Circuit breaker mechanical state monitoring data cleaning method

Abstract

The invention discloses a circuit breaker mechanical state monitoring data cleaning method, and relates to the technical field of data cleaning. The invention discloses a method for cleaning mechanical state monitoring data of a circuit breaker. The method comprises the following steps of: 1, establishing a data cleaning model by utilizing an EMD algorithm; 2, acquiring and processing real-time monitoring data; 3, decomposing all monitoring data one by one; step 4, clustering all monitoring data decomposition results; and step 5, removing noise point data outside the data set. According to the method, the problem that the fault diagnosis of the circuit breaker is inaccurate due to the fact that noise and loss easily exist in the monitoring data obtained in the prior art is solved, the cleaned data are cleaner and tidier, the characteristics of the circuit breaker state data monitored in real time are better extracted, the data quality is effectively improved, and the fault diagnosis accuracy is improved. It is ensured that the operation state of the circuit breaker is better reflected by the data reserved after data cleaning, and the accuracy of fault diagnosis of the circuit breaker is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of data cleaning, and in particular to a method for cleaning data of a circuit breaker mechanical state monitoring. Background Art

[0002] Circuit breakers are important components of power grids and have a great impact on regional power supply. If they fail to operate, it may cause power shortages or power outages in most areas. More seriously, it may cause great losses to society and the economy. In order to avoid circuit breaker failures, it is necessary to monitor the status of circuit breakers.

[0003] The Chinese patent with the publication number CN109061463A discloses a method for mechanical state monitoring and fault diagnosis of a high-voltage circuit breaker, comprising the steps of: 1) performing artificial fault simulation experiments on some common high-voltage circuit breakers according to the statistical results of historical fault data of the high-voltage circuit breaker to obtain fault data; 2) analyzing the current data of the fault, and extracting the current data characteristic quantity of each mechanism fault as one of the basis for state diagnosis classification; 3) analyzing the state data of the circuit breaker monitored by vibration in real time of the fault, and using wavelet packet decomposition and sample entropy to process the high and low frequency components of the state data of the circuit breaker monitored by vibration in real time to obtain the corresponding wavelet packet relative energy and sample entropy as the characteristic quantity of the state data of the circuit breaker monitored by vibration in real time; 4) using principal component analysis to reduce the dimension of the characteristic quantity of the state data of the circuit breaker monitored by vibration in real time as one of the basis for state diagnosis classification, and using support vector machine to perform state diagnosis on the circuit breaker, effectively utilizing the multi-dimensional information of the fault state, and promoting the development of multi-parameter and multi-dimensional mapping fault diagnosis of the circuit breaker.

[0004] In actual use of the above patent, due to the operating environment of the circuit breaker and the loss of the communication line, the acquired monitoring data is prone to noise and omissions, resulting in inaccurate fault diagnosis of the circuit breaker; therefore, it does not meet the existing needs. In this regard, we propose a method for cleaning circuit breaker mechanical status monitoring data. Summary of the invention

[0005] The purpose of the present invention is to provide a method for cleaning data of mechanical status monitoring of a circuit breaker, so that the cleaned data is cleaner and neater, a lot of unnecessary work can be saved, thereby saving time and resources, better extracting the characteristics of the circuit breaker status data monitored in real time, providing more detailed analysis, effectively improving data quality, ensuring that the data retained after data cleaning better reflects the operating status of the circuit breaker, improving the accuracy of circuit breaker fault diagnosis, and solving the problems raised in the above-mentioned background technology.

[0006] To achieve the above object, the present invention provides the following technical solution: a method for cleaning data of mechanical condition monitoring of a circuit breaker, comprising the following steps:

[0007] Step 1: Obtain historical circuit breaker mechanical condition monitoring data, and establish a data cleaning model using the EMD algorithm based on the acquired historical circuit breaker mechanical condition monitoring data;

[0008] Step 2: Acquire the real-time monitoring data of the mechanical state of the circuit breaker, process the acquired monitoring data, and store the monitoring data after the processing to form a monitoring data column;

[0009] Step 3: Input the monitoring data column into the data cleaning model, and use the data cleaning model to decompose all monitoring data one by one;

[0010] Step 4: Save all monitoring data decomposition results, and cluster all monitoring data decomposition results to obtain multiple clustered data sets;

[0011] Step 5: Remove the noise point data outside the multiple clustered data sets to obtain the cleaned circuit breaker mechanical condition monitoring data.

[0012] Preferably, the circuit breaker mechanical state monitoring data is multidimensional characteristic data of the circuit breaker mechanical state, transmission mechanism data and energy storage mechanism data. The multidimensional characteristic data of the circuit breaker mechanical state mainly includes opening and closing data, stroke and time characteristics, circuit breaker state data of real-time vibration monitoring, coil current and speed. The collection of opening and closing data specifically includes: using a timer to collect the opening and closing time of the circuit breaker, using a photoelectric stroke sensor to collect the opening and closing displacement of the circuit breaker, and using an audio collector to collect the opening and closing sound data of the circuit breaker.

[0013] Preferably, the acquiring of real-time monitoring data of the mechanical state of the circuit breaker and processing of the acquired monitoring data specifically includes:

[0014] Real-time collection of multi-dimensional characteristic data of the mechanical state of the circuit breaker, transmission mechanism data and energy storage mechanism data of the circuit breaker;

[0015] The communication interference, white noise interference and random pulse interference in the collected data are removed, wherein the white noise interference is filtered out by the average filtering method, and the communication interference and random pulse interference are filtered out by the median filtering algorithm;

[0016] Data enhancement is performed on the data after interference removal, so that the characteristic parameters in the data after interference removal are enhanced.

[0017] Preferably, the step of acquiring real-time monitoring data of the mechanical state of the circuit breaker and processing the acquired monitoring data further includes:

[0018] Extract the data information after communication interference, white noise interference and random pulse interference are removed from the data;

[0019] Retrieving the corresponding signal-to-noise ratio parameters in the multi-dimensional characteristic data of the mechanical state of the circuit breaker, the transmission mechanism data and the energy storage mechanism data after the communication interference, white noise interference and random pulse interference in the completion data are removed;

[0020] The signal-to-noise ratio evaluation coefficient is obtained by using the corresponding signal-to-noise ratio parameters in the multi-dimensional characteristic data of the mechanical state of the circuit breaker, the transmission mechanism data and the energy storage mechanism data; wherein the signal-to-noise ratio evaluation coefficient is obtained by the following formula:

[0021]

[0022] Wherein, S represents the signal-to-noise ratio evaluation coefficient; n represents the number of data contained in the multidimensional characteristic data of the mechanical state of the circuit breaker, the transmission mechanism data and the energy storage mechanism data after the communication interference, white noise interference and random pulse interference in the data are removed; S hi represents the signal-to-noise ratio of the i-th data after denoising; S xi represents the signal-to-noise ratio of the i-th data before denoising; S xb represents the standard deviation of the signal-to-noise ratio corresponding to n data of the multidimensional characteristic data of the mechanical state of the circuit breaker, the transmission mechanism data and the energy storage mechanism data before denoising; S hb represents the signal-to-noise ratio standard deviation corresponding to n data of the multidimensional characteristic data of the mechanical state of the circuit breaker, the transmission mechanism data and the energy storage mechanism data after denoising; S hmax and S hmin represents the maximum value of the signal-to-noise ratio and the minimum value of the signal-to-noise ratio corresponding to n data after denoising of the multi-dimensional characteristic data of the mechanical state of the circuit breaker, the transmission mechanism data and the energy storage mechanism data; S xmax and S xmin Indicates the maximum value of the signal-to-noise ratio and the minimum value of the signal-to-noise ratio corresponding to n data of the multidimensional characteristic data of the mechanical state of the circuit breaker, the transmission mechanism data and the energy storage mechanism data before denoising;

[0023] Comparing the signal-to-noise ratio evaluation coefficient with a preset evaluation coefficient threshold;

[0024] When the signal-to-noise ratio evaluation coefficient is lower than a preset evaluation coefficient threshold, denoising is performed on the multidimensional feature data of the mechanical state of the circuit breaker, the transmission mechanism data and the energy storage mechanism data after the communication interference, the white noise interference and the random pulse interference in the completed data until the signal-to-noise ratio evaluation coefficient corresponding to the multidimensional feature data of the mechanical state of the circuit breaker, the transmission mechanism data and the energy storage mechanism data after the denoising exceeds the preset evaluation coefficient threshold;

[0025] When the signal-to-noise ratio evaluation coefficient is not lower than a preset evaluation coefficient threshold, a data distortion determination is performed, and a data quality abnormality alarm is issued when the data distortion is too large and the data quality is lower than a preset data quality requirement.

[0026] Preferably, when the signal-to-noise ratio evaluation coefficient is not lower than a preset evaluation coefficient threshold, a data distortion determination is performed, and when the data distortion is too large and the data quality is lower than a preset data quality requirement, a data quality abnormality alarm is performed, including:

[0027] When the signal-to-noise ratio evaluation coefficient is lower than a preset evaluation coefficient threshold, extracting the data gain ratio of the multi-dimensional feature data of the mechanical state of the circuit breaker, the transmission mechanism data and the energy storage mechanism data;

[0028] Obtaining a distortion determination parameter by combining the multidimensional characteristic data of the mechanical state of the circuit breaker, the transmission mechanism data and the energy storage mechanism data after denoising using the data gain ratio of the multidimensional characteristic data of the mechanical state of the circuit breaker, the transmission mechanism data and the energy storage mechanism data;

[0029] The distortion determination parameter is obtained by the following formula:

[0030]

[0031] Wherein, Z represents the distortion determination parameter; n represents the number of data contained in the multi-dimensional characteristic data of the mechanical state of the circuit breaker, the transmission mechanism data and the energy storage mechanism data after the communication interference, white noise interference and random pulse interference in the data are removed; S hi represents the signal-to-noise ratio of the i-th data after denoising; S xi represents the signal-to-noise ratio of the i-th data before denoising; Z i Indicates the data gain ratio after the ith data is denoised; Z c Indicates the preset data gain ratio reference value; S hb represents the signal-to-noise ratio standard deviation corresponding to n data of the multidimensional characteristic data of the mechanical state of the circuit breaker, the transmission mechanism data and the energy storage mechanism data after denoising; Z b represents the data gain ratio standard deviation corresponding to n data of the multi-dimensional characteristic data of the mechanical state of the circuit breaker, the transmission mechanism data and the energy storage mechanism data after denoising; Z min represents the minimum value of the data gain ratio corresponding to n data of the multi-dimensional characteristic data of the mechanical state of the circuit breaker, the transmission mechanism data and the energy storage mechanism data after denoising; S zx Indicates the signal-to-noise ratio change ratio of the data corresponding to the minimum value of the data gain ratio before and after the noise removal processing;

[0032] Comparing the distortion determination parameter with a preset parameter threshold;

[0033] When the distortion determination parameter exceeds a preset parameter threshold, it is determined that the data distortion is too large, resulting in the data quality being lower than a preset data quality requirement;

[0034] When data distortion is too large and the data quality is lower than the preset data quality requirements, a data quality abnormality alarm will be issued.

[0035] Preferably, the use of the EMD algorithm to establish a data cleaning model specifically includes:

[0036] Improve the EMD algorithm and receive the processed real-time monitoring data of the mechanical status of the circuit breaker;

[0037] The real-time monitoring data of the mechanical state of the circuit breaker is decomposed into multiple intrinsic mode functions through the improved EMD algorithm;

[0038] Screen the intrinsic mode functions, remove the intrinsic mode functions containing noise, and retain the intrinsic mode functions containing useful information;

[0039] The filtered intrinsic mode functions are superimposed, and the superimposed intrinsic mode functions are superimposed for training to obtain a data cleaning model.

[0040] Preferably, the improvement of the EMD algorithm specifically includes:

[0041] The improved wavelet denoising method is used to denoise the real-time monitored circuit breaker state data, and the denoised real-time monitored circuit breaker state data is adaptively decomposed into a series of intrinsic mode function components;

[0042] Select appropriate prediction models for components of different frequencies, and finally superimpose the prediction values ​​of each component to obtain the final prediction value;

[0043] The EMD algorithm is improved by adding Gaussian noise and cubic spline interpolation sampling, and removing the first intrinsic mode function.

[0044] Preferably, the use of a data cleaning model to decompose all monitoring data one by one specifically includes:

[0045] Column renaming: If there are two column names with the same name or the same meaning in the monitoring data column, rename the name of one of the data columns;

[0046] Delete duplicate values: If there are duplicate data values ​​in the monitoring data column, delete the duplicate data values ​​and keep the first record of the duplicate data when deleting;

[0047] Missing value processing: If there are missing values ​​in the data in the monitoring data column, use statistical methods to fill in the missing values ​​in the data;

[0048] Consistency processing: If there are inconsistent data or naming rules in the data column, use the split column function to split the data values ​​in the inconsistent data column;

[0049] Data sorting and processing: filter and sort the data in the data column, and identify and process outliers in the data.

[0050] Preferably, clustering all monitoring data decomposition results specifically includes:

[0051] The multi-dimensional feature data of the mechanical state of the circuit breaker in the monitoring data column is extracted by using multi-scale entropy to extract features useful for cluster analysis.

[0052] Convert the features of different dimensions or magnitudes in the multi-dimensional feature data of the mechanical state of the circuit breaker to the same scale;

[0053] The K-Means clustering algorithm is used to cluster the feature information in the multi-dimensional feature data of the circuit breaker mechanical state after feature transformation.

[0054] Preferably, the method of extracting features from the multi-dimensional feature data of the mechanical state of the circuit breaker in the monitoring data column by using multi-scale entropy specifically includes:

[0055] Inputting the multi-dimensional characteristic data of the mechanical state of the circuit breaker into the empirical mode decomposition algorithm, and initializing the parameters of the empirical mode decomposition algorithm to obtain multiple target decomposition parameters;

[0056] Performing modal decomposition on the multi-dimensional characteristic data of the mechanical state of the circuit breaker according to multiple target decomposition parameters to obtain multiple intrinsic modal component data;

[0057] The multi-scale entropy algorithm is used to analyze the characteristics of multiple intrinsic mode component data respectively, and the multi-scale entropy value characteristics of each intrinsic mode component data are obtained;

[0058] The multi-scale entropy value features of each intrinsic mode component data are classified by a support vector machine model to obtain the feature classification result of each intrinsic mode component data;

[0059] Determine multiple inherent mode functions of multidimensional characteristic data of the mechanical state of the circuit breaker according to the characteristic classification result of each intrinsic mode component data; iteratively analyze the multiple inherent mode functions to obtain multiple components of the multidimensional characteristic data of the mechanical state of the circuit breaker;

[0060] State features are extracted from components of a plurality of multi-dimensional feature data of mechanical states of circuit breakers to obtain state features of each of the plurality of multi-dimensional feature data of mechanical states of circuit breakers.

[0061] Compared with the prior art, the present invention has the following beneficial effects:

[0062] The present invention removes interference from the circuit breaker status monitoring data collected in real time, so that the cleaned data is cleaner and neater, which can save a lot of unnecessary work, thereby saving time and resources. By using the improved EMD algorithm to establish a data cleaning model, the characteristics of the circuit breaker status data monitored in real time can be better extracted, and more detailed analysis can be provided. Through multi-scale entropy analysis, potential problems of the circuit breaker can be discovered earlier, so that maintenance measures can be taken in advance, equipment failures can be avoided, equipment service life can be extended, maintenance costs can be reduced, data quality can be effectively improved, and it is ensured that the data retained after data cleaning better reflects the operating status of the circuit breaker, thereby improving the accuracy of circuit breaker fault diagnosis. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figure 1 It is a schematic diagram of a circuit breaker mechanical state monitoring data cleaning method of the present invention;

[0064] Figure 2 It is a schematic diagram of feature extraction of multi-dimensional feature data of mechanical state of circuit breaker according to the present invention. DETAILED DESCRIPTION

[0065] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0066] In order to solve the problem that in the actual use of the existing patent, due to the operating environment of the circuit breaker and the loss of the communication line, the monitoring data obtained is prone to noise and loss, resulting in inaccurate fault diagnosis of the circuit breaker, please refer to Figure 1-Figure 2 , this embodiment provides the following technical solutions:

[0067] A method for cleaning data of mechanical condition monitoring of a circuit breaker comprises the following steps:

[0068] Step 1: Obtain historical circuit breaker mechanical condition monitoring data, and establish a data cleaning model using the EMD algorithm based on the acquired historical circuit breaker mechanical condition monitoring data;

[0069] Step 2: Acquire the real-time monitoring data of the mechanical state of the circuit breaker, process the acquired monitoring data, and store the monitoring data after the processing to form a monitoring data column;

[0070] Step 3: Input the monitoring data column into the data cleaning model, and use the data cleaning model to decompose all monitoring data one by one;

[0071] Step 4: Save all monitoring data decomposition results, and cluster all monitoring data decomposition results to obtain multiple clustered data sets;

[0072] Step 5: Remove the noise point data outside the multiple clustered data sets to obtain the cleaned circuit breaker mechanical condition monitoring data.

[0073] Obtain real-time monitoring data of the mechanical state of the circuit breaker and process the acquired monitoring data, including:

[0074] Real-time collection of multi-dimensional characteristic data of the mechanical state of the circuit breaker, transmission mechanism data and energy storage mechanism data of the circuit breaker. The multi-dimensional characteristic data of the mechanical state of the circuit breaker mainly includes opening and closing data, travel and time characteristics, circuit breaker state data of real-time vibration monitoring, coil current and speed;

[0075] The communication interference, white noise interference and random pulse interference in the collected data are removed, wherein the white noise interference is filtered out by the average filtering method, and the communication interference and random pulse interference are filtered out by the median filtering algorithm;

[0076] Data enhancement is performed on the data after interference removal so that the characteristic parameters in the data after interference removal are enhanced. By removing the interference in the data, invalid, duplicate and erroneous data can be removed, and erroneous data can be corrected and null values ​​can be filled, which can make the data more accurate and reliable, make the cleaned data more in line with the actual situation, and make the results of data analysis more accurate and reliable.

[0077] The circuit breaker mechanical state monitoring data includes the multi-dimensional feature data of the circuit breaker mechanical state, the transmission mechanism data and the energy storage mechanism data, and the collection of the opening and closing data, specifically including: using a timer to collect the opening and closing time of the circuit breaker, using a photoelectric travel sensor to collect the opening and closing displacement of the circuit breaker, and using an audio collector to collect the opening and closing sound data of the circuit breaker. In the process of collecting the sound data, echo cancellation technology is used to eliminate the echo of the sound data and an adaptive filter is used to filter it. Then, the traditional real-time monitoring circuit breaker state data processing is combined with deep learning technology to further suppress the residual echo after the echo cancellation adaptive filtering.

[0078] Specifically, the step of acquiring real-time monitoring data of the mechanical state of the circuit breaker and processing the acquired monitoring data further includes:

[0079] Extract the data information after communication interference, white noise interference and random pulse interference are removed from the data;

[0080] Retrieving the corresponding signal-to-noise ratio parameters in the multi-dimensional characteristic data of the mechanical state of the circuit breaker, the transmission mechanism data and the energy storage mechanism data after the communication interference, white noise interference and random pulse interference in the completion data are removed;

[0081] The signal-to-noise ratio evaluation coefficient is obtained by using the corresponding signal-to-noise ratio parameters in the multi-dimensional characteristic data of the mechanical state of the circuit breaker, the transmission mechanism data and the energy storage mechanism data; wherein the signal-to-noise ratio evaluation coefficient is obtained by the following formula:

[0082]

[0083] Wherein, S represents the signal-to-noise ratio evaluation coefficient; n represents the number of data contained in the multidimensional characteristic data of the mechanical state of the circuit breaker, the transmission mechanism data and the energy storage mechanism data after the communication interference, white noise interference and random pulse interference in the data are removed; S hi represents the signal-to-noise ratio of the i-th data after denoising; S xi represents the signal-to-noise ratio of the i-th data before denoising; S xb represents the standard deviation of the signal-to-noise ratio corresponding to n data of the multidimensional characteristic data of the mechanical state of the circuit breaker, the transmission mechanism data and the energy storage mechanism data before denoising; S hb represents the signal-to-noise ratio standard deviation corresponding to n data of the multidimensional characteristic data of the mechanical state of the circuit breaker, the transmission mechanism data and the energy storage mechanism data after denoising; S hmax and S hmin represents the maximum value of the signal-to-noise ratio and the minimum value of the signal-to-noise ratio corresponding to n data after denoising of the multi-dimensional characteristic data of the mechanical state of the circuit breaker, the transmission mechanism data and the energy storage mechanism data; S xmax and S xmin Indicates the maximum value of the signal-to-noise ratio and the minimum value of the signal-to-noise ratio corresponding to n data of the multidimensional characteristic data of the mechanical state of the circuit breaker, the transmission mechanism data and the energy storage mechanism data before denoising;

[0084] Comparing the signal-to-noise ratio evaluation coefficient with a preset evaluation coefficient threshold;

[0085] When the signal-to-noise ratio evaluation coefficient is lower than a preset evaluation coefficient threshold, denoising is performed on the multidimensional feature data of the mechanical state of the circuit breaker, the transmission mechanism data and the energy storage mechanism data after the communication interference, the white noise interference and the random pulse interference in the completed data until the signal-to-noise ratio evaluation coefficient corresponding to the multidimensional feature data of the mechanical state of the circuit breaker, the transmission mechanism data and the energy storage mechanism data after the denoising exceeds the preset evaluation coefficient threshold;

[0086] When the signal-to-noise ratio evaluation coefficient is not lower than a preset evaluation coefficient threshold, a data distortion determination is performed, and a data quality abnormality alarm is issued when the data distortion is too large and the data quality is lower than a preset data quality requirement.

[0087] The technical effect of the above technical solution is: by extracting and removing communication interference, white noise interference and random pulse interference, the accuracy of real-time monitoring data of the mechanical state of the circuit breaker can be effectively improved. The data obtