer.

[0133] In an exemplary embodiment, the method may further include: obtaining the average interval time of each first heartbeat message in a historical time period, and obtaining a candidate heartbeat synchronization period based on the average interval time; the first heartbeat message is sent by the monitoring node corresponding to the transformer when the local signal amplitude of the partial discharge pulse signal of the detected partial discharge event is greater than a second threshold; a heartbeat synchronization timer is started based on the minimum value of the candidate heartbeat synchronization period and the maximum heartbeat synchronization period; and a second heartbeat message is sent to other monitoring nodes corresponding to the transformer when triggered by the heartbeat synchronization timer.

[0134] In one exemplary embodiment, acquiring partial discharge monitoring data of a transformer may include: acquiring electric field monitoring data of the transformer using an ultra-high frequency sensor; acquiring mechanical wave monitoring data of the transformer using an ultrasonic partial discharge sensor; and obtaining partial discharge monitoring data of the transformer based on the electric field monitoring data and the mechanical wave monitoring data.

[0135] In an exemplary embodiment, obtaining the transformer fault classification result based on the partial discharge pulse signal of the current partial discharge event may include: extracting the waveform features of the partial discharge pulse signal; matching the waveform features with each feature template in the waveform fingerprint feature library to obtain the matching degree between the waveform features and each feature template; and obtaining the transformer fault classification result based on the fault type corresponding to the feature template with the highest matching degree.

[0136] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages in other steps. It is understood that the steps in different embodiments can be freely combined as needed, and all non-contradictory solutions formed by such combinations are within the scope of protection of this application.

[0137] Based on the same inventive concept, this application also provides a transformer fault classification and early warning device based on heartbeat synchronization for implementing the aforementioned transformer fault classification and early warning method based on heartbeat synchronization. The solution provided by this device is similar to the solution described in the above method. Therefore, the specific limitations of one or more embodiments of the transformer fault classification and early warning device based on heartbeat synchronization provided below can be found in the limitations of the transformer fault classification and early warning method based on heartbeat synchronization described above, and will not be repeated here.

[0138] In one exemplary embodiment, such as Figure 9 As shown, a transformer fault classification and early warning device based on heartbeat synchronization is provided, comprising:

[0139] The fault classification module 901 is used to acquire partial discharge monitoring data of the transformer. When the partial discharge monitoring data indicates that a partial discharge event has occurred in the transformer, the fault classification result of the transformer is obtained based on the partial discharge pulse signal of the current partial discharge event.

[0140] The first transmitting module 902 is used to transmit local event information of the current partial discharge event to each neighboring monitoring node; the local event information includes the local detection time of the current partial discharge event and the local signal amplitude of the partial discharge pulse signal.

[0141] The first receiving module 903 is used to receive neighboring event information from the neighboring monitoring nodes, and obtain neighboring event information corresponding to the current partial discharge event based on the neighboring detection time and the local detection time included in the neighboring event information; wherein, the monitoring nodes corresponding to the transformer synchronize their time through heartbeat messages.

[0142] The amplitude statistics module 904 is used to statistically analyze the amplitude of the neighboring signals contained in the neighboring event information corresponding to the current partial discharge event, and obtain the neighboring average amplitude.

[0143] The second sending module 905 is used to send fault classification warning information, which includes the local detection time, the local signal amplitude and the fault classification result, to the warning platform when the difference between the local signal amplitude and the neighboring average amplitude is greater than a first threshold.

[0144] In an exemplary embodiment, the apparatus further includes: an offset calculation module, configured to, upon receiving a heartbeat message from another monitoring node corresponding to the transformer, obtain a clock offset based on the reception time of the heartbeat message, a reference time indicated by the heartbeat message, and a transmission delay; a clock correction module, configured to correct the local logic clock based on the clock offset; a drift rate calculation module, configured to obtain the clock drift rate of the local logic clock based on the clock offset and the historical clock offset corresponding to the previous correction; and a clock compensation module, configured to compensate for the timekeeping of the local logic clock using the clock drift rate.

[0145] In an exemplary embodiment, the apparatus further includes a third transmitting module, configured to send a first heartbeat message containing the local detection time of the current partial discharge event to other monitoring nodes corresponding to the transformer when the local signal amplitude of the partial discharge pulse signal of the current partial discharge event is greater than a second threshold.

[0146] In an exemplary embodiment, the apparatus further includes: a candidate period acquisition module, configured to acquire the average interval time of each first heartbeat message in a historical time period, and obtain a candidate heartbeat synchronization period based on the average interval time; the first heartbeat message is sent by the monitoring node corresponding to the transformer when the local signal amplitude of the partial discharge pulse signal of the detected partial discharge event is greater than a second threshold; a timer start module, configured to start a heartbeat synchronization timer based on the minimum value of the candidate heartbeat synchronization period and the maximum heartbeat synchronization period; and a fourth sending module, configured to send a second heartbeat message to other monitoring nodes corresponding to the transformer under the triggering of the heartbeat synchronization timer.

[0147] In an exemplary embodiment, the fault classification module 901 is configured to: acquire electric field monitoring data of the transformer using an ultra-high frequency sensor; acquire mechanical wave monitoring data of the transformer using an ultrasonic partial discharge sensor; and obtain partial discharge monitoring data of the transformer based on the electric field monitoring data and the mechanical wave monitoring data.

[0148] In an exemplary embodiment, the fault classification module 901 is configured to: extract waveform features of the partial discharge pulse signal; match the waveform features with each feature template in the waveform fingerprint feature library to obtain the matching degree between the waveform features and each feature template; and obtain the fault classification result of the transformer according to the fault type corresponding to the feature template with the highest matching degree.

[0149] The modules in the aforementioned transformer fault classification and early warning device based on heartbeat synchronization can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of the electronic device in hardware form or independently of it, or stored in the memory of the electronic device in software form, so that the processor can call and execute the corresponding operations of each module.

[0150] In one exemplary embodiment, an electronic device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 10 As shown, this electronic device includes a processor, memory, input / output interfaces, and a communication interface. The processor, memory, and input / output interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the input / output interfaces. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system and computer programs. The internal memory provides the environment for the operation of the operating system and computer programs stored in the non-volatile storage medium. The input / output interfaces are used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, Near Field Communication (NFC), or other technologies. When the computer program is executed by the processor, it implements a transformer fault classification and early warning method based on heartbeat synchronization.

[0151] Those skilled in the art will understand that Figure 10 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the electronic device to which the present application is applied. The specific electronic device may include more or fewer components than shown in the figure, or combine certain components, or have different component arrangements.

[0152] In one embodiment, an electronic device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0153] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon that, when executed by a processor, implements the steps in the above method embodiments.

[0154] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0155] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0156] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0157] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0158] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A transformer fault classification and early warning method based on heartbeat synchronization, characterized in that, The method includes: Acquire partial discharge monitoring data of the transformer; when the partial discharge monitoring data indicates that a partial discharge event has occurred in the transformer, obtain the fault classification result of the transformer based on the partial discharge pulse signal of the current partial discharge event. The local event information of the current partial discharge event is sent to each neighboring monitoring node; the local event information includes the local detection time of the current partial discharge event and the local signal amplitude of the partial discharge pulse signal. The system receives neighboring event information from the neighboring monitoring nodes, and obtains neighboring event information corresponding to the current partial discharge event based on the neighboring detection time included in the neighboring event information and the local detection time; wherein, the monitoring nodes corresponding to the transformer synchronize their time through heartbeat messages. The amplitude values ​​of the neighboring signals contained in the neighboring event information corresponding to the current partial discharge event are statistically analyzed to obtain the average amplitude of the neighboring signals. If the difference between the local signal amplitude and the neighboring average amplitude is greater than a first threshold, a fault classification warning message containing the local detection time, the local signal amplitude, and the fault classification result is sent to the warning platform.

