Transformer fault diagnosis method and system based on sparrow algorithm, storage medium and electronic equipment

Abstract

The invention provides a transformer fault diagnosis method and system based on a sparrow algorithm, a storage medium and electronic equipment, and relates to the technical field of intelligent power grid and power system monitoring. A sparrow algorithm is used for constructing a fault diagnosis model, setting foraging and reconnaissance sparrow roles and optimizing positions, combining oil pressure, load current, temperature data and characteristic parameters, generating a fitness function, selecting an optimal parameter combination, collecting current data and the characteristic parameters, and comparing differences to generate fault indexes; and if so, determining that the transformer fails. According to the invention, rapid and accurate transformer fault detection is realized by acquiring and preprocessing transformer operation data, extracting fault related characteristics, constructing a diagnosis model by using a sparrow algorithm, and setting fitness evaluation and threshold judgment.

Description

Technical Field

[0001] The present invention relates to the field of smart grid and power system monitoring technology, and specifically to a transformer fault diagnosis method, system, storage medium and electronic equipment based on a sparrow algorithm. Background Art

[0002] Transformers are key equipment in power systems, and their operating status directly affects the stability and safety of power systems. Effective fault diagnosis methods can timely detect potential transformer faults and prevent further development of faults, thereby improving the reliability and safety of power systems. Through early fault diagnosis, power companies can carry out targeted preventive maintenance instead of passively repairing faults that have already occurred. This can not only significantly reduce maintenance costs, but also reduce power outage time, thereby reducing economic losses caused by power outages.

[0003] In the prior art, publication number CN116842837A discloses a transformer fault diagnosis method, system, electronic device and storage medium, which obtains transformer status performance data corresponding to the transformer to be diagnosed in real time; inputs the transformer status performance data into a pre-trained transformer fault classifier model to obtain a transformer fault diagnosis result; wherein the transformer fault classifier model is based on an optimized integrated support vector recursive neural network model algorithm; and feeds back the transformer fault diagnosis result to a user to achieve real-time monitoring of the transformer status.

[0004] Insufficient existing technology:

[0005] The transformer fault classifier model pre-trained in the prior art may be difficult to adapt to the various complex and dynamic conditions of the transformer in the actual operating environment, resulting in limited accuracy of the fault diagnosis results. The operating state and environment of the transformer may change, which reduces the prediction accuracy of the model and fails to reflect the actual situation in a timely manner. This lack of adaptability may lead to false positives or false negatives, affecting users' trust and reliance on the system.

[0006] In addition, real-time acquisition and processing of a large amount of transformer status performance data places high demands on the hardware and software performance of the system, which may lead to performance bottlenecks in the data processing process and affect the timeliness of fault diagnosis. There may be noise and outliers in the data. If they are not effectively preprocessed, these data will affect the diagnostic accuracy of the model, leading to fault identification errors or delays. This lack of real-time performance and processing capabilities may result in the failure to detect faults in a timely manner, thereby increasing the risk of transformer operation and possibly causing serious equipment damage and economic losses.

[0007] Therefore, it is necessary to provide a transformer fault diagnosis method, system, storage medium and electronic device based on the sparrow algorithm to solve the above problems.

[0008] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not constitute the prior art that is already known to one of ordinary skill in the art. Summary of the invention

[0009] The object of the present invention is to provide a transformer fault diagnosis method, system, storage medium and electronic device based on the sparrow algorithm to solve the problems raised in the above background technology.

[0010] To achieve the above object, the present invention provides the following technical solutions:

[0011] A transformer fault diagnosis method based on the sparrow algorithm, the specific steps include:

[0012] Step 1: Obtain historical operation data of the transformer, pre-process the acquired data, and remove noise and abnormal value data, wherein the historical operation data includes oil pressure, load current and temperature data of the transformer;

[0013] Step 2: extracting features from the processed operating data and calculating characteristic parameters related to transformer faults, wherein the characteristic parameters include the temperature rise rate, the concentration change rate of dissolved gas in oil, and the current change rate;

[0014] Step 3: Build a fault diagnosis model based on the sparrow algorithm, generate an initial sparrow population, define the roles of foraging sparrows and scout sparrows, determine the initial position, and iteratively optimize the initial position of each sparrow through the position update formula;

[0015] Step 4: Generate a fitness function based on the acquired oil pressure, load current, and temperature data and combine the currently measured temperature rise rate, the concentration change rate of dissolved gas in the oil, and the current change rate to select the best parameter combination;

[0016] Step 5: Collect current operating data and characteristic parameters, compare the difference between the current parameter combination and the optimal parameter combination, and generate fault indicators based on the difference;

[0017] Step 6: Set a fault threshold to determine whether the fault index exceeds the threshold. If it exceeds the fault threshold, it is determined that the transformer is faulty.

