A substation fault detection method and device

Abstract

The application provides a substation fault detection method and device, and belongs to the technical field of fault detection. The method comprises the following steps: acquiring electrical quantity data and equipment operation data of a substation; performing first fault determination on the electrical quantity data and the equipment operation data based on a first determination condition to obtain a fault probability; the first determination condition comprises threshold values corresponding to the electrical quantity data and the equipment operation data respectively; if the fault probability is lower than a first fault probability threshold value, performing second fault determination on the electrical quantity data and the equipment operation data; if the second fault determination result is a fault, inputting the electrical quantity data and the equipment operation data into a fault detection model to generate a fault detection result; if the fault probability is not lower than the first fault probability threshold value, inputting the electrical quantity data and the equipment operation data into the fault detection model to generate the fault detection result; and the fault detection model is obtained by training based on a historical fault data set. The application can improve the fault detection efficiency of the substation.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of fault detection, and more particularly relates to a substation fault detection method and device. BACKGROUND

[0002] As the core hub of the power system, the operation state of the substation directly affects the safety and stability of power supply, and fault detection is a key link to ensure the reliable operation of the substation. Electrical quantity data and equipment operation data are core parameters reflecting the state of the substation, and the fault detection method based on these parameters is the research focus of the prior art.

[0003] In the prior art, the substation fault detection adopts a single determination mode, for example, some methods only determine the electrical quantity or equipment operation data by presetting a fixed threshold, and when the data exceeds the threshold, it is determined as a fault. However, this method is difficult to cope with complex working conditions such as load fluctuation and equipment aging, and is prone to misjudgment or omission due to normal data fluctuation; some other methods use machine learning models to detect all data, which can improve the determination accuracy, but need to process a large amount of redundant data, and have the problems of large amount of calculation and low detection efficiency.

[0004] Therefore, how to improve the fault detection efficiency of the substation while ensuring the detection accuracy has become a technical problem to be solved in the prior art. SUMMARY

[0005] The purpose of the present application is to provide a substation fault detection method and device to improve the fault detection efficiency of the substation.

[0006] The first aspect of the embodiment of the present application provides a substation fault detection method, comprising:

[0007] Obtaining electrical quantity data and equipment operation data of the substation, performing first fault determination on the electrical quantity data and the equipment operation data based on a first determination condition to obtain a fault probability; the first determination condition includes a threshold corresponding to each of the electrical quantity data and the equipment operation data;

[0008] If the fault probability is lower than a first fault probability threshold, performing second fault determination on the electrical quantity data and the equipment operation data; if the second fault determination result is a fault, inputting the electrical quantity data and the equipment operation data into a fault detection model to generate a fault detection result;

[0009] If the fault probability is not lower than the first fault probability threshold, inputting the electrical quantity data and the equipment operation data into the fault detection model to generate the fault detection result; the fault detection model is obtained by training based on a historical fault data set, and the historical fault data set includes historical electrical quantity data, historical equipment operation data and a fault type label.

[0010] In a second aspect, the application provides a substation fault detection device, comprising:

[0011] a first fault determination module configured to acquire electrical quantity data and equipment operation data of the substation, perform first fault determination on the electrical quantity data and the equipment operation data based on a first determination condition, and obtain a fault probability; the first determination condition comprises a threshold corresponding to each of the electrical quantity data and the equipment operation data;

[0012] a second fault determination module configured to, if the fault probability is lower than a first fault probability threshold, perform second fault determination on the electrical quantity data and the equipment operation data; and if the second fault determination result is fault, input the electrical quantity data and the equipment operation data into a fault detection model to generate a fault detection result;

[0013] a third fault determination module configured to, if the fault probability is not lower than the first fault probability threshold, input the electrical quantity data and the equipment operation data into the fault detection model to generate the fault detection result; the fault detection model is trained based on a historical fault data set, and the historical fault data set comprises historical electrical quantity data, historical equipment operation data, and a fault type label.

[0014] In a third aspect, the application provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and running on the processor; when the processor executes the computer program, the steps of the substation fault detection method described above are implemented.

[0015] In a fourth aspect, the application provides a computer readable storage medium, which stores a computer program; when the computer program is executed by a processor, the steps of the substation fault detection method described above are implemented.

[0016] The substation fault detection method and device provided by the application have the beneficial effect that the application significantly improves the comprehensive performance of substation fault detection through the collaborative design of hierarchical determination and model application.

[0017] On the one hand, the first fault determination designed by the application calculates the fault probability based on the threshold condition of electrical quantity and equipment operation data, which can quickly screen out high-risk samples and reduce the amount of data analysis entering the complex model. For the case where the fault probability is not lower than the first threshold, the application selects to directly input the data into the model for processing, ensuring the detection accuracy of high-risk scenarios; for the low probability case, the application selects to further verify through the second fault determination, avoiding invalid model calls caused by single threshold misjudgment, and effectively reducing redundant calculation.

[0018] On the other hand, the model trained by the historical fault data set provides data support for the final detection result, and in combination with the screening effect of the previous hierarchical determination, the invalid data proportion of the model processing is reduced while ensuring the detection accuracy, and the overall detection cycle is shortened.