2. The method according to claim 1, characterized in that, The method further includes: Upon receiving a heartbeat message from another monitoring node corresponding to the transformer, the clock offset is obtained based on the reception time of the heartbeat message, the reference time indicated by the heartbeat message, and the transmission delay. Correct the local logic clock based on the clock offset; The clock drift rate of the local logic clock is obtained based on the clock offset and the historical clock offset corresponding to the previous correction. The clock drift rate is used to compensate for the timekeeping of the local logic clock.

3. The method according to claim 2, characterized in that, The method further includes: If the local signal amplitude of the partial discharge pulse signal of the current partial discharge event is greater than the second threshold, a first heartbeat message containing the local detection time of the current partial discharge event is sent to other monitoring nodes corresponding to the transformer.

4. The method according to claim 2, characterized in that, The method further includes: The average interval time of each first heartbeat message in the historical period is obtained, and a candidate heartbeat synchronization period is obtained based on the average interval time; the first heartbeat message is sent by the monitoring node corresponding to the transformer when the local signal amplitude of the partial discharge pulse signal of the detected partial discharge event is greater than the second threshold. The heartbeat synchronization timer is started based on the minimum value of the candidate heartbeat synchronization period and the maximum heartbeat synchronization period; When the heartbeat synchronization timer is triggered, a second heartbeat message is sent to other monitoring nodes corresponding to the transformer.

5. The method according to claim 1, characterized in that, The acquisition of partial discharge monitoring data of the transformer includes: Electric field monitoring data of the transformer are acquired using an ultra-high frequency sensor; Mechanical wave monitoring data of the transformer are acquired using an ultrasonic partial discharge sensor; Based on the electric field monitoring data and the mechanical wave monitoring data, the partial discharge monitoring data of the transformer is obtained.

6. The method according to any one of claims 1 to 5, characterized in that, The step of obtaining the fault classification result of the transformer based on the partial discharge pulse signal of the current partial discharge event includes: Extract the waveform features of the partial discharge pulse signal; The waveform feature is matched with each feature template in the waveform fingerprint feature library to obtain the matching degree between the waveform feature and each feature template; The fault classification result of the transformer is obtained based on the fault type corresponding to the feature template with the highest matching degree.

7. A transformer fault classification and early warning system based on heartbeat synchronization, characterized in that, The system includes multiple monitoring nodes and an early warning platform, and the monitoring nodes synchronize their time through heartbeat messages. The monitoring node is used to acquire partial discharge monitoring data of the transformer. When the partial discharge monitoring data indicates that a partial discharge event has occurred in the transformer, the fault classification result of the transformer is obtained based on the partial discharge pulse signal of the current partial discharge event. The monitoring node is also used to send local event information of the current partial discharge event to each neighboring monitoring node; the local event information includes the local detection time of the current partial discharge event and the local signal amplitude of the partial discharge pulse signal; The monitoring node is also configured to receive neighboring event information from the neighboring monitoring nodes, and obtain neighboring event information corresponding to the current partial discharge event based on the neighboring detection time included in the neighboring event information and the local detection time; wherein, the monitoring nodes corresponding to the transformer synchronize their time through heartbeat messages; The monitoring node is also used to statistically analyze the amplitude of the neighboring signal contained in the neighboring event information corresponding to the current partial discharge event, and obtain the neighboring average amplitude. The monitoring node is also used to send fault classification warning information, including the local detection time, the local signal amplitude, and the fault classification result, to the early warning platform when the difference between the local signal amplitude and the neighboring average amplitude is greater than a first threshold. The early warning platform is used to receive the fault classification early warning information and to visualize the fault classification early warning information. The early warning platform is also used to generate an insulation status trend report of the transformer based on the fault classification early warning information.

8. The system according to claim 7, characterized in that, The monitoring node includes a composite power supply module; the composite power supply module is used to supply power to the circuits within the monitoring node using thermal and mechanical energy collected from the operating environment of the transformer.

9. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images