[0018] Furthermore, the acquired data is preprocessed to remove noise and outlier data, based on the following method:

[0019] Use sensors and monitoring equipment to collect transformer oil pressure, load current and temperature data in real time, and transmit the collected data to the data center for storage. Use a mean filter to filter the acquired data to remove noise in the data.

[0020] The triple standard deviation method is used to detect whether there are abnormalities in the denoised running data. The logical formula is as follows:

[0021]

[0022] Among them, Q represents the logical value for judging whether the running data after denoising is abnormal. When Q = 0, it indicates that the running data after denoising is abnormal, that is, y t Abnormal data are removed and the missing data are supplemented by the average interpolation method; when Q = 1, it indicates that there is no abnormality in the running data after denoising; μ represents the mean of the running data after denoising, σ represents the standard deviation of the running data after denoising, and y t Represents the historical operating data at time t.

[0023] Furthermore, feature extraction is performed on the processed operating data to calculate characteristic parameters related to transformer faults, based on the following formula:

[0024]

[0025]

[0026] Among them, V T (t), V C (t), V I (t) respectively represent the temperature increase rate, the concentration change rate of dissolved gas in oil, and the current change rate at time t, ΔT t Indicates the change from the temperature at time t-1 to the temperature at time t, ΔC t Indicates the change from the concentration at time t-1 to the concentration at time t, ΔI t Indicates the change in current from time t-1 to time t, Δt T , Δt C , Δt I They respectively represent the time change of their respective parameters from time t-1 to time t.

[0027] Furthermore, a fault diagnosis model is constructed to generate an initial sparrow population, define the roles of foraging sparrows and scout sparrows, and determine the initial position, based on the following method:

[0028] The size of the population and the dimensions of the feature combination are set, and the initial sparrow population is randomly generated. The position of each sparrow represents a feature parameter combination. The sparrow species are divided into foraging sparrows and scout sparrows. Foraging sparrows are used to explore the feature space, find better feature combinations, and determine the characteristic ranges of the temperature increase rate, the concentration change rate of dissolved gas in oil, and the current change rate. Scout sparrows are used to monitor the operating status and early warning, detect abnormal changes in parameters, and monitor data in real time according to the detected abnormalities. The formula is:

[0029]

[0030] Among them, X i represents the initial position of the i-th sparrow, They represent the temperature increase rate, the change rate of dissolved gas concentration in oil and the change rate of current in the position vector of the i-th sparrow, respectively. They represent the upper limits of the characteristic ranges of temperature rise rate, concentration change rate and current change rate, respectively. They represent the lower bounds of the characteristic ranges of temperature rise rate, concentration change rate, and current change rate, respectively. rand(0,1) represents a uniformly distributed random number randomly generated in the interval [0,1], which is used to randomly initialize characteristic parameters within the specified range.

[0031] Furthermore, the initial position of each sparrow is iteratively optimized through the position update formula, and the formula based on it is:

[0032]

[0033] Among them, X i (k) represents the position of the i-th sparrow at the t-th iteration, S represents the control variable, which ranges from [0,1] and is used to distinguish between foraging sparrows and scout sparrows. When S = 1, the formula represents the updating mechanism of the foraging sparrow position; when S = 0, the formula represents the foraging mechanism of the scout sparrow, α represents the control parameter, and D max represents the maximum number of iterations, β represents the learning factor, and X best (k) represents the global optimal position found in the kth iteration.

[0034] Furthermore, according to the acquired oil pressure, load current, temperature data and combined with the currently measured temperature rise rate, the concentration change rate of dissolved gas in oil and the current change rate, a fitness function is generated to select the best parameter combination. The formula is as follows:

[0035] F=β0*C+β1*T+β2*I+β3*V T +β4*V C +β5*δ I

[0036] Among them, F represents the fitness function, β0, β1, β2, β3, β4, and β5 represent weight coefficients, reflecting the contribution of each variable to the evaluation results, and C, T, I, and V T 、V C , δ I They represent the acquired data of oil pressure, temperature, load current, temperature rise rate, concentration change rate of gas dissolved in oil, and current change rate, respectively, and β5>β4>β3>β2>β1>β0;

[0037] Based on the sparrow algorithm, the current operating data and characteristic parameters are optimized to obtain the best parameter combination. The method is as follows:

[0038] Initialize the sparrow population, each sparrow is a vector containing a weight coefficient, and calculate its fitness value based on the acquired oil pressure, temperature, load current, temperature rise rate, concentration change rate of dissolved gas in oil and current change rate data, and perform minimization to obtain the best parameter combination.