[0019] In summary, the embodiment of the present application considers the accuracy and efficiency of fault detection through the hierarchical architecture of rapid preliminary screening, accurate verification and model detection, solves the problem of low efficiency caused by single determination mode or full data modeling in the prior art, and is suitable for the high-efficiency fault detection demand of complex working conditions of a transformer substation. BRIEF DESCRIPTION OF DRAWINGS

[0020] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0021] Figure 1 A flowchart of a transformer substation fault detection method provided by an embodiment of the present application;

[0022] Figure 2 A structural block diagram of a transformer substation fault detection device provided by an embodiment of the present application;

[0023] Figure 3 A schematic block diagram of an electronic device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0024] In the following description, specific details are set forth in order to provide a thorough understanding of the embodiments of the present application. However, persons skilled in the art should understand that the present application can also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted so as not to obscure the description of the present application with unnecessary details.

[0025] In order to make the purpose, technical scheme and advantages of the present application clearer, the following will be described by specific embodiments in conjunction with the drawings.

[0026] Please refer to Figure 1 , Figure 1 A flowchart of a transformer substation fault detection method provided by an embodiment of the present application, the method can be executed by an electronic device, specifically, the method can include S101-S103.

[0027] S101: Obtain electrical quantity data and equipment operation data of the transformer substation, perform first fault determination on the electrical quantity data and the equipment operation data based on a first determination condition, and obtain a fault probability; the first determination condition includes threshold values corresponding to the electrical quantity data and the equipment operation data respectively.

[0028] In the embodiment, the electrical quantity data includes current data and voltage data in a first time period, and the equipment operation data includes equipment temperature and equipment vibration frequency in the first time period; the first fault determination on the electrical quantity data and the equipment operation data based on the first determination condition includes: comparing the current data, the voltage data, the equipment temperature and the equipment vibration frequency in the first time period with the threshold values corresponding thereto respectively; determining the fault probability according to the total number of times that the current data, the voltage data, the equipment temperature and the equipment vibration frequency exceed the threshold values corresponding thereto respectively in the first time period; and the fault probability is positively correlated with the total number of times.

[0029] In the embodiment, the transformer substation refers to a power facility for transforming voltage and distributing electric energy, and can include devices such as transformers and circuit breakers. The electrical quantity data refers to data reflecting electrical characteristics of a power system, including current data and voltage data, wherein the current data is, for example, three-phase current instantaneous value, effective value and current mutation amplitude, and the voltage data is, for example, line voltage, phase voltage and voltage fluctuation value. The equipment operation data refers to data representing the state of the equipment, for example, the equipment operation data can include equipment temperature and equipment vibration frequency, wherein the equipment temperature is, for example, transformer winding temperature and circuit breaker surface temperature, and the equipment vibration frequency is, for example, transformer core vibration frequency and switch mechanism vibration frequency.

[0030] In the embodiment, the first determination condition refers to a preliminary fault determination rule based on a preset standard, and can specifically include threshold values corresponding to each item of data, for example, the current threshold value can be set according to the rated current of the equipment and the upper and lower limit threshold values of the load fluctuation, the voltage threshold value can be set according to the rated voltage of the power grid and the allowable deviation, the temperature threshold value can be set according to the temperature resistance of the equipment material, and the vibration frequency threshold value can be set according to the frequency range of the equipment during normal operation. The first fault determination refers to a process of analyzing the data by the first determination condition to determine the possibility of failure. The fault probability refers to the possibility of the transformer substation failure in the determination result. The first time period refers to a continuous monitoring period for data collection. The total number of times that the data exceeds the threshold value refers to the cumulative number of times that the current, voltage, temperature and vibration frequency exceed the threshold values respectively in the first time period. The current, voltage, temperature and vibration frequency data can each include multiple data at different time nodes.

[0031] In the embodiment, the preliminary quantitative determination of the fault is realized by hierarchical data monitoring and threshold comparison. First, the electrical quantity and equipment operation data are continuously collected in the first time period, and each data is compared with the preset threshold to count the total number of times of exceeding the threshold. The more the number of times of exceeding the threshold, the more significant the abnormal characteristics of the equipment, and the higher the probability of the fault. The logic behind this is that the fault of the substation equipment is often accompanied by the abnormality of multiple parameters. In addition to the fault factor, the temporary exceeding of a single parameter is more likely to be caused by external environmental interference. The exceeding of multiple parameters or multiple / continuous exceeding can be more indicative of a real fault. The embodiment can comprehensively reflect the abnormality degree of the equipment by counting the total number of times of exceeding the threshold, avoid the misjudgment of a single parameter, and quickly generate the probability of the fault in a simple quantitative manner to provide a basis for subsequent processing, taking into account efficiency and preliminary accuracy.

[0032] For example, the first time period can be set to 10 minutes, and the current data, line voltage data, transformer surface temperature data and circuit breaker operating mechanism vibration frequency data of the 10kV outgoing line cabinet in the period can be collected. The embodiment can preset the threshold values: the lower limit threshold value of the current is 50A, the upper limit threshold value of the current is 400A, the upper limit threshold value of the voltage is 10.7kV, the lower limit threshold value of the voltage is 9.3kV, the lower limit threshold value of the temperature is -10℃, the upper limit threshold value of the temperature is 70℃, the lower limit threshold value of the vibration frequency is 10Hz, and the upper limit threshold value of the vibration frequency is 100Hz, etc.

[0033] The sampling data in 10 minutes can be compared point by point. It is assumed that the current data exceeds 400A for 3 times, the voltage data is lower than 9.3kV for 2 times, the temperature data exceeds 70℃ for 1 time, and the vibration frequency does not exceed the limit. The total number of times of exceeding the threshold is 6 times. Further, the probability of the fault can be determined according to the total number of times. Since the total number of times is 6, it is in the range of slight abnormality, and the probability of the fault is set to 10%. The total number of times of data exceeding the limit is positively correlated with the probability of the fault, and the probability of the fault can be determined by a preset formula or a preset discrete data mapping table, such as 0 times corresponding to 0% probability and 10 times corresponding to 30% probability. The probability of the fault obtained by the embodiment can be used as a basis for subsequent second fault determination or directly input into the fault detection model.

[0034] S102: If the probability of the fault is lower than the first fault probability threshold value, the electrical quantity data and the equipment operation data are subjected to the second fault determination. If the result of the second fault determination is a fault, the electrical quantity data and the equipment operation data are input into the fault detection model to generate a fault detection result. The fault detection model is trained based on a historical fault data set including historical electrical quantity data, historical equipment operation data and fault type labels.

[0035] In the embodiment, the first fault probability threshold is a critical value for triggering the second fault determination, and the first fault probability threshold can be set based on historical fault data and system safety requirements. The second fault determination refers to a further verification process for low fault probability data. The second fault determination result is the conclusion of whether the substation has a fault output by the second fault determination process, and the fault indicates that the determination finds abnormal characteristics. The fault detection model is an algorithm model for accurately identifying fault types, which can be implemented using a classification algorithm. The historical fault data set is the basic data set for model training, where the historical electrical quantity data can include current and voltage data before and after the past substation fault, and the historical equipment operation data can include device temperature and device vibration data before and after the past substation fault. The fault type label is used to label the specific fault category of the historical fault data, such as short circuit and insulation aging. The fault detection result can include the fault type.

[0036] In the embodiment, when the fault probability is lower than the first fault probability threshold, the embodiment can further analyze the data trend characteristics through the second fault determination, and if significant fault characteristics are found (i.e., determined as a fault), the data is input into the fault detection model for more accurate fault detection. The logic behind this is that a relatively low fault probability still has potential progressive faults (such as slow parameter changes of device aging), and a single threshold determination is prone to missed detection. The embodiment captures such potential abnormalities through data fault trend analysis by the second fault determination, and finally, for data determined to have a fault / abnormality, the fault detection is performed in combination with the refined identification capability of the model. The embodiment can reduce missed detection while ensuring fault detection efficiency through layered detection and conditional threshold determination.

[0037] For example, the embodiment can set the first fault probability threshold to 40%, and the first fault determination result of a certain substation circuit breaker is a fault probability of 35%. Since it is lower than the first fault probability threshold, the second fault determination is triggered.

[0038] The embodiment can extract electrical quantity data of the circuit breaker for about 20 minutes, including three-phase current change slope and line voltage fluctuation trend; the embodiment can extract equipment operation data of the circuit breaker for about 20 minutes, including arc chamber temperature rise rate and operating mechanism vibration frequency drift characteristics. The embodiment compares the above trend characteristics with the pre-fault characteristics of various faults in the historical fault trend library, finds that the current change slope continuously increases, and the temperature rise rate has a similarity of 90% with the pre-fault trend of contact aging fault, which is higher than a preset threshold (such as 70%), and determines that the second fault determination result is a fault. The embodiment inputs the electrical quantity data and equipment operation data extracted within the 20 minutes into the fault detection model. The fault detection model compares the real-time characteristics with the feature mode of historical contact aging cases, outputs the fault detection result as contact aging of the circuit breaker, and completes the detection process.

[0039] S103: If the fault probability is not lower than the first fault probability threshold, input the electrical quantity data and the equipment operation data into the fault detection model to generate a fault detection result.

[0040] In this embodiment, the fault detection result is the determination information about the fault output by the model, which may include, for example, a specific fault type and a determination confidence, etc., for characterizing the specific situation of the fault.

[0041] In this embodiment, when the fault probability is not lower than the first fault probability threshold, the electrical quantity data and the equipment operation data are directly input into the fault detection model, and the fault detection result is output by the fault detection model. The consideration behind this embodiment is that when the fault probability is not lower than the threshold, it indicates that there are significant abnormal features in the data, and no additional determination step is needed. The learning ability of the fault detection model for historical fault features can be directly utilized to quickly and accurately identify the fault, ensuring accuracy while improving detection efficiency, and avoiding the influence of redundant processes on the response speed of high-risk faults.

[0042] For example, this embodiment can set the first fault probability threshold to 50% by analyzing the historical fault data of the substation and the safety operation requirements. When the first fault determination result of a certain transformer is 60%, which is not lower than the threshold, the subsequent process is triggered. This embodiment can extract the electrical quantity data of the transformer within the current 15 minutes, including the three-phase current effective value, the line voltage fluctuation value, and the equipment operation data, including the winding temperature change curve and the iron core vibration frequency sequence. This embodiment can input the extracted data into the fault detection model, the fault detection model can preprocess the input data into a feature vector, and call the historical fault data set to compare the preprocessed real-time feature vector with the historical features of the inter-turn short circuit fault, and calculate the feature matching degree. For example, the historical features of the inter-turn short circuit fault may include a current imbalance degree higher than 5%, a temperature rising rate higher than 2 ℃ / min, and a vibration main frequency offset of 10 Hz, etc. Finally, the fault detection model outputs the fault detection result according to the feature matching degree, for example, the fault type is the inter-turn short circuit of the transformer.

[0043] From the above, it can be concluded that this embodiment significantly improves the comprehensive efficiency of substation fault detection through the collaborative design of hierarchical determination and model application. On the one hand, the first fault determination designed in this embodiment calculates the fault probability based on the threshold condition of electrical quantity and equipment operation data, which can quickly screen out high-risk samples and reduce the amount of data analysis entering the complex model. For the case where the fault probability is not lower than the first threshold, this embodiment selects to directly input the data into the model for processing to ensure the detection accuracy of high-risk scenarios; for the low probability case, this embodiment selects to further verify through the second fault determination to avoid invalid model calling due to single threshold misjudgment, effectively reducing redundant calculation.

[0044] On the other hand, the model trained by the historical fault data set provides data support for the final detection result, combined with the screening effect of the previous hierarchical judgment, reduces the proportion of invalid data processed by the model while ensuring the detection accuracy, and shortens the overall detection cycle.

[0045] In summary, the embodiment takes into account the accuracy and efficiency of fault detection through the hierarchical architecture of rapid preliminary screening, accurate verification and model detection, solves the problem of low efficiency caused by single judgment mode or full data modeling in the prior art, and is suitable for efficient fault detection demand under complex working conditions of a transformer substation.

[0046] In an embodiment of the present application, the second fault judgment is performed on the electrical quantity data and the equipment operation data, including: extracting electrical trend features based on the electrical quantity data; extracting equipment operation trend features based on the equipment operation data; calculating a first fitting degree of the electrical trend features and the electrical fault trend features; calculating a second fitting degree of the equipment operation trend features and the equipment operation fault trend features; determining the second fault judgment on the electrical quantity data and the equipment operation data based on the first fitting degree and the second fitting degree.

[0047] In the embodiment, the second fault judgment is determined on the electrical quantity data and the equipment operation data based on the first fitting degree and the second fitting degree, specifically including:

[0048] The first fitting degree and the second fitting degree are weighted and summed to obtain a third fitting degree;

[0049] If the first fitting degree is higher than the first fitting degree threshold or the second fitting degree is higher than the first fitting degree threshold, and the third fitting degree is higher than the second fitting degree threshold, it is determined that the second fault judgment result is fault;

[0050] If the first fitting degree is not higher than the first fitting degree threshold, the second fitting degree is not higher than the first fitting degree threshold, and the third fitting degree is higher than the second fitting degree threshold, it is determined that the second fault judgment result is no fault.

[0051] In the embodiment, the first fitting degree refers to the matching degree of the electrical trend features and the electrical fault trend features, which can include, for example, the similarity of trend slope, fluctuation amplitude, etc. The second fitting degree refers to the matching degree of the equipment operation trend features and the equipment operation fault trend features, which can include, for example, the matching degree of temperature rise rate, vibration frequency drift, etc. The third fitting degree is a comprehensive matching value obtained by weighted sum of the first fitting degree and the second fitting degree, which is used to comprehensively reflect the overall trend matching. The first fitting degree threshold is a critical value for determining whether a single trend matches, and the second fitting degree threshold is a critical value for determining whether a comprehensive trend matches, both of which are set according to historical fault trend data.

[0052] In the embodiment, the fitting degrees of the electrical and equipment operation trend features and the corresponding fault trend features are calculated, a comprehensive fitting degree is obtained by weighting, and the second fault determination result is determined according to the comparison result of the fitting degree and the threshold. The consideration behind the embodiment is that if only a single trend feature is considered, there is a risk of accidental matching, and the comprehensive fitting degree of the electrical and equipment operation two dimensions can reduce misjudgment. The embodiment determines that there is a fault when only a single threshold is met and the comprehensive threshold is met, and determines that there is no fault when only the comprehensive threshold is met but the single threshold is not met, which not only captures significant single trend abnormalities, but also avoids misjudgment caused by a virtual high comprehensive value, thereby improving the determination accuracy.

[0053] For example, the first fitting degree threshold can be set to 0.7, the second fitting degree threshold can be set to 0.65, the electrical trend feature weight can be set to 0.6, and the equipment operation trend feature weight can be set to 0.4 by analyzing historical data. The electrical trend feature (current slope change) and the equipment operation trend feature (temperature rise rate) of a certain circuit breaker can be extracted, the first fitting degree can be calculated as 0.75, the second fitting degree can be calculated as 0.6, and the third fitting degree can be calculated as 0.75*0.6+0.6*0.4=0.69. Since the first fitting degree 0.75>0.7, the single threshold condition is met, and the second fault determination result is determined to be a fault. If the first fitting degree is 0.65, the second fitting degree is 0.6, and the third fitting degree 0.63 is less than 0.65 in another case, the second fault determination result is determined to be no fault.

[0054] The embodiment determines the trend fitting degree of the electrical and equipment operation two dimensions, captures potential faults by using significant single trend abnormalities, integrates the two-dimensional information by weighted summation, avoids misjudgment caused by accidental single feature abnormalities, and further reduces the misjudgment risk of a virtual high comprehensive value by designing that only the comprehensive threshold is met but the single threshold is not met, thereby significantly improving the accuracy and reliability of the second fault determination and providing accurate basis for efficient calling of the subsequent fault detection model.

[0055] In an embodiment of the present application, after the second fault determination on the electrical quantity data and the equipment operation data, the method further includes: if the second fault determination result is no fault, updating the fault probability corresponding to the electrical quantity data and the equipment operation data; updating the first determination condition based on the electrical quantity data, the equipment operation data, and the updated fault probability to obtain an updated first determination condition.

[0056] In the embodiment, the fault probability corresponding to the electrical quantity data and the equipment operation data is updated, specifically including: updating the fault probability based on a first step; the updated fault probability is lower than the updated fault probability; if the updated fault probability is lower than the first fault probability threshold, the first fault probability threshold is used as the updated fault probability.

[0057] In the embodiment, the first step length is a fixed amplitude for adjusting the failure probability, which may include a percentage or a specific value determined based on historical statistics of non-failure cases, for example. The updated failure probability refers to the failure probability value adjusted by the first step length, which is used to reflect the risk level after the second failure determination is non-failure. The failure probability before updating refers to the original failure probability before adjustment, which is the result of the first failure determination.

[0058] In the embodiment, when the second failure determination result is non-failure, the embodiment reduces the failure probability by the first step length. If the updated probability is lower than the first failure probability threshold, the embodiment takes the threshold as the final updated result, and optimizes the first determination condition based on the relevant data and the updated probability. The consideration behind the embodiment is that the second failure determination is non-failure, which indicates that the risk is reduced, so the probability is lowered. The embodiment limits the minimum value to the first threshold to avoid overestimating the risk. The embodiment updates the first determination condition, so that the initial determination is more accurate and the subsequent invalid determination is reduced.

[0059] For example, the embodiment can set the first step length to 5%, the first failure probability threshold to 30%, the failure probability of the first failure determination of a certain device to 35%, and the second failure determination result to non-failure. The embodiment can adjust by the first step length, and the probability before updating is 35% minus 5% to 30%. Since 30% is equal to the first threshold, the updated failure probability is 30%. The embodiment can combine the electrical quantity and device operation data of the device, and the updated 30% probability to increase the original current threshold from 400A to 420A, and obtain the updated first determination condition.

[0060] The embodiment reduces the failure probability after non-failure determination by the step length, and reasonably limits the minimum value, which accurately reflects the actual situation of risk reduction and avoids overestimating the risk. At the same time, the embodiment optimizes the first determination condition based on the data and the updated probability, so that the initial determination is more accurate, the subsequent invalid process is reduced, and the adaptability and reliability of the overall detection are improved.

[0061] In an embodiment of the present application, the first determination condition is updated based on the electrical quantity data, the device operation data and the updated failure probability to obtain the updated first determination condition, including: extracting the electrical trend feature based on the electrical quantity data, and extracting the device operation trend feature based on the device operation data; adding the electrical trend feature, the device operation trend feature and the updated failure probability to the first determination condition to obtain the updated first determination condition.

[0062] In the embodiment, the substation fault detection method further comprises: obtaining target electrical quantity data and target equipment operation data in a second time period, performing first fault determination on the target electrical quantity data and the target equipment operation data in the second time period based on the updated first determination condition, and obtaining a fault probability.

[0063] The target electrical quantity data comprises current data and voltage data in the second time period, and the target equipment operation data comprises equipment temperature and equipment vibration frequency in the second time period.

[0064] The electrical feature matching degree between the target electrical trend feature and the electrical trend feature is calculated, the equipment feature matching degree between the target equipment operation trend feature and the equipment operation trend feature is calculated, and the target feature matching degree is obtained based on the electrical feature matching degree and the equipment feature matching degree.

[0065] If the target feature matching degree is higher than the feature matching degree threshold, the updated fault probability corresponding to the electrical trend feature and the equipment operation trend feature is taken as the fault probability.

[0066] If the target feature matching degree is not higher than the feature matching degree threshold, the current data, the voltage data, the equipment temperature and the equipment vibration frequency in the second time period are compared with respective threshold values.

[0067] The fault probability is determined according to the total number of times that the current data, the voltage data, the equipment temperature and the equipment vibration frequency in the second time period exceed respective threshold values.

[0068] In the embodiment, the target electrical quantity data refers to electrical quantity data in a second time period, which may comprise current data and voltage data in the second time period, etc. The target equipment operation data refers to equipment operation data in the second time period, which may comprise equipment temperature and equipment vibration frequency in the second time period, etc. The second time period refers to a new monitoring time period for collecting target data. The target electrical trend feature refers to a trend feature extracted from the target electrical quantity data, and the target equipment operation trend feature refers to a trend feature extracted from the target equipment operation data. The electrical feature matching degree refers to the matching degree between the target electrical trend feature and the original electrical trend feature, and the equipment feature matching degree refers to the matching degree between the target equipment operation trend feature and the original equipment operation trend feature. The target feature matching degree is a comprehensive matching value obtained based on the electrical feature matching degree and the equipment feature matching degree, and the feature matching degree threshold is a critical value for determining whether the target feature matching degree meets the standard.

[0069] In the embodiment, the updated first determination condition adds the electrical trend feature, the equipment operation trend feature, and the updated failure probability; for the target data of the second time period, the embodiment can first calculate the matching degree of the target feature and the original feature, and if the matching degree is higher than the threshold, directly adopt the corresponding updated failure probability; otherwise, determine the failure probability by the original way through threshold comparison and number statistics. The consideration behind the embodiment is that: through trend feature matching, the determination result of the historical failure-free scene can be quickly reused to improve the determination efficiency of similar scenes; the threshold comparison mode is retained to cope with new scenes, and flexibility and accuracy are considered, so that the first determination is more suitable for the dynamic operation state of the equipment. If the first determination condition is not updated, the similar failure-free situation will still be determined as a failure because the total number of data exceeding the threshold is high, and the updated first determination condition can effectively solve this problem.

[0070] For example, the embodiment can extract the electrical trend feature (such as the current slope being stable at 0.1 A / min) and the equipment operation trend feature (such as the temperature fluctuation amplitude being lower than 2°C) based on the historical data, combine the updated failure probability 30%, form the updated first determination condition, and set the feature matching degree threshold to 0.8. The embodiment can obtain the target electrical quantity data and the target equipment operation data of the second time period (such as 10 minutes), extract the target electrical trend feature (such as the current slope 0.12 A / min) and the target equipment operation trend feature (such as the temperature fluctuation amplitude 1.8°C).

[0071] The embodiment can calculate the electrical feature matching degree as 0.9 (highly similar to the original electrical trend feature), the equipment feature matching degree as 0.85, obtain the target feature matching degree 0.88 (higher than 0.8), and take 30% as the target failure probability. If the target feature matching degree of another target data is 0.75 (lower than 0.8), compare the current with the updated current threshold, the temperature with the updated temperature threshold, and the like, and count the exceeding number as 0, and the corresponding failure probability is 5%.

[0072] The embodiment adds the trend feature corresponding to the historical determination error failure-free scene in the first determination condition, so that the similar failure-free scene can directly reuse the historical determination result, and the determination efficiency is significantly improved; at the same time, the embodiment still retains the threshold comparison mode to cope with new scenes, avoiding the limitation of a single mode. Through the combination of feature matching and threshold statistics, the embodiment reduces redundant calculation, guarantees adaptability to complex working conditions, effectively improves the accuracy and efficiency of the first failure determination, and provides a more reliable initial basis for the subsequent detection process.

[0073] A substation failure detection method corresponding to the above embodiment, Figure 2A structural block diagram of a substation fault detection device is provided for an embodiment of the present application. For ease of illustration, only parts related to the embodiments of the present application are shown. For reference Figure 2 The substation fault detection device 20 includes a first fault determination module 21, a second fault determination module 22, and a third fault determination module 23.

[0074] The first fault determination module 21 is configured to obtain electrical quantity data and equipment operation data of the substation, perform first fault determination on the electrical quantity data and the equipment operation data based on first determination conditions, and obtain a fault probability; the first determination conditions include respective threshold values of the electrical quantity data and the equipment operation data.

[0075] The second fault determination module 22 is configured to perform second fault determination on the electrical quantity data and the equipment operation data if the fault probability is lower than a first fault probability threshold value; and input the electrical quantity data and the equipment operation data into a fault detection model to generate a fault detection result if the second fault determination result is fault.

[0076] The third fault determination module 23 is configured to input the electrical quantity data and the equipment operation data into the fault detection model to generate the fault detection result if the fault probability is not lower than the first fault probability threshold value; the fault detection model is obtained based on a historical fault data set, and the historical fault data set includes historical electrical quantity data, historical equipment operation data, and a fault type label.

[0077] In an embodiment of the present application, the electrical quantity data includes current data and voltage data of a first time period, and the equipment operation data includes equipment temperature and equipment vibration frequency of the first time period; and the first fault determination module 21 is specifically configured to:

[0078] respectively compare the current data, the voltage data, the equipment temperature, and the equipment vibration frequency of the first time period with respective threshold values; determine the fault probability according to a total number of times that the current data, the voltage data, the equipment temperature, and the equipment vibration frequency of the first time period exceed the respective threshold values; and the fault probability is positively correlated with the total number of times.

[0079] In an embodiment of the present application, the second fault determination module 22 is specifically configured to:

[0080] extract electrical trend features based on the electrical quantity data; extract equipment operation trend features based on the equipment operation data; calculate a first fitting degree of the electrical trend features and electrical fault trend features; calculate a second fitting degree of the equipment operation trend features and equipment operation fault trend features; and determine the second fault determination on the electrical quantity data and the equipment operation data based on the first fitting degree and the second fitting degree.

[0081] In an embodiment of the present application, the secondary fault determination module 22 is further configured to:

[0082] The first fitting degree and the second fitting degree are weighted and summed to obtain a third fitting degree; if the first fitting degree is higher than the first fitting degree threshold or the second fitting degree is higher than the first fitting degree threshold, and the third fitting degree is higher than a second fitting degree threshold, it is determined that the second fault determination result is fault; if the first fitting degree is not higher than the first fitting degree threshold, the second fitting degree is not higher than the first fitting degree threshold, and the third fitting degree is higher than the second fitting degree threshold, it is determined that the second fault determination result is no fault.

[0083] In an embodiment of the present application, after the second fault determination on the electrical quantity data and the equipment operation data, the substation fault detection device 20 further comprises an adaptive updating module configured to:

[0084] If the second fault determination result is no fault, the fault probability corresponding to the electrical quantity data and the equipment operation data is updated; and the first determination condition is updated based on the electrical quantity data and the equipment operation data and the updated fault probability to obtain an updated first determination condition.

[0085] In an embodiment of the present application, the adaptive updating module is specifically configured to:

[0086] The fault probability is updated based on a first step length; the updated fault probability is lower than the fault probability before updating; and if the updated fault probability is lower than a first fault probability threshold, the first fault probability threshold is taken as the updated fault probability.

[0087] In an embodiment of the present application, the adaptive updating module is further configured to:

[0088] The electrical trend feature is extracted based on the electrical quantity data, and the equipment operation trend feature is extracted based on the equipment operation data; the electrical trend feature, the equipment operation trend feature and the updated fault probability are added to the first determination condition to obtain the updated first determination condition.

[0089] Referring to Figure 3 , Figure 3 A schematic block diagram of an electronic device provided in an embodiment of the present application is shown in FIG. 1. As shown in FIG. 1, the electronic device comprises a substation fault detection device 20. Figure 3The electronic device 300 in the embodiment shown can include one or more processors 301, one or more input devices 302, one or more output devices 303, and one or more memories 304. The above-mentioned processors 301, input devices 302, output devices 303, and memories 304 complete communication with each other through a communication bus 305. The memory 304 is configured to store a computer program, and the computer program includes program instructions. The processor 301 is configured to execute the program instructions stored in the memory 304. Specifically, the processor 301 is configured to invoke the program instructions to execute the functions of various modules in the above-mentioned device embodiments, for example Figure 2 The functions of the primary fault determination module 21, the secondary fault determination module 22, and the tertiary fault determination module 23 shown.

[0090] It should be understood that, in the embodiments of the present application, the processor 301 can be a central processing unit (CPU), and the processor can also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor.

[0091] The input device 302 can include a touchpad, a fingerprint collection sensor (for collecting fingerprint information and direction information of a fingerprint of a user), a microphone, etc., and the output device 303 can include a display (LCD, etc.), a speaker, etc.

[0092] The memory 304 can include read-only memory and random access memory, and provide instructions and data for the processor 301. A portion of the memory 304 can also include non-volatile random access memory. For example, the memory 304 can also store information of a device type of a substation.

[0093] In specific implementations, the processor 301, the input device 302, and the output device 303 described in the embodiments of the present application can execute the implementation manners described in the embodiments of the substation fault detection method provided by the embodiments of the present application, and can also execute the implementation manners of the electronic device 300 described in the embodiments of the present application, which will not be described here.

[0094] In another embodiment of the present application, a computer readable storage medium is provided, which stores a computer program. The computer program includes program instructions, which, when executed by a processor, implement all or part of the processes of the above-mentioned embodiment methods. The computer program can also instruct related hardware to complete the implementation. The computer program can be stored in a computer readable storage medium. When the computer program is executed by the processor, the steps of the above-mentioned various method embodiments can be implemented. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or some intermediate form. The computer readable medium can include any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal, and software distribution medium, etc.

[0095] The computer readable storage medium can be an internal storage unit of the electronic device of any of the preceding embodiments, such as a hard disk or a memory of the electronic device. The computer readable storage medium can also be an external storage device of the electronic device, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. Further, the computer readable storage medium can include both the internal storage unit and the external storage device of the electronic device. The computer readable storage medium is used to store the computer program and other programs and data required by the electronic device. The computer readable storage medium can also be used to temporarily store data that has been output or will be output.

[0096] Those skilled in the art can appreciate that the modules / units and algorithm steps of the examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. In order to clearly illustrate the interchangeability of hardware and software, the components and steps of the examples have been described in general terms in the above description. Whether the functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. A person skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0097] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the electronic device and the units described above can refer to the corresponding processes in the above-mentioned method embodiments, which will not be described here.

[0098] In several embodiments provided in the present application, it should be understood that the disclosed electronic device and method can be implemented in other manners. For example, the division of the above-described apparatus embodiments is merely an example, and there can be other division manners. For example, multiple modules or units can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the displayed or discussed mutual couplings or direct couplings or communication connections can be indirect couplings or communication connections through some interfaces, or can be in electrical, mechanical or other forms.

[0099] The modules or units illustrated as separate parts can or can not be physically separate, and the parts illustrated as modules or units can or can not be physical modules or units, i.e., can be located in one place, or can be distributed on multiple network modules or units. Some or all of the modules or units can be selected according to actual needs to achieve the purposes of the embodiments of the present application.

[0100] In addition, each functional module / unit in each embodiment of the present application can be integrated into one processing module / unit, or each module / unit can exist physically separately, or two or more modules / units can be integrated into one module / unit. The integrated module / unit can be implemented in the form of hardware or in the form of a software functional module / unit.

[0101] The above is merely specific embodiments of the present application, and the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of various equivalent modifications or replacements within the technical scope disclosed in the present application, and these modifications or replacements should be covered in the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A substation fault detection method characterized by, The method comprises the following steps: obtaining electrical quantity data and equipment operation data of a substation, performing first fault determination on the electrical quantity data and the equipment operation data based on a first determination condition to obtain a fault probability; the first determination condition comprises a threshold corresponding to the electrical quantity data and the equipment operation data respectively; if the fault probability is lower than a first fault probability threshold, extracting an electrical trend feature based on the electrical quantity data; extracting an equipment operation trend feature based on the equipment operation data; calculating a first fitting degree of the electrical trend feature and an electrical fault trend feature; calculating a second fitting degree of the equipment operation trend feature and an equipment operation fault trend feature; performing weighted summation on the first fitting degree and the second fitting degree to obtain a third fitting degree; if the first fitting degree is higher than a first fitting degree threshold or the second fitting degree is higher than a first fitting degree threshold, and the third fitting degree is higher than a second fitting degree threshold, determining that a second fault determination result is fault; if the first fitting degree is not higher than a first fitting degree threshold, the second fitting degree is not higher than a first fitting degree threshold, and the third fitting degree is higher than a second fitting degree threshold, determining that a second fault determination result is no fault; if the second fault determination result is fault, inputting the electrical quantity data and the equipment operation data into a fault detection model to generate a fault detection result; if the fault probability is not lower than the first fault probability threshold, inputting the electrical quantity data and the equipment operation data into a fault detection model to generate a fault detection result; the fault detection model is obtained by training based on a historical fault data set, and the historical fault data set comprises historical electrical quantity data, historical equipment operation data and a fault type label; the electrical quantity data comprises current data and voltage data in a first time period, and the equipment operation data comprises equipment temperature and equipment vibration frequency in the first time period; the first fault determination based on the first determination condition comprises: comparing the current data, the voltage data, the equipment temperature and the equipment vibration frequency in the first time period with the corresponding thresholds respectively; determining the fault probability according to the total number of times that the current data, the voltage data, the equipment temperature and the equipment vibration frequency in the first time period exceed the corresponding thresholds respectively; the fault probability is positively correlated with the total number of times.

2. A substation fault detection method as claimed in claim 1, characterized by, After the second fault determination on the electrical quantity data and the equipment operation data, the method further comprises the following steps: if the second fault determination result is no fault, updating the fault probability corresponding to the electrical quantity data and the equipment operation data; updating the first determination condition based on the electrical quantity data, the equipment operation data and the updated fault probability to obtain an updated first determination condition.

3. A substation fault detection method as claimed in claim 2, characterized by, The updating of the fault probability corresponding to the electrical quantity data and the equipment operation data comprises the following steps: updating the fault probability based on a first step; the updated fault probability is lower than the previous fault probability. If the updated failure probability is lower than the first failure probability threshold, then the first failure probability threshold is used as the updated failure probability.

4. A substation fault detection method as claimed in claim 3, characterized by, The step of updating the first determination condition based on the electrical quantity data, the equipment operation data, and the updated fault probability to obtain the updated first determination condition includes: Electrical trend features are extracted based on the electrical quantity data, and equipment operation trend features are extracted based on the equipment operation data. The electrical trend characteristics, the equipment operation trend characteristics, and the updated fault probability are added to the first determination condition to obtain the updated first determination condition.

5. A substation fault detection apparatus characterized by, include: A first-level fault determination module is used to acquire electrical quantity data and equipment operation data of the substation, and perform a first fault determination on the electrical quantity data and the equipment operation data based on a first determination condition to obtain the fault probability; the first determination condition includes the threshold corresponding to each of the electrical quantity data and the equipment operation data; the electrical quantity data includes current data and voltage data for a first time period, and the equipment operation data includes equipment temperature and equipment vibration frequency for a first time period; The first-level fault determination module is specifically used to compare the current data, voltage data, equipment temperature, and equipment vibration frequency in the first time period with their respective corresponding thresholds; and to determine the fault probability based on the total number of times the current data, voltage data, equipment temperature, and equipment vibration frequency in the first time period exceed their respective corresponding thresholds. The probability of failure is positively correlated with the total number of data occurrences. The secondary fault determination module is used to extract electrical trend features based on the electrical quantity data; extract equipment operation trend features based on the equipment operation data; calculate a first degree of fit between the electrical trend features and the electrical fault trend features; calculate a second degree of fit between the equipment operation trend features and the equipment operation fault trend features; and perform a weighted summation of the first degree of fit and the second degree of fit to obtain a third degree of fit. If the first degree of fit is higher than the first degree of fit threshold or the second degree of fit is higher than the first degree of fit threshold, and the third degree of fit is higher than the second degree of fit threshold, then the second fault determination result is determined to be a fault. If the first degree of fit is not higher than the first degree of fit threshold, the second degree of fit is not higher than the first degree of fit threshold, and the third degree of fit is higher than the second degree of fit threshold, then the second fault determination result is determined to be no fault. If the second fault determination result is a fault, then the electrical quantity data and the equipment operation data are input into the fault detection model to generate a fault detection result; The three-level fault determination module is used to input the electrical quantity data and the equipment operation data into the fault detection model to generate a fault detection result if the fault probability is not lower than the first fault probability threshold. The fault detection model is trained based on a historical fault dataset, which includes historical electrical quantity data, historical equipment operation data, and fault type labels.

6. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, The computer program is executed by the processor to implement the steps of the method according to any one of claims 1 to 4.

7. A computer-readable storage medium storing a computer program, wherein the computer program comprises the following steps of: receiving a request for a resource from a client; determining whether the client is authorized to access the resource; and if the client is authorized to access the resource, providing the resource to the client. The computer program is executed by the processor to implement the steps of the method according to any one of claims 1 to 4.

Images