A circuit breaker mechanical state non-intervention multi-dimensional perception optimal arrangement method

Abstract

The application discloses a kind of circuit breaker mechanical state non-interventional multidimensional perception optimal arrangement method, it is related to circuit breaker optimal arrangement technical field, to solve the problem that when arranging circuit breaker, there is no targeted arrangement according to the actual situation of circuit breaker.The confirmation of the arrangement position of circuit breaker is automatically carried out by the constructed bayesian network and performance influence relationship, the subjectivity and uncertainty of artificial decision are reduced, the multi-dimensional characteristic data of circuit breaker and performance evaluation data are combined, the influence relationship between data is analyzed using multivariate statistical method, the objectivity and accuracy of decision are improved, different types of data are extracted using special feature extraction method, to ensure that the key information of each data type can be effectively captured.This targeted feature extraction method helps to improve the accuracy and efficiency of subsequent analysis, through in-depth analysis of multi-dimensional characteristic data, early diagnosis of circuit breaker mechanical state and potential fault prediction are realized.

Description

Technical Field

[0001] This invention relates to the field of optimal circuit breaker arrangement technology, specifically a non-intrusive multi-dimensional sensing optimal arrangement method for the mechanical state of circuit breakers. Background Technology

[0002] Optimal circuit breaker placement refers to the rational arrangement of the location and configuration of circuit breakers in a power system to achieve the best system performance and operating efficiency.

[0003] Chinese patent CN116298844B discloses a semi-dynamic substation high-voltage circuit breaker condition monitoring system and method. It primarily achieves long-term differentiated monitoring of some circuit breakers by applying statically arranged circuit breaker online monitoring devices based on different principles. On the other hand, it uses dynamically arranged circuit breaker latent fault monitoring devices to perform non-long-term (several days to several months) circuit breaker defect monitoring for circuit breakers requiring urgent monitoring (those with unresolved family defects or those whose maintenance cycles have exceeded the limit). While the above patent solves the data monitoring problem, the following issues still exist in practical operation:

[0004] 1. When acquiring mechanical status data of circuit breakers, the data acquisition is inaccurate due to imperfect data collection methods.

[0005] 2. The mechanical condition data of the acquired circuit breakers was not subjected to targeted data feature extraction, which led to inaccurate analysis results in the later data analysis.

[0006] 3. The circuit breaker placement was not optimized and the model was not built based on the placement location, which made it inconvenient to adjust the circuit breakers later. Summary of the Invention

[0007] The purpose of this invention is to provide a non-intrusive, multi-dimensional sensing optimal layout method for circuit breaker mechanical conditions. By constructing a Bayesian network and analyzing performance impact relationships, the method automatically confirms the circuit breaker placement, reducing the subjectivity and uncertainty of human decision-making. Combining multi-dimensional feature data and performance evaluation data, and utilizing multivariate statistical methods to analyze the influence relationships between data, improves the objectivity and accuracy of decision-making. Specialized feature extraction methods are employed for different types of data to ensure that key information of each data type is effectively captured. This targeted feature extraction approach helps improve the accuracy and efficiency of subsequent analysis. Through in-depth analysis of multi-dimensional feature data, early diagnosis of the circuit breaker's mechanical condition and prediction of potential faults can be achieved, thus solving problems in existing technologies.

[0008] To achieve the above objectives, the present invention provides the following technical solution:

[0009] A non-intrusive multi-dimensional sensing optimal arrangement method for the mechanical state of circuit breakers includes the following steps:

[0010] S1: Circuit breaker data acquisition: Monitor and acquire the original mechanical status data of the circuit breaker, and mark the acquired original mechanical status data of the circuit breaker as status data to be processed;

[0011] S2: Circuit breaker data analysis: The status data to be processed is preprocessed. After the data preprocessing is completed, the index data of the circuit breaker is extracted. After the index data is extracted, the multi-dimensional feature data of the circuit breaker is obtained.

[0012] S3: Circuit breaker optimization layout: Based on the data of the circuit breakers to be arranged, the positions of the circuit breakers are arranged and a three-dimensional model is constructed to obtain the standard circuit breaker position data;

[0013] S4: Circuit Breaker Data Evaluation: Real-time monitoring data of circuit breakers in standard circuit breaker location data is collected, and the collected real-time monitoring data is evaluated. The effectiveness of the circuit breaker location arrangement is judged based on the evaluation results.

[0014] Preferably, the monitoring and acquisition of the original mechanical state data of the circuit breaker in S1 includes:

[0015] The mechanical condition data of a circuit breaker includes temperature data, acoustic data, vibration data, electrical data, optical data, gas analysis data, environmental data, and mechanical characteristic data.

[0016] The data includes temperature data (infrared thermal imaging and resistance temperature data), acoustic data (acoustic emission signals and sound spectrum analysis), vibration data (vibration waveforms and spectrum analysis), electrical data (partial discharge signals, current and voltage waveforms), optical data (visual images and optical sensors), gas analysis data (SF6 gas composition analysis), environmental data (humidity and temperature), and mechanical characteristic data (operating time and operating force).

[0017] Preferably, the monitoring and acquisition of the original mechanical state data of the circuit breaker in S1 further includes: monitoring the mechanical state data of the circuit breaker using a non-invasive sensor;

[0018] Specifically, temperature data is monitored using a non-contact infrared thermal imager and temperature sensor; acoustic data is monitored using an acoustic emission sensor and an external microphone; vibration data is monitored using a MEMS accelerometer; electrical data is monitored using a high-frequency current transformer and voltage transformer; optical data is monitored using a high-definition camera and laser rangefinder; gas analysis data is monitored using a gas sampler; environmental data is monitored using temperature and humidity sensors; and mechanical property data is monitored using displacement sensors and force sensors.

[0019] The monitored temperature data, acoustic data, vibration data, electrical data, optical data, gas analysis data, environmental data, and mechanical characteristic data are uniformly labeled as data to be processed.

[0020] Preferably, in step S2, the status data to be processed is preprocessed, and after the data preprocessing is completed, the indicator data of the circuit breaker is extracted, including:

[0021] Perform data preprocessing on the status data to be processed;

[0022] The data preprocessing process includes: denoising, calibrating, normalizing, and interpolating the temperature data in the data to be processed; denoising, enhancing, and synchronizing the acoustic data in the data to be processed; denoising, calibrating, and reducing the dimensionality of the vibration data in the data to be processed; denoising, enhancing, and synchronizing the electrical data in the data to be processed; enhancing and denoising the optical data in the data to be processed; correcting and denoising the gas analysis data in the data to be processed; denoising and calibrating the environmental data in the data to be processed; and denoising and calibrating the mechanical property data in the data to be processed.