[0039] Furthermore, the current operating data and characteristic parameters are collected, the difference between the current parameter combination and the optimal parameter combination is compared, and the fault index is generated according to the difference, and the formula is as follows:

[0040]

[0041] Among them, T1, C1, I1, V1 T 、V1 C ,δ1 I Indicates the oil pressure, temperature, load current, temperature rise rate, concentration change rate of dissolved gas in oil, current change rate data in the best parameter combination, T2, C2, I2, V2 T 、V2 C , δ2 I It indicates the oil pressure, temperature, load current, temperature rise rate, concentration change rate of dissolved gas in oil, and current change rate data in the current parameter combination. GZ indicates the fault indicator.

[0042] Furthermore, the logic formula for judging whether the transformer is faulty is as follows:

[0043]

[0044] Wherein, U represents a logical value for judging whether there is a fault in the transformer. When U=0, the fault index does not exceed the preset fault threshold, indicating that there is no fault; when U=1, the fault index exceeds the preset fault threshold, indicating that there is a fault.

[0045] The present invention further provides a transformer fault diagnosis system based on the sparrow algorithm, wherein the fault diagnosis system is used to execute the above-mentioned transformer fault diagnosis method based on the sparrow algorithm, comprising:

[0046] A data preprocessing module, the data preprocessing module is used to obtain historical operation data of the transformer, preprocess the obtained data, and remove noise and abnormal value data, the historical operation data includes oil pressure, load current and temperature data of the transformer;

[0047] A feature extraction module, the feature extraction module is used to extract features from the processed operating data and calculate feature parameters related to transformer faults, the feature parameters including the temperature rise rate, the concentration change rate of dissolved gas in oil, and the current change rate;

[0048] A model building module, wherein the model building module builds a fault diagnosis model based on the sparrow algorithm, generates an initial sparrow population, defines the roles of foraging sparrows and scout sparrows, determines an initial position, and iteratively optimizes the initial position of each sparrow through a position update formula;

[0049] A parameter evaluation module, which generates a fitness function and selects the best parameter combination based on the acquired oil pressure, load current, temperature data and the currently measured temperature rise rate, the concentration change rate of dissolved gas in oil, and the current change rate;

[0050] A fault indicator generation module, wherein the fault indicator generation module collects current operating data and characteristic parameters, compares the difference between the current parameter combination and the optimal parameter combination, and generates a fault indicator according to the difference;

[0051] A fault judgment module is used to set a fault threshold and judge whether the fault index exceeds the threshold. If it exceeds the fault threshold, it is determined that the transformer is faulty.

[0052] The present invention also provides a computer-readable storage medium for transformer fault diagnosis based on the sparrow algorithm, on which a computer program is stored. When the computer program is executed by a processor, the above-mentioned transformer fault diagnosis method based on the sparrow algorithm is implemented.

[0053] The present invention also provides a transformer fault diagnosis electronic device based on the sparrow algorithm, the electronic device comprising:

[0054] A memory storing a computer program;

[0055] A processor is communicatively connected to the memory and executes the transformer fault diagnosis method based on the sparrow algorithm when calling the computer program.

[0056] Compared with the prior art, the present invention has the following beneficial effects:

[0057] The present invention solves the problems of model adaptability and update challenges in existing fault diagnosis technologies by introducing the sparrow algorithm. By optimizing the initialization and iteration of the model, the diagnostic model can more flexibly adapt to the changes of the transformer under different operating environments. This adaptability improves the accuracy of fault diagnosis results, reduces false positives and false negatives, and increases users' trust in the system.

[0058] In terms of data processing and real-time performance, the present invention uses mean filtering and triple standard deviation method for data preprocessing to effectively remove noise and outliers. At the same time, the sparrow algorithm is used for feature extraction and fault threshold setting to ensure that the real-time processing of data is more efficient and accurate. This not only improves the timeliness of fault diagnosis, but also reduces the operating risk of the system and the possibility of equipment damage.

[0059] The present invention optimizes the transformer fault diagnosis model through the sparrow algorithm, improves the accuracy and real-time performance of the diagnosis, and enhances the adaptability and reliability of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 It is a schematic diagram of the overall method flow of the present invention.

[0061] Figure 2 It is a schematic diagram of the system module flow of the present invention. DETAILED DESCRIPTION

[0062] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with specific embodiments.

[0063] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the present invention should be understood by people with ordinary skills in the field to which the present invention belongs. The words "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0064] Example:

[0065] See also Figure 1, the present invention provides a technical solution:

[0066] A transformer fault diagnosis method based on the sparrow algorithm, the specific steps include:

[0067] Step 1: Obtain historical operation data of the transformer, pre-process the acquired data, and remove noise and abnormal value data, wherein the historical operation data includes oil pressure, load current and temperature data of the transformer;

[0068] Step 2: extracting features from the processed operating data and calculating characteristic parameters related to transformer faults, wherein the characteristic parameters include the temperature rise rate, the concentration change rate of dissolved gas in oil, and the current change rate;

[0069] Step 3: Build a fault diagnosis model based on the sparrow algorithm, generate an initial sparrow population, define the roles of foraging sparrows and scout sparrows, determine the initial position, and iteratively optimize the initial position of each sparrow through the position update formula;

[0070] Step 4: Generate a fitness function based on the acquired oil pressure, load current, and temperature data and combine the currently measured temperature rise rate, the concentration change rate of dissolved gas in the oil, and the current change rate to select the best parameter combination;

[0071] Step 5: Collect current operating data and characteristic parameters, compare the difference between the current parameter combination and the optimal parameter combination, and generate fault indicators based on the difference;

[0072] Step 6: Set a fault threshold to determine whether the fault index exceeds the threshold. If it exceeds the fault threshold, it is determined that the transformer is faulty.

[0073] It should be noted that data preprocessing removes noise through a mean filter to ensure that the data input to the fault diagnosis model is purer and more accurate. High-quality data is the basis for the model to make accurate predictions. The triple standard deviation method is a commonly used statistical method for detecting outliers in data. It is based on the statistical characteristics of data distribution. For most normally distributed data, the data within the range of three times the standard deviation accounts for more than 99.7%, so it can effectively detect outliers. By detecting and removing outliers through the triple standard deviation method, the interference of abnormal data on model training and prediction can be avoided, thereby improving the accuracy of fault diagnosis. Abnormal data often causes false positives or false negatives, and removing these data can significantly reduce the misjudgment rate and reduce interference with the model.

[0074] Therefore, it is necessary to preprocess the acquired data to remove noise and outlier data. The method is as follows:

[0075] Use sensors and monitoring equipment to collect transformer oil pressure, load current and temperature data in real time, and transmit the collected data to the data center for storage. Use a mean filter to filter the acquired data to remove noise in the data.

[0076] The triple standard deviation method is used to detect whether there are abnormalities in the denoised running data. The logical formula is as follows:

[0077]

[0078] Among them, Q represents the logical value for judging whether the running data after denoising is abnormal. When Q = 0, it indicates that the running data after denoising is abnormal, that is, y t Abnormal data are removed and the missing data are supplemented by the average interpolation method; when Q = 1, it indicates that there is no abnormality in the running data after denoising; μ represents the mean of the running data after denoising, σ represents the standard deviation of the running data after denoising, and y t Represents the historical operating data at time t.

[0079] It should be noted that by simplifying the data dimensions and extracting key features related to faults from a large amount of operating data, the complexity of the data can be greatly simplified, making subsequent analysis and model training more efficient. In addition, by extracting key characteristic parameters such as the temperature rise rate, the rate of change of the concentration of dissolved gas in the oil, and the abnormal rate of change of the current waveform, the abnormal operating state of the transformer can be more accurately identified, thereby improving the accuracy and reliability of fault diagnosis.

[0080] Therefore, it is necessary to extract features from the processed operating data and calculate the characteristic parameters related to transformer faults based on the following formula:

[0081]

[0082] Among them, V T (t), V C (t), V I (t) respectively represent the temperature increase rate, the concentration change rate of dissolved gas in oil, and the current change rate at time t, ΔT t Indicates the change from the temperature at time t-1 to the temperature at time t, ΔC t Indicates the change from the concentration at time t-1 to the concentration at time t, ΔI t Indicates the change in current from time t-1 to time t, Δt T , Δt C , Δt I They respectively represent the time change of their respective parameters from time t-1 to time t.

[0083] It should be noted that by simulating the foraging behavior of sparrows, the optimal feature combination can be explored in the feature space, thereby improving the accuracy of fault diagnosis. This optimization process can find the best feature parameter combination to help the model identify fault modes more effectively. The monitoring role of the scout sparrow can help the model adjust in real time, adjust the feature parameters according to the changes in the operating state, and enhance the model's adaptability to dynamic data. Through the reasonable setting of the initial position and role, the convergence speed of the algorithm is accelerated, making fault detection more efficient.