[0023] Finally, we obtain the preprocessed data and the data awaiting processing.

[0024] Preferably, the acoustic data in the state data to be processed is subjected to sound enhancement processing, including:

[0025] Extract the acoustic data from the state data to be processed;

[0026] Extract a preset initial volume threshold from the database; wherein the initial volume threshold is not lower than a preset minimum volume value that can meet data quality requirements;

[0027] A volume threshold is set based on the acoustic data in the data to be processed, combined with an initial volume threshold, wherein the volume threshold is obtained by the following formula:

[0028]

[0029] Where F represents the volume threshold, and when the volume threshold is lower than the preset minimum volume value that can meet the data quality requirements, F = 1.08F. x And, F x This indicates the preset minimum volume level required to achieve the desired data quality; F yc The initial volume threshold is represented by ; n represents the number of frames contained in the acoustic data of the state data to be processed; m represents the number of frames in the acoustic data of the state data to be processed that are below the initial volume threshold; Fi F represents the volume value corresponding to the i-th frame of acoustic data in the state data to be processed; i This represents the intermediate volume value corresponding to n frames of acoustic data in the state data to be processed; F max and F min This represents the maximum and minimum volume values ​​corresponding to n frames of acoustic data in the state data to be processed; F b This represents the standard deviation of volume corresponding to n frames of acoustic data in the state data to be processed.

[0030] The acoustic data in the state data to be processed is compared with the volume threshold.

[0031] Extract acoustic data below the volume threshold and perform volume enhancement processing on the acoustic data below the volume threshold.

[0032] Preferably, the acoustic data below the volume threshold is extracted, and the acoustic data below the volume threshold is subjected to volume enhancement processing, including:

[0033] Extract acoustic data below the volume threshold;

[0034] Extract the preset minimum volume value that meets data quality requirements;

[0035] The first volume adjustment coefficient corresponding to the acoustic data below the volume threshold is obtained by using the volume value corresponding to the acoustic data below the volume threshold and the preset minimum volume value that can meet the data quality requirements.

[0036] The first volume adjustment coefficient is obtained by the following formula:

[0037]

[0038] Among them, f 01 Indicates the first volume adjustment coefficient; k represents the number of frames corresponding to acoustic data below the volume threshold; F j This represents the volume value corresponding to the acoustic data in frame j that is below the volume threshold; F represents the volume threshold; F yc Indicates the initial volume threshold; F x This indicates the minimum volume value preset to meet data quality requirements;

[0039] A second volume adjustment coefficient is set for the acoustic data below the volume threshold using the quantitative relationship between the volume value corresponding to the acoustic data below the volume threshold and the volume threshold; wherein, the second volume adjustment coefficient is obtained by the following formula:

[0040]

[0041] Among them, f 02 Indicates the second volume adjustment coefficient; k represents the number of frames corresponding to acoustic data below the volume threshold; F j This represents the volume value corresponding to the acoustic data in frame j that is below the volume threshold; F represents the volume threshold; F x This represents the preset minimum volume value that meets data quality requirements; c represents the adjustment factor, which is obtained using the following formula:

[0042]

[0043] Where c represents the adjustment factor; F min This represents the minimum volume value corresponding to n frames of acoustic data in the state data to be processed; F m This represents the maximum volume in the acoustic data of frames k that are below the volume threshold; F p This represents the average volume of acoustic data in k frames that are below the volume threshold;

[0044] The volume values ​​corresponding to the acoustic data below the volume threshold are adjusted using the first volume adjustment coefficient and the second volume adjustment coefficient to obtain the adjusted acoustic data; wherein, the volume value of the adjusted acoustic data is obtained by the following formula:

[0045]

[0046] Among them, F t This represents the adjusted acoustic data volume value corresponding to each frame of acoustic data below the stated volume threshold; f 01 f represents the first volume adjustment factor; 02 The value represents the second volume adjustment coefficient; F represents the volume threshold; F0 represents the volume value of the acoustic data before adjustment corresponding to the acoustic data below the volume threshold in each frame.

[0047] Preferably, the process of preprocessing the status data to be processed in step S2, followed by extracting the circuit breaker's index data after the preprocessing is complete, further includes:

[0048] Extract data features from the preprocessed data to be processed state data;

[0049] During data feature extraction, features are extracted based on the data attributes of the state data to be processed.

[0050] Among them, feature extraction of temperature data involves hot spot detection, temperature gradient calculation, temperature difference identification, and time series identification of infrared thermal imaging data; and temperature change trend analysis and outlier identification of resistance temperature data.

[0051] Feature extraction from acoustic data involves frequency and signal energy identification of acoustic emission signals; sound spectrum analysis involves spectral feature identification and time-frequency identification.

[0052] Vibration data feature extraction involves performing time-domain feature identification, frequency-domain feature identification, and time-frequency feature identification on vibration waveforms and spectrum analysis.

[0053] The feature extraction of electrical data involves identifying the number of discharges, the phase of discharge, and the energy of discharge in partial discharge signals; and identifying the waveform features and frequency features of current and voltage waveforms.

[0054] Feature extraction from optical data involves recognizing the shape, texture, and color features of visual images; and distance measurement and recognition using optical sensors.

[0055] Feature extraction of gas analysis data involves the identification of concentration features and concentration change features;

[0056] Feature extraction of environmental data involves the identification of statistical and correlational features;

[0057] Feature extraction from mechanical property data is used to identify statistical features and operating patterns;

[0058] After feature extraction of the state data to be processed, feature selection is performed on the extracted features. The feature selection uses principal component analysis to calculate the principal components of the extracted features.

[0059] After principal component calculation, the index data in the state data to be processed is obtained and labeled as multidimensional feature data of the circuit breaker.

[0060] Preferably, in step S3, the circuit breakers are positioned according to the data of the circuit breakers to be arranged, and a three-dimensional model is constructed, including:

[0061] The circuit breaker's performance is evaluated based on its multidimensional characteristic data. The performance evaluation includes contact performance evaluation, insulator performance evaluation, operating mechanism performance evaluation, and mechanical condition evaluation.

[0062] A Bayesian network is constructed based on the performance evaluation results and the location layout rules, wherein the location layout rules are obtained from the database.

[0063] Once the Bayesian network is constructed, each node and each edge in the Bayesian network is identified.

[0064] Each node represents an option point for the performance parameters and location layout of the circuit breaker; each edge represents an associated edge for the performance parameters and location layout.

[0065] The performance impact of multidimensional characteristic data and performance evaluation data of circuit breakers is judged, and the influence relationship between multidimensional characteristic data and performance evaluation data of circuit breakers is analyzed by multivariate statistical method.

[0066] The circuit breaker placement locations are confirmed based on the constructed Bayesian network and performance impact relationships, and simulated circuit breaker location data is obtained after confirmation.

[0067] Preferably, for step S3, which involves arranging the circuit breakers in positions based on the data of the circuit breakers to be arranged and constructing a three-dimensional model, the method further includes:

[0068] Construct a 3D model from the simulated location data of the circuit breaker;

[0069] This involves retrieving basic information about the circuit breaker from a database, including its size, shape, and material; and then using a 3D scanner to acquire 3D coordinate data based on this information.

[0070] Use the Geomagic tool to convert 3D coordinate data into an editable 3D model;

[0071] The converted editable 3D model is then rendered in 3D, and the standard circuit breaker location data is obtained after the 3D rendering is completed.

[0072] Preferably, real-time monitoring data is collected for the circuit breakers in the standard circuit breaker location data in S4, and the collected real-time monitoring data is evaluated. The effectiveness of the circuit breaker location arrangement is judged based on the evaluation results.

[0073] The circuit breakers are positioned according to the standard circuit breaker position data, and the circuit breakers are monitored in real time after the position is arranged.

[0074] Retrieve historical fault records of circuit breakers from the database and construct a fault diagnosis model based on the historical fault records;

[0075] The real-time monitoring data is used to identify faults using the constructed fault diagnosis model;

[0076] The effectiveness level of the location layout is judged based on the fault identification results;

[0077] If the location layout effect level is within the adjustment range, the circuit breaker will be repositioned; if the location layout effect level is not within the adjustment range, the circuit breaker does not need to be repositioned.

[0078] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0079] 1. This invention employs a non-intrusive, multi-dimensional sensing optimal arrangement method for monitoring the mechanical state of circuit breakers. By using non-intrusive sensors for monitoring, it avoids the drawbacks of traditional monitoring methods that require direct contact with the internal structure of the circuit breaker, thus reducing the impact on normal equipment operation and potential safety risks. Simultaneously, it also reduces the risk of equipment failure due to monitoring operations.

[0080] 2. This invention employs a non-intrusive, multi-dimensional sensing optimal arrangement method for the mechanical state of circuit breakers. It utilizes specialized feature extraction methods for different data types to ensure that key information for each data type is effectively captured. This targeted feature extraction approach helps improve the accuracy and efficiency of subsequent analysis. Through in-depth analysis of multi-dimensional feature data, early diagnosis of the mechanical state of circuit breakers and prediction of potential faults can be achieved.

[0081] 3. This invention employs a non-intrusive, multi-dimensional sensing optimal layout method for circuit breakers based on their mechanical state. It automatically confirms the placement of circuit breakers according to a constructed Bayesian network and performance impact relationships, reducing the subjectivity and uncertainty of human decision-making. By combining multi-dimensional characteristic data and performance evaluation data of the circuit breakers and utilizing multivariate statistical methods to analyze the impact relationships between the data, the objectivity and accuracy of decision-making are improved. Attached Figure Description

[0082] Figure 1 This is a schematic diagram of the steps of the optimal circuit breaker arrangement method of the present invention;

[0083] Figure 2 This is a schematic diagram of the optimal circuit breaker arrangement method of the present invention. Detailed Implementation

[0084] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention. To solve the problem of inaccurate data acquisition when acquiring mechanical state data of circuit breakers in the prior art due to imperfect data acquisition methods, please refer to... Figure 1 and Figure 2 This embodiment provides the following technical solution:

[0085] A non-intrusive multi-dimensional sensing optimal arrangement method for the mechanical state of circuit breakers includes the following steps:

[0086] S1: Circuit breaker data acquisition: Monitor and acquire the original mechanical status data of the circuit breaker, and mark the acquired original mechanical status data of the circuit breaker as status data to be processed;

[0087] Among them, multi-dimensional monitoring methods can more comprehensively reflect the actual operating status of circuit breakers and improve the accuracy and timeliness of fault diagnosis;

[0088] S2: Circuit breaker data analysis: The status data to be processed is preprocessed. After the data preprocessing is completed, the index data of the circuit breaker is extracted. After the index data is extracted, the multi-dimensional feature data of the circuit breaker is obtained.

[0089] Specifically, specialized feature extraction methods are used for different types of data to ensure that key information of each data type can be effectively captured.

[0090] S3: Circuit breaker optimization layout: Based on the data of the circuit breakers to be arranged, the positions of the circuit breakers are arranged and a three-dimensional model is constructed to obtain the standard circuit breaker position data;

[0091] Among them, intelligent location arrangement can ensure that the circuit breaker operates in the optimal position, thereby improving the operating efficiency and reliability of the equipment;

[0092] S4: Circuit breaker data evaluation: Real-time monitoring data of circuit breakers in standard circuit breaker location data is collected, and the collected real-time monitoring data is evaluated. The effectiveness of the circuit breaker location arrangement is judged based on the evaluation results.

[0093] Among them, real-time monitoring and fault diagnosis can detect and deal with potential faults in advance, avoiding system downtime or accidents caused by the escalation of faults.

[0094] The monitoring and acquisition of the original mechanical condition data of the circuit breaker in S1 includes:

[0095] The mechanical condition data of a circuit breaker includes temperature data, acoustic data, vibration data, electrical data, optical data, gas analysis data, environmental data, and mechanical characteristic data.

[0096] The data includes temperature data (infrared thermal imaging and resistance temperature data), acoustic data (acoustic emission signals and sound spectrum analysis), vibration data (vibration waveforms and spectrum analysis), electrical data (partial discharge signals, current and voltage waveforms), optical data (visual images and optical sensors), gas analysis data (SF6 gas composition analysis), environmental data (humidity and temperature), and mechanical characteristic data (operating time and operating force).

[0097] The mechanical condition data of the circuit breaker is monitored using non-invasive sensors;

[0098] Specifically, temperature data is monitored using a non-contact infrared thermal imager and temperature sensor; acoustic data is monitored using an acoustic emission sensor and an external microphone; vibration data is monitored using a MEMS accelerometer; electrical data is monitored using a high-frequency current transformer and voltage transformer; optical data is monitored using a high-definition camera and laser rangefinder; gas analysis data is monitored using a gas sampler; environmental data is monitored using temperature and humidity sensors; and mechanical property data is monitored using displacement sensors and force sensors.

[0099] The monitored temperature data, acoustic data, vibration data, electrical data, optical data, gas analysis data, environmental data, and mechanical characteristic data are uniformly labeled as data to be processed.

[0100] Specifically, by combining the monitoring of multiple physical quantities such as temperature, acoustics, and vibration, the operating status of the circuit breaker can be comprehensively reflected, improving the accuracy and reliability of fault diagnosis. Data from different dimensions can corroborate each other, forming a more complete fault profile, which helps to discover hidden problems that are difficult to capture by single monitoring methods. The use of advanced sensors and data analysis technology improves measurement accuracy and stability. By integrating multiple sensors such as infrared thermal imagers, temperature sensors, acoustic emission sensors, external microphones, MEMS accelerometers, high-frequency current transformers, voltage transformers, high-definition cameras, laser rangefinders, gas samplers, temperature and humidity sensors, displacement sensors, and force sensors, comprehensive monitoring of the circuit breaker's mechanical state across multiple dimensions, including temperature, acoustics, vibration, electrical, optical, gas analysis, environmental, and mechanical characteristics, is achieved. This multi-dimensional monitoring method can more comprehensively reflect the actual operating status of the circuit breaker, improving the accuracy and timeliness of fault diagnosis. The use of non-invasive sensors avoids the shortcomings of traditional monitoring methods that require direct contact with the internal structure of the circuit breaker, reducing the impact on normal equipment operation and potential safety risks. Simultaneously, it reduces the risk of equipment failure due to monitoring operations. Modern sensor technologies, such as infrared thermal imagers, MEMS accelerometers, and high-frequency current transformers, possess high precision and sensitivity, capable of capturing minute state changes, thereby improving the accuracy and reliability of monitoring data. This is crucial for the timely detection of potential equipment failures and anomalies. To address the issue in existing technologies where targeted feature extraction of acquired circuit breaker mechanical state data is not performed, leading to inaccurate analysis results in subsequent data analysis, please refer to [link to relevant documentation]. Figure 1 and Figure 2 This embodiment provides the following technical solution:

[0101] In S2, the status data to be processed is preprocessed, and after the data preprocessing is completed, the circuit breaker's index data is extracted, including:

[0102] Perform data preprocessing on the status data to be processed;

[0103] The data preprocessing process includes: denoising, calibrating, normalizing, and interpolating the temperature data in the data to be processed; denoising, enhancing, and synchronizing the acoustic data in the data to be processed; denoising, calibrating, and reducing the dimensionality of the vibration data in the data to be processed; denoising, enhancing, and synchronizing the electrical data in the data to be processed; enhancing and denoising the optical data in the data to be processed; correcting and denoising the gas analysis data in the data to be processed; denoising and calibrating the environmental data in the data to be processed; and denoising and calibrating the mechanical property data in the data to be processed.

[0104] Finally, we obtain the preprocessed data and the data awaiting processing.

[0105] Extract data features from the preprocessed data to be processed state data;

[0106] During data feature extraction, features are extracted based on the data attributes of the state data to be processed.

[0107] Among them, feature extraction of temperature data involves hot spot detection, temperature gradient calculation, temperature difference identification, and time series identification of infrared thermal imaging data; and temperature change trend analysis and outlier identification of resistance temperature data.

[0108] Feature extraction from acoustic data involves frequency and signal energy identification of acoustic emission signals; sound spectrum analysis involves spectral feature identification and time-frequency identification.

[0109] Vibration data feature extraction involves performing time-domain feature identification, frequency-domain feature identification, and time-frequency feature identification on vibration waveforms and spectrum analysis.

[0110] The feature extraction of electrical data involves identifying the number of discharges, the phase of discharge, and the energy of discharge in partial discharge signals; and identifying the waveform features and frequency features of current and voltage waveforms.

[0111] Feature extraction from optical data involves recognizing the shape, texture, and color features of visual images; and distance measurement and recognition using optical sensors.

[0112] Feature extraction of gas analysis data involves the identification of concentration features and concentration change features;

[0113] Feature extraction of environmental data involves the identification of statistical and correlational features;

[0114] Feature extraction from mechanical property data is used to identify statistical features and operating patterns;

[0115] After feature extraction of the state data to be processed, feature selection is performed on the extracted features. The feature selection uses principal component analysis to calculate the principal components of the extracted features.

[0116] After principal component calculation, the index data in the state data to be processed is obtained and labeled as multidimensional feature data of the circuit breaker.

[0117] Specifically, denoising effectively removes noise signals from the original data, reduces interference, and improves data purity and accuracy. Calibration ensures data accuracy and consistency, especially for physical quantities requiring precise measurement (such as temperature and vibration), eliminating measurement bias. Normalization converts data to the same dimension and range, aiding subsequent data analysis and processing. In machine learning and deep learning applications, normalization is a crucial step in improving model performance. Imputation techniques fill in gaps in missing data, maintaining data integrity and avoiding analytical biases caused by missing data. For high-dimensional data (such as vibration data), dimensionality reduction techniques reduce data complexity and computational load while retaining key information, facilitating subsequent data analysis and processing. A rigorous data preprocessing workflow allows for timely detection and correction of errors and anomalies in the data, improving the reliability and stability of the entire system. Especially in the condition monitoring and fault diagnosis of critical equipment such as circuit breakers, accurate data is the foundation for ensuring the safe operation of the system. Through multi-dimensional data feature extraction, including infrared thermal imaging, resistance temperature, acoustic, vibration, electrical, optical, gas analysis, environmental, and mechanical characteristics, comprehensive and detailed monitoring of the mechanical condition of circuit breakers can be achieved. This multi-dimensional data collection and analysis helps to more accurately identify the operating status and potential problems of the equipment. Specialized feature extraction methods are used for different types of data, such as hotspot detection, frequency identification, and time-frequency feature identification, ensuring that key information of each data type is effectively captured. This targeted feature extraction method helps improve the accuracy and efficiency of subsequent analysis. Through in-depth analysis of multi-dimensional feature data, early diagnosis of the mechanical condition of circuit breakers and prediction of potential faults can be achieved. This helps to take measures in advance to prevent faults from occurring and improve the reliability and safety of equipment operation. Specifically, the acoustic data in the condition data to be processed undergoes sound enhancement processing, including:

[0118] Extract the acoustic data from the state data to be processed;

[0119] Extract a preset initial volume threshold from the database; wherein the initial volume threshold is not lower than a preset minimum volume value that can meet data quality requirements;

[0120] A volume threshold is set based on the acoustic data in the data to be processed, combined with an initial volume threshold, wherein the volume threshold is obtained by the following formula:

[0121]

[0122] Where F represents the volume threshold, and when the volume threshold is lower than the preset minimum volume value that can meet the data quality requirements, F = 1.08F. x And, F x This indicates the preset minimum volume level required to achieve the desired data quality; F yc The initial volume threshold is represented by ; n represents the number of frames contained in the acoustic data of the state data to be processed; m represents the number of frames in the acoustic data of the state data to be processed that are below the initial volume threshold; F i F represents the volume value corresponding to the i-th frame of acoustic data in the state data to be processed; i This represents the intermediate volume value corresponding to n frames of acoustic data in the state data to be processed; F max and F min This represents the maximum and minimum volume values ​​corresponding to n frames of acoustic data in the state data to be processed; F b This represents the standard deviation of volume corresponding to n frames of acoustic data in the state data to be processed.

[0123] The acoustic data in the state data to be processed is compared with the volume threshold.

[0124] Extract acoustic data below the volume threshold and perform volume enhancement processing on the acoustic data below the volume threshold.

[0125] The technical effect of the above solution is as follows: By comprehensively considering the acoustic data characteristics (such as frame count, volume distribution, standard deviation, etc.) in the data to be processed, as well as the preset initial volume threshold and minimum volume requirement, a suitable volume threshold is dynamically calculated. This adaptive method can more accurately identify the low-volume parts that need enhancement, avoiding the over-enhancement or under-enhancement problems that may be caused by a single fixed threshold. By ensuring that the volume threshold is not lower than the preset minimum volume value that can meet the data quality requirements, this technical solution ensures that the processed acoustic data reaches at least a high quality standard in terms of volume, which is crucial for subsequent applications such as sound analysis and speech recognition.

[0126] This targeted approach enhances volume only for acoustic data below a specific volume threshold. This not only reduces unnecessary computational resource consumption but also prevents sound quality degradation in non-low-volume areas, effectively improving the clarity and intelligibility of low-volume portions while maintaining overall sound quality. In applications such as audio processing, speech recognition, and voice communication, low-volume portions are often easily overlooked or misinterpreted. This volume enhancement technology significantly improves the auditory experience of these low-volume areas, enhancing the overall user experience. The formulas and parameters in this technology (such as the initial volume threshold, minimum volume value, and volume standard deviation) can be adjusted according to specific application scenarios and needs, offering high flexibility and scalability to adapt to diverse sound processing requirements.

[0127] In summary, this technical solution effectively improves the quality and discernibility of acoustic data by adaptively setting a volume threshold and only performing volume enhancement processing on acoustic data below the threshold, while maintaining the stability of overall sound quality. This is of great significance for improving user experience and meeting specific application needs.

[0128] Specifically, the acoustic data below the volume threshold is extracted, and the acoustic data below the volume threshold is subjected to volume enhancement processing, including:

[0129] Extract acoustic data below the volume threshold;

[0130] Extract the preset minimum volume value that meets data quality requirements;

[0131] The first volume adjustment coefficient corresponding to the acoustic data below the volume threshold is obtained by using the volume value corresponding to the acoustic data below the volume threshold and the preset minimum volume value that can meet the data quality requirements.

[0132] The first volume adjustment coefficient is obtained by the following formula:

[0133]

[0134] Among them, f 01 Indicates the first volume adjustment coefficient; k represents the number of frames corresponding to acoustic data below the volume threshold; F j This represents the volume value corresponding to the acoustic data in frame j that is below the volume threshold; F represents the volume threshold; F yc Indicates the initial volume threshold; F x This indicates the minimum volume value preset to meet data quality requirements;

[0135] A second volume adjustment coefficient is set for the acoustic data below the volume threshold using the quantitative relationship between the volume value corresponding to the acoustic data below the volume threshold and the volume threshold; wherein, the second volume adjustment coefficient is obtained by the following formula:

[0136]

[0137] Among them, f 02 Indicates the second volume adjustment coefficient; k represents the number of frames corresponding to acoustic data below the volume threshold; F j This represents the volume value corresponding to the acoustic data in frame j that is below the volume threshold; F represents the volume threshold; F x This represents the preset minimum volume value that meets data quality requirements; c represents the adjustment factor, which is obtained using the following formula:

[0138]

[0139] Where c represents the adjustment factor; F min This represents the minimum volume value corresponding to n frames of acoustic data in the state data to be processed; F m This represents the maximum volume in the acoustic data of frames k that are below the volume threshold; F p This represents the average volume of acoustic data in k frames that are below the volume threshold;

[0140] The volume values ​​corresponding to the acoustic data below the volume threshold are adjusted using the first volume adjustment coefficient and the second volume adjustment coefficient to obtain the adjusted acoustic data; wherein, the volume value of the adjusted acoustic data is obtained by the following formula:

[0141]

[0142] Among them, F t This represents the adjusted acoustic data volume value corresponding to each frame of acoustic data below the stated volume threshold; f 01 f represents the first volume adjustment factor; 02 The value represents the second volume adjustment coefficient; F represents the volume threshold; F0 represents the volume value of the acoustic data before adjustment corresponding to the acoustic data below the volume threshold in each frame.

[0143] The technical effect of the above solution is as follows: By introducing a first volume adjustment coefficient and a second volume adjustment coefficient, this solution achieves more refined volume adjustment for acoustic data below a volume threshold. The first volume adjustment coefficient is based on the volume threshold, the initial volume threshold, a preset minimum volume value, and the volume value for each frame, ensuring that the adjusted volume at least meets the preset minimum requirement. The second volume adjustment coefficient further considers the volume distribution characteristics within the acoustic data below the volume threshold. By introducing the adjustment factor, the volume adjustment better matches the characteristics of the data itself, avoiding over-adjustment or under-adjustment.

[0144] By finely adjusting volume, this technical solution can significantly improve the sound quality of acoustic data below a certain volume threshold, bringing it to or exceeding the preset minimum requirement, thereby improving the overall data quality. This is particularly important for applications in speech recognition, voice communication, and audio analysis, as low-volume data often affects the accuracy and efficiency of these applications. During volume enhancement, this solution not only considers increasing the volume value but also maintains a balance in sound quality by setting a second volume adjustment coefficient and adjustment factor. That is, while increasing volume, it minimizes damage to the original sound quality characteristics, such as avoiding distortion and noise. The formulas and parameters in this solution (such as the first and second volume adjustment coefficients and adjustment factors) are highly flexible and can be adjusted according to specific application scenarios and needs. For example, the preset minimum volume value and initial volume threshold can be adjusted based on different audio types and acquisition environments to adapt to different sound processing requirements. By improving the sound quality and intelligibility of low-volume data, this technical solution can significantly enhance the user experience in scenarios such as speech recognition, voice communication, and audio playback. Users no longer need to strain to hear the low-volume parts, and the possibility of misunderstandings or misidentifications caused by low-volume data is reduced.

[0145] In summary, this technical solution effectively improves the quality and discernibility of acoustic data below the volume threshold through refined volume adjustment and adaptive parameter settings, while maintaining the balance and stability of sound quality. This is of great significance for enhancing user experience and meeting specific application needs.

[0146] To address the issues in existing technologies where circuit breaker placement is not optimized and modeling is not based on the placement location, leading to inconvenience in subsequent circuit breaker adjustments, please refer to [link to relevant documentation]. Figure 1 and Figure 2 This embodiment provides the following technical solution:

[0147] In S3, the circuit breakers are positioned based on the data of the circuit breakers to be installed, and a 3D model is constructed, including:

[0148] The circuit breaker's performance is evaluated based on its multidimensional characteristic data. The performance evaluation includes contact performance evaluation, insulator performance evaluation, operating mechanism performance evaluation, and mechanical condition evaluation.

[0149] A Bayesian network is constructed based on the performance evaluation results and the location layout rules, wherein the location layout rules are obtained from the database.

[0150] Once the Bayesian network is constructed, each node and each edge in the Bayesian network is identified.

[0151] Each node represents an option point for the performance parameters and location layout of the circuit breaker; each edge represents an associated edge for the performance parameters and location layout.

[0152] The performance impact of multidimensional characteristic data and performance evaluation data of circuit breakers is judged, and the influence relationship between multidimensional characteristic data and performance evaluation data of circuit breakers is analyzed by multivariate statistical method.

[0153] The circuit breaker placement locations are confirmed based on the constructed Bayesian network and performance impact relationships, and simulated circuit breaker location data is obtained after confirmation.

[0154] Construct a 3D model from the simulated location data of the circuit breaker;

[0155] This involves retrieving basic information about the circuit breaker from a database, including its size, shape, and material; and then using a 3D scanner to acquire 3D coordinate data based on this information.

[0156] Use the Geomagic tool to convert 3D coordinate data into an editable 3D model;

[0157] The converted editable 3D model is then rendered in 3D, and the standard circuit breaker location data is obtained after the 3D rendering is completed.

[0158] Specifically, by evaluating contact performance, insulator performance, operating mechanism performance, and mechanical condition from multiple dimensions, a comprehensive and in-depth understanding of the circuit breaker's overall performance can be achieved. Using multi-dimensional characteristic data of the circuit breaker for performance evaluation ensures the scientific rigor and accuracy of the results. Modeling circuit breaker performance parameters and location layout using Bayesian networks captures the complex relationships between these parameters, providing strong support for optimized layout. Based on the constructed Bayesian network and performance influence relationships, the circuit breaker's location is automatically confirmed, reducing the subjectivity and uncertainty of human decision-making. Combining multi-dimensional characteristic data and performance evaluation data, and using multivariate statistical methods to analyze the influence relationships between data, improves the objectivity and accuracy of decision-making. Intelligent location layout ensures that the circuit breaker operates in the optimal position, thereby improving equipment operating efficiency and reliability. Using a 3D scanner to acquire the circuit breaker's 3D coordinate data ensures high accuracy and detail. 3D scanners, characterized by non-contact measurement, high scanning speed, and high precision, can accurately capture the outline and detailed features of circuit breakers. Professional tools like Geomagic convert the 3D coordinate data into an editable 3D model, further enhancing the model's detail and realism. Geomagic and similar tools offer rich functionality and powerful capabilities in data processing and model optimization, ensuring model quality. The 3D scanner quickly acquires the circuit breaker's 3D coordinate data, while Geomagic efficiently converts this data into an editable 3D model. This process significantly shortens modeling time and improves work efficiency. The converted, editable 3D model allows users to modify and adjust it as needed. This provides users with greater flexibility and creative space, allowing for model optimization and improvement based on actual requirements. 3D rendering technology can present the circuit breaker's 3D model in an intuitive and vivid way. This visualization helps users better understand the circuit breaker's structure and layout, improving the efficiency of fault diagnosis and maintenance. The completed standard circuit breaker location data can be applied to multiple scenarios, such as simulation analysis, fault prediction, and maintenance guidance. This flexibility and versatility make this solution have broader application prospects.

[0159] For S4, real-time monitoring data is collected from the circuit breakers in the standard circuit breaker location data, and the collected real-time monitoring data is evaluated. The effectiveness of the circuit breaker location arrangement is judged based on the evaluation results.

[0160] The circuit breakers are positioned according to the standard circuit breaker position data, and the circuit breakers are monitored in real time after the position is arranged.

[0161] Retrieve historical fault records of circuit breakers from the database and construct a fault diagnosis model based on the historical fault records;

[0162] The real-time monitoring data is used to identify faults using the constructed fault diagnosis model;

[0163] The effectiveness level of the location layout is judged based on the fault identification results;

[0164] If the location layout effect level is within the adjustment range, the circuit breaker will be repositioned; if the location layout effect level is not within the adjustment range, the circuit breaker does not need to be repositioned.

[0165] Specifically, by acquiring historical fault records of circuit breakers, a fault diagnosis model is constructed. This model learns the patterns and characteristics of fault occurrence, improving the accuracy and efficiency of fault diagnosis. Using this model to identify faults in real-time monitoring data, the potential fault types and locations of circuit breakers can be quickly identified, providing strong support for subsequent maintenance and replacement. Based on the fault identification results, the effectiveness of the circuit breaker placement is assessed using quantitative evaluation standards, making the evaluation results more objective and scientific. If the placement effectiveness level is within the adjustment range, the circuit breakers are promptly rearranged to optimize the placement; if the effectiveness level is outside the adjustment range, the current placement is deemed reasonable and no rearrangement is necessary. This dynamic adjustment mechanism ensures continuous optimization and adaptability of circuit breaker placement. Through real-time monitoring and fault diagnosis, potential faults can be detected and addressed early, preventing system downtime or accidents caused by escalating faults and improving the overall system operating efficiency.

[0166] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0167] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention.

Claims

1. A non-intrusive multi-dimensional sensing optimal arrangement method for the mechanical state of circuit breakers, characterized in that, Includes the following steps: S1: Circuit breaker data acquisition: Monitor and acquire the original mechanical status data of the circuit breaker, and mark the acquired original mechanical status data of the circuit breaker as status data to be processed; S2: Circuit breaker data analysis: The status data to be processed is preprocessed. After the data preprocessing is completed, the index data of the circuit breaker is extracted. After the index data is extracted, the multi-dimensional feature data of the circuit breaker is obtained. S3: Circuit breaker optimization layout: Based on the data of the circuit breakers to be arranged, the positions of the circuit breakers are arranged and a three-dimensional model is constructed to obtain the standard circuit breaker position data; S4: Circuit breaker data evaluation: Real-time monitoring data of circuit breakers in standard circuit breaker location data is collected, and the collected real-time monitoring data is evaluated. The effectiveness of the circuit breaker location arrangement is judged based on the evaluation results. The acoustic data in the state data to be processed is subjected to sound enhancement processing, including: Extract the acoustic data from the state data to be processed; Extract a preset initial volume threshold from the database; wherein the initial volume threshold is not lower than a preset minimum volume value that can meet data quality requirements; A volume threshold is set based on the acoustic data in the data to be processed, combined with an initial volume threshold, wherein the volume threshold is obtained by the following formula: ； Where F represents the volume threshold, and when the volume threshold is lower than the preset minimum volume value that can meet the data quality requirements, F = 1.08F. x And, F x This indicates the minimum preset volume level required to achieve the desired data quality; F yc The initial volume threshold is represented by ; n represents the number of frames contained in the acoustic data of the state data to be processed; m represents the number of frames in the acoustic data of the state data to be processed that are below the initial volume threshold; F i F represents the volume value corresponding to the i-th frame of acoustic data in the state data to be processed; i This represents the intermediate volume value corresponding to n frames of acoustic data in the state data to be processed; F max and F min This represents the maximum and minimum volume values ​​corresponding to n frames of acoustic data in the state data to be processed; F b This represents the standard deviation of volume corresponding to n frames of acoustic data in the state data to be processed. The acoustic data in the state data to be processed is compared with the volume threshold. Extracting acoustic data below the volume threshold and performing volume enhancement processing on the acoustic data below the volume threshold, including: Extract acoustic data below the volume threshold; Extract the preset minimum volume value that meets data quality requirements; The first volume adjustment coefficient corresponding to the acoustic data below the volume threshold is obtained by using the volume value corresponding to the acoustic data below the volume threshold and the preset minimum volume value that can meet the data quality requirements. The first volume adjustment coefficient is obtained by the following formula: ； Among them, f 01 Indicates the first volume adjustment coefficient; k represents the number of frames corresponding to acoustic data below the volume threshold; F j This represents the volume value corresponding to the acoustic data in frame j that is below the volume threshold; F represents the volume threshold; F yc Indicates the initial volume threshold; F x This indicates the minimum volume value preset to meet data quality requirements; A second volume adjustment coefficient is set for the acoustic data below the volume threshold using the quantitative relationship between the volume value corresponding to the acoustic data below the volume threshold and the volume threshold. The second volume adjustment coefficient is obtained using the following formula: ； Among them, f 02 Indicates the second volume adjustment coefficient; k represents the number of frames corresponding to acoustic data below the volume threshold; F j This represents the volume value corresponding to the acoustic data in frame j that is below the volume threshold; F represents the volume threshold; F x This represents the preset minimum volume value that meets data quality requirements; c represents the adjustment factor, which is obtained using the following formula: ； Where c represents the adjustment factor; F min This represents the minimum volume value corresponding to n frames of acoustic data in the state data to be processed; F m This represents the maximum volume in the acoustic data of frames k that are below the volume threshold; F p This represents the average volume of acoustic data in k frames that are below the volume threshold; The volume values ​​corresponding to the acoustic data below the volume threshold are adjusted using the first volume adjustment coefficient and the second volume adjustment coefficient to obtain the adjusted acoustic data; wherein, the volume value of the adjusted acoustic data is obtained by the following formula: ； Among them, F t This represents the adjusted acoustic data volume value corresponding to each frame of acoustic data below the stated volume threshold; f 01 f represents the first volume adjustment factor; 02 This represents the second volume adjustment coefficient; F represents the volume threshold; F0 represents the volume value of the acoustic data before adjustment corresponding to each frame of acoustic data that is lower than the volume threshold; In S3, the circuit breakers are positioned based on the data of the circuit breakers to be installed, and a 3D model is constructed, including: The circuit breaker's performance is evaluated based on its multidimensional characteristic data. The performance evaluation includes contact performance evaluation, insulator performance evaluation, operating mechanism performance evaluation, and mechanical condition evaluation. A Bayesian network is constructed based on the performance evaluation results and the location layout rules, wherein the location layout rules are obtained from the database. Once the Bayesian network is constructed, each node and each edge in the Bayesian network is identified. Each node represents an option point for the performance parameters and location layout of the circuit breaker; each edge represents an associated edge for the performance parameters and location layout. The performance impact of multidimensional characteristic data and performance evaluation data of circuit breakers is judged, and the influence relationship between multidimensional characteristic data and performance evaluation data of circuit breakers is analyzed by multivariate statistical method. The circuit breaker placement locations are confirmed based on the constructed Bayesian network and performance impact relationships, and simulated circuit breaker location data is obtained after confirmation.

2. The non-intrusive multi-dimensional sensing optimal arrangement method for the mechanical state of a circuit breaker according to claim 1, characterized in that: The monitoring and acquisition of the original mechanical condition data of the circuit breaker in S1 includes: The mechanical condition data of a circuit breaker includes temperature data, acoustic data, vibration data, electrical data, optical data, gas analysis data, environmental data, and mechanical characteristic data. The data includes temperature data (infrared thermal imaging and resistance temperature data), acoustic data (acoustic emission signals and sound spectrum analysis), vibration data (vibration waveforms and spectrum analysis), electrical data (partial discharge signals, current and voltage waveforms), optical data (visual images and optical sensors), gas analysis data (SF6 gas composition analysis), environmental data (humidity and temperature of the surrounding environment), and mechanical characteristic data (operating time and operating force).

3. The non-intrusive multi-dimensional sensing optimal arrangement method for the mechanical state of a circuit breaker according to claim 2, characterized in that: The monitoring and acquisition of the original mechanical condition data of the circuit breaker in S1 also includes: The mechanical condition data of the circuit breaker is monitored using non-invasive sensors; Specifically, temperature data is monitored using a non-contact infrared thermal imager and temperature sensor; acoustic data is monitored using an acoustic emission sensor and an external microphone; vibration data is monitored using a MEMS accelerometer; electrical data is monitored using a high-frequency current transformer and voltage transformer; optical data is monitored using a high-definition camera and laser rangefinder; gas analysis data is monitored using a gas sampler; environmental data is monitored using temperature and humidity sensors; and mechanical property data is monitored using displacement sensors and force sensors. The monitored temperature data, acoustic data, vibration data, electrical data, optical data, gas analysis data, environmental data, and mechanical characteristic data are uniformly labeled as data to be processed.

4. The non-intrusive multi-dimensional sensing optimal arrangement method for the mechanical state of a circuit breaker according to claim 3, characterized in that: In S2, the status data to be processed is preprocessed, and after the data preprocessing is completed, the circuit breaker's index data is extracted, including: Perform data preprocessing on the status data to be processed; The data preprocessing process includes: denoising, calibrating, normalizing, and interpolating the temperature data in the data to be processed; denoising, enhancing, and synchronizing the acoustic data in the data to be processed; denoising, calibrating, and reducing the dimensionality of the vibration data in the data to be processed; denoising, enhancing, and synchronizing the electrical data in the data to be processed; enhancing and denoising the optical data in the data to be processed; correcting and denoising the gas analysis data in the data to be processed; denoising and calibrating the environmental data in the data to be processed; and denoising and calibrating the mechanical property data in the data to be processed. Finally, we obtain the preprocessed data and the data awaiting processing.

5. The non-intrusive multi-dimensional sensing optimal arrangement method for the mechanical state of a circuit breaker according to claim 4, characterized in that: In S2, data preprocessing is performed on the status data to be processed. After the data preprocessing is completed, the circuit breaker's index data is extracted. This also includes: Extract data features from the preprocessed data to be processed state data; During data feature extraction, features are extracted based on the data attributes of the state data to be processed. Among them, feature extraction of temperature data involves hot spot detection, temperature gradient calculation, temperature difference identification, and time series identification of infrared thermal imaging data; and temperature change trend analysis and outlier identification of resistance temperature data. Feature extraction from acoustic data involves frequency and signal energy identification of acoustic emission signals; sound spectrum analysis involves spectral feature identification and time-frequency identification. Vibration data feature extraction involves performing time-domain feature identification, frequency-domain feature identification, and time-frequency feature identification on vibration waveforms and spectrum analysis. The feature extraction of electrical data involves identifying the number of discharges, the phase of discharge, and the energy of discharge in partial discharge signals; and identifying the waveform features and frequency features of current and voltage waveforms. Feature extraction from optical data involves recognizing the shape, texture, and color features of visual images; and distance measurement and recognition using optical sensors. Feature extraction of gas analysis data involves the identification of concentration features and concentration change features; Feature extraction of environmental data involves the identification of statistical and correlational features; Feature extraction from mechanical property data is used to identify statistical features and operating patterns; After feature extraction of the state data to be processed, feature selection is performed on the extracted features. The feature selection uses principal component analysis to calculate the principal components of the extracted features. After principal component calculation, the index data in the state data to be processed is obtained and labeled as multidimensional feature data of the circuit breaker.

6. The non-intrusive multi-dimensional sensing optimal arrangement method for the mechanical state of a circuit breaker according to claim 5, characterized in that: For S3, which involves arranging circuit breakers based on their data and constructing a 3D model, the following are also included: Construct a 3D model from the simulated location data of the circuit breaker; This includes retrieving basic information about the circuit breaker from the database; Based on the basic information of the circuit breaker, three-dimensional coordinate data is obtained using a 3D scanner. Use the Geomagic tool to convert 3D coordinate data into an editable 3D model; The converted editable 3D model is then rendered in 3D, and the standard circuit breaker location data is obtained after the 3D rendering is completed.

7. The non-intrusive multi-dimensional sensing optimal arrangement method for the mechanical state of a circuit breaker according to claim 6, characterized in that: For S4, real-time monitoring data is collected from the circuit breakers in the standard circuit breaker location data, and the collected real-time monitoring data is evaluated. The effectiveness of the circuit breaker location arrangement is judged based on the evaluation results. The circuit breakers are positioned according to the standard circuit breaker position data, and the circuit breakers are monitored in real time after the position is arranged. Retrieve historical fault records of circuit breakers from the database and construct a fault diagnosis model based on the historical fault records; The real-time monitoring data is used to identify faults using the constructed fault diagnosis model; The effectiveness level of the location layout is judged based on the fault identification results; If the location layout effect level is within the adjustment range, the circuit breaker will be repositioned; if the location layout effect level is not within the adjustment range, the circuit breaker does not need to be repositioned.

Images