[0084] Therefore, it is necessary to build a fault diagnosis model, generate an initial sparrow population, define the roles of foraging sparrows and scout sparrows, and determine the initial position. The method is based on:

[0085] The size of the population and the dimensions of the feature combination are set, and the initial sparrow population is randomly generated. The position of each sparrow represents a feature parameter combination. The sparrow species are divided into foraging sparrows and scout sparrows. Foraging sparrows are used to explore the feature space, find better feature combinations, and determine the characteristic ranges of the temperature increase rate, the concentration change rate of dissolved gas in oil, and the current change rate. Scout sparrows are used to monitor the operating status and early warning, detect abnormal changes in parameters, and monitor data in real time according to the detected abnormalities. The formula is:

[0086]

[0087] Among them, X i represents the initial position of the i-th sparrow, They represent the temperature increase rate, the change rate of dissolved gas concentration in oil and the change rate of current in the position vector of the i-th sparrow, respectively. They represent the upper limits of the characteristic ranges of temperature rise rate, concentration change rate and current change rate, respectively. They represent the lower bounds of the characteristic ranges of the temperature rise rate, concentration change rate, and current change rate, respectively. rand(0,1) represents a uniformly distributed random number randomly generated in the interval [0,1], which is used to randomly initialize characteristic parameters within the specified range. In the above formula, the diversity of the initial population is ensured by randomly initializing the characteristic parameter combination, which helps to explore a wider characteristic space and avoid falling into the local optimum. The foraging sparrow focuses on exploring the optimal solution in the characteristic space, simulating the foraging behavior in nature, and improving the optimization ability of the characteristic combination. The scout sparrow is used to monitor the operating status, detect anomalies in time, warn of potential failures, and ensure the safety of the system.

[0088] It should be noted that the position update formula can effectively promote the sparrow to approach the global optimal solution in the feature space through the iterative optimization mechanism, thereby improving the accuracy of the fault diagnosis model. By dynamically adjusting the position of the sparrow, the model can converge to the optimal solution more quickly, reduce the calculation time, and improve the diagnosis efficiency. In addition, by controlling the setting of the variable s, the optimization strategies of different roles are realized, which increases the flexibility and diversity of the algorithm.

[0089] Therefore, it is necessary to iteratively optimize the initial position of each sparrow through the position update formula, based on the following formula:

[0090]

[0091] Among them, X i (k) represents the position of the i-th sparrow at the t-th iteration, S represents the control variable, which ranges from [0,1] and is used to distinguish between foraging sparrows and scout sparrows. When s = 1, the formula represents the update mechanism of the foraging sparrow position; when S = 0, the formula represents the foraging mechanism of the scout sparrow, α represents the control parameter, and D max represents the maximum number of iterations, β represents the learning factor, and X best (k) represents the global optimal position found in the kth iteration.

[0092] It should be noted that the fitness function is the key objective function in the optimization process. It reveals the importance of each factor to the transformer operating status by comprehensively evaluating oil pressure, temperature, load current and other key indicators. Optimizing the weight coefficient can improve the accuracy of fault diagnosis and help timely warn and prevent potential risks. As the core of the optimization algorithm, the fitness function ensures the optimization of parameter combinations, so that the system maintains efficient operation and management capabilities under different conditions.

[0093] Therefore, it is necessary to generate a fitness function based on the acquired oil pressure, load current, temperature data and the currently measured temperature rise rate, the concentration change rate of dissolved gas in oil and the current change rate to select the best parameter combination. The formula is as follows:

[0094] F=β0*C+β1*T+β2*I+β3*V T +β4*V C ++β5*δ I

[0095] Among them, F represents the fitness function, β0, β1, β2, β3, β4, and β5 represent weight coefficients, reflecting the contribution of each variable to the evaluation results, and C, T, I, and V T 、V C , δ IThey represent the acquired oil pressure, temperature, load current, temperature rise rate, concentration change rate of dissolved gas in oil, and current change rate data respectively, and β5>β4>β3>β2>β1>β0. The weights are set in this way to reflect the differences in the importance of each variable to the fault assessment results. According to the formula, weight β5 corresponds to the current change rate, and its value is the largest, indicating that the current change rate has the greatest impact on fault assessment; followed by the concentration change rate of dissolved gas in oil β4, and the temperature rise rate β3, which are generally considered to be more direct and sensitive fault indicators. The load current β2, temperature β1, and oil pressure β0 decrease in turn, indicating that their relative importance is relatively low and belong to the influence of environmental factors on the transformer. This weight distribution helps to identify faults more accurately;

[0096] Based on the sparrow algorithm, the current operating data and characteristic parameters are optimized to obtain the best parameter combination. The method is as follows:

[0097] Initialize the sparrow population, each sparrow is a vector containing a weight coefficient, and calculate its fitness value based on the acquired oil pressure, temperature, load current, temperature rise rate, concentration change rate of dissolved gas in oil and current change rate data, and perform minimization to obtain the best parameter combination.

[0098] It should be noted that by comparing the difference between the current operating data and the optimal parameter combination, the deviation of the transformer state can be quantified. The specific formula GZ generates a comprehensive fault index by calculating the difference between the oil pressure, temperature, load current and their optimal values, as well as the temperature rise rate, the concentration change rate of dissolved gas in the oil, and the current change rate and their optimal values. This indicator can quickly and accurately reflect the abnormality of the transformer operation state and is used for timely warning and maintenance.

[0099] Therefore, it is necessary to collect the current operating data and characteristic parameters, compare the difference between the current parameter combination and the optimal parameter combination, and generate the fault index according to the difference. The formula is as follows:

[0100]

[0101] Among them, T1, C1, I1, V1 T 、V1 C ,δ1 I Indicates the oil pressure, temperature, load current, temperature rise rate, concentration change rate of dissolved gas in oil, current change rate data in the best parameter combination, T2, C2, I2, V2 T 、V2 C , δ2 IIt indicates the oil pressure, temperature, load current, temperature rise rate, concentration change rate of dissolved gas in oil, and current change rate data in the current parameter combination. GZ indicates the fault indicator.

[0102] It should be noted that the logic formula U clearly defines the fault state and the normal state through simple conditional judgment, and the preset threshold is set according to the actual operation data and fault characteristics, ensuring that potential faults can be warned in time when the fitness value exceeds the threshold. By judging whether the fitness value meets the preset conditions, system faults can be accurately identified and warned. The reasonable setting of the logic formula U and the fitness threshold G ensures the accuracy and timeliness of fault detection and optimizes the resource allocation and maintenance efficiency of the system. This step is an important link in achieving efficient and accurate fault diagnosis.

[0103] Therefore, it is necessary to judge whether the fitness value reaches the preset conditions, and the logical formula is as follows:

[0104]

[0105] Wherein, U represents a logical value for judging whether there is a fault in the transformer. When U=0, the fitness value does not exceed the preset fault threshold, indicating that there is no fault; when U=1, the fitness value exceeds the preset fault threshold, indicating that there is a fault.

[0106] The present invention further provides a transformer fault diagnosis system based on the sparrow algorithm, wherein the fault diagnosis system is used to execute the above-mentioned transformer fault diagnosis method based on the sparrow algorithm, comprising:

[0107] A data preprocessing module, the data preprocessing module is used to obtain historical operation data of the transformer, preprocess the obtained data, and remove noise and abnormal value data, the historical operation data includes oil pressure, load current and temperature data of the transformer;

[0108] A feature extraction module, the feature extraction module is used to extract features from the processed operating data and calculate feature parameters related to transformer faults, the feature parameters including the temperature rise rate, the concentration change rate of dissolved gas in oil, and the current change rate;

[0109] A model building module, wherein the model building module builds a fault diagnosis model based on the sparrow algorithm, generates an initial sparrow population, defines the roles of foraging sparrows and scout sparrows, determines an initial position, and iteratively optimizes the initial position of each sparrow through a position update formula;

[0110] A parameter evaluation module, which generates a fitness function and selects the best parameter combination based on the acquired oil pressure, load current, temperature data and the currently measured temperature rise rate, the concentration change rate of dissolved gas in oil, and the current change rate;

[0111] A fault indicator generation module, wherein the fault indicator generation module collects current operating data and characteristic parameters, compares the difference between the current parameter combination and the optimal parameter combination, and generates a fault indicator according to the difference;

[0112] A fault judgment module is used to set a fault threshold and judge whether the fault index exceeds the threshold. If it exceeds the fault threshold, it is determined that the transformer is faulty.

[0113] The present invention also provides a computer-readable storage medium for transformer fault diagnosis based on the sparrow algorithm, on which a computer program is stored. When the computer program is executed by a processor, the above-mentioned transformer fault diagnosis method based on the sparrow algorithm is implemented.

[0114] The present invention also provides a transformer fault diagnosis electronic device based on the sparrow algorithm, the electronic device comprising:

[0115] A memory storing a computer program;

[0116] A processor is communicatively connected to the memory and executes the transformer fault diagnosis method based on the sparrow algorithm when calling the computer program.

[0117] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0118] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination thereof. When implemented by software, the above embodiments may be implemented in whole or in part in the form of a computer program product. Those skilled in the art may appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein may be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed by hardware or software methods depends on the specific application and design constraints of the technical solution.

[0119] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, and may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0120] The above description is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application.

Claims

1. A transformer fault diagnosis method based on the sparrow algorithm, characterized in that: The specific steps include: Step 1: Obtain historical operation data of the transformer, pre-process the acquired data, and remove noise and abnormal value data, wherein the historical operation data includes oil pressure, load current and temperature data of the transformer; Step 2: extracting features from the processed operating data and calculating characteristic parameters related to transformer faults, wherein the characteristic parameters include the temperature rise rate, the concentration change rate of dissolved gas in oil, and the current change rate; Step 3: Build a fault diagnosis model based on the sparrow algorithm, generate an initial sparrow population, define the roles of foraging sparrows and scout sparrows, determine the initial position, and iteratively optimize the initial position of each sparrow through the position update formula; Step 4: Generate a fitness function based on the acquired oil pressure, load current, and temperature data and combine the currently measured temperature rise rate, the concentration change rate of dissolved gas in the oil, and the current change rate to select the best parameter combination; Step 5: Collect current operating data and characteristic parameters, compare the difference between the current parameter combination and the optimal parameter combination, and generate fault indicators based on the difference; Step 6: Set a fault threshold to determine whether the fault index exceeds the threshold. If it exceeds the fault threshold, it is determined that the transformer is faulty.

2. A transformer fault diagnosis method based on the sparrow algorithm according to claim 1, characterized in that: The acquired data is preprocessed to remove noise and outlier data, based on the following method: Use sensors and monitoring equipment to collect transformer oil pressure, load current and temperature data in real time, and transmit the collected data to the data center for storage. Use a mean filter to filter the acquired data to remove noise in the data. The triple standard deviation method is used to detect whether there are abnormalities in the denoised running data. The logical formula is as follows: Among them, Q represents the logical value for judging whether there is an abnormality in the running data after denoising. When Q = 0, it indicates that there is an abnormality in the running data after denoising, that is, y t Abnormal data are removed and the missing data are supplemented by the average interpolation method; when Q = 1, it indicates that there is no abnormality in the running data after denoising; μ represents the mean of the running data after denoising, σ represents the standard deviation of the running data after denoising, and y t Represents the historical operating data at time t.

3. A transformer fault diagnosis method based on sparrow algorithm according to claim 2, characterized in that: The processed operating data is subjected to feature extraction to calculate the characteristic parameters related to transformer faults, based on the following formula: Among them, V T (t), V C (t), V I (t) respectively represent the temperature increase rate, the concentration change rate of dissolved gas in oil, and the current change rate at time t, ΔT t Indicates the change from the temperature at time t-1 to the temperature at time t, ΔC t Indicates the change from the concentration at time t-1 to the concentration at time t, ΔI t Indicates the change in current from time t-1 to time t, Δt T , Δt C , Δt I They respectively represent the time change of their respective parameters from time t-1 to time t.

4. A transformer fault diagnosis method based on the sparrow algorithm according to claim 3, characterized in that: The fault diagnosis model is constructed to generate the initial sparrow population, define the roles of foraging sparrows and scout sparrows, and determine the initial position based on the following method: The size of the population and the dimensions of the feature combination are set, and the initial sparrow population is randomly generated. The position of each sparrow represents a feature parameter combination. The sparrow species are divided into foraging sparrows and scout sparrows. Foraging sparrows are used to explore the feature space, find better feature combinations, and determine the characteristic ranges of the temperature increase rate, the concentration change rate of dissolved gas in oil, and the current change rate. Scout sparrows are used to monitor the operating status and early warning, detect abnormal changes in parameters, and monitor data in real time according to the detected abnormalities. The formula is: Among them, X i represents the initial position of the i-th sparrow, They represent the temperature increase rate, the change rate of dissolved gas concentration in oil and the change rate of current in the position vector of the i-th sparrow, respectively. They represent the upper limits of the characteristic ranges of temperature rise rate, concentration change rate and current change rate, respectively. They represent the lower bounds of the characteristic ranges of the temperature rise rate, concentration change rate, and current change rate, respectively. rand(0,1) represents a uniformly distributed random number randomly generated in the interval [0,1], which is used to randomly initialize characteristic parameters within the specified range. The initial position of each sparrow is iteratively optimized through the position update formula, based on the formula: Among them, X i (k) represents the position of the i-th sparrow at the t-th iteration, S represents the control variable, which ranges from [0,1] and is used to distinguish between foraging sparrows and scout sparrows. When S = 1, the formula represents the updating mechanism of the foraging sparrow position; when S = 0, the formula represents the foraging mechanism of the scout sparrow, α represents the control parameter, and D max represents the maximum number of iterations, β represents the learning factor, and X best (k) represents the global optimal position found in the tth iteration.

5. The transformer fault diagnosis method based on the sparrow algorithm according to claim 1 is characterized in that: Based on the acquired oil pressure, load current, and temperature data and combined with the currently measured temperature rise rate, the concentration change rate of dissolved gas in the oil, and the current change rate, a fitness function is generated to select the best parameter combination. The formula is as follows: F=β0*C+β1*T+β2*I+β3*V T +β4*V C +β5*δ I Among them, F represents the fitness function, β0, β1, β2, β3, β4, and β5 represent weight coefficients, reflecting the contribution of each variable to the evaluation results, and C, T, I, and V T 、V C , δ I They represent the acquired data of oil pressure, temperature, load current, temperature rise rate, concentration change rate of gas dissolved in oil, and current change rate, respectively, and β5>β4>β3>β2>β1>β0; Based on the sparrow algorithm, the current operating data and characteristic parameters are optimized to obtain the best parameter combination. The method is as follows: Initialize the sparrow population. Each sparrow is a vector containing a weight coefficient. Calculate its fitness value based on the acquired oil pressure, temperature, load current, temperature rise rate, concentration change rate of dissolved gas in oil and current change rate data. Minimize the fitness value as the optimization goal to obtain the best parameter combination.

6. A transformer fault diagnosis method based on the sparrow algorithm according to claim 7, characterized in that: Collect the current operating data and characteristic parameters, compare the difference between the current parameter combination and the optimal parameter combination, and generate fault indicators based on the difference. The formula is: Among them, T1, C1, I1, V1 T 、V1 C ,δ1 I Indicates the oil pressure, temperature, load current, temperature rise rate, concentration change rate of dissolved gas in oil, current change rate data in the best parameter combination, Y2, C2, I2, V2 T 、V2 C , δ2 I It indicates the oil pressure, temperature, load current, temperature rise rate, concentration change rate of dissolved gas in oil, and current change rate data in the current parameter combination. GZ indicates the fault indicator.

7. A transformer fault diagnosis method based on the sparrow algorithm according to claim 6, characterized in that: The logical formula for judging whether the transformer is faulty is: Wherein, U represents a logical value for judging whether there is a fault in the transformer. When U=0, the fault index does not exceed the preset fault threshold, indicating that there is no fault; when U=1, the fault index exceeds the preset fault threshold, indicating that there is a fault.

8. A transformer fault diagnosis system based on the sparrow algorithm, characterized in that: The fault diagnosis system is used to execute the transformer fault diagnosis method based on the sparrow algorithm according to any one of claims 1 to 7, comprising: A data preprocessing module, the data preprocessing module is used to obtain historical operation data of the transformer, preprocess the obtained data, and remove noise and abnormal value data, the historical operation data includes oil pressure, load current and temperature data of the transformer; A feature extraction module, the feature extraction module is used to extract features from the processed operating data and calculate feature parameters related to transformer faults, the feature parameters including the temperature rise rate, the concentration change rate of dissolved gas in oil, and the current change rate; A model building module, wherein the model building module builds a fault diagnosis model based on the sparrow algorithm, generates an initial sparrow population, defines the roles of foraging sparrows and scout sparrows, determines an initial position, and iteratively optimizes the initial position of each sparrow through a position update formula; A parameter evaluation module, which generates a fitness function and selects the best parameter combination based on the acquired oil pressure, load current, temperature data and the currently measured temperature rise rate, the concentration change rate of dissolved gas in oil, and the current change rate; A fault indicator generation module, wherein the fault indicator generation module collects current operating data and characteristic parameters, compares the difference between the current parameter combination and the optimal parameter combination, and generates a fault indicator according to the difference; A fault judgment module is used to set a fault threshold and judge whether the fault index exceeds the threshold. If it exceeds the fault threshold, it is determined that the transformer is faulty.

9. A computer-readable storage medium for transformer fault diagnosis based on the sparrow algorithm, on which a computer program is stored, characterized in that: When the computer program is executed by a processor, the transformer fault diagnosis method based on the sparrow algorithm described in any one of claims 1 to 9 is implemented.

10. A transformer fault diagnosis electronic device based on the sparrow algorithm, characterized in that: The electronic device comprises: A memory storing a computer program; A processor is communicatively connected to the memory, and executes the transformer fault diagnosis method based on the sparrow algorithm according to any one of claims 1 to 9 when calling the computer program.

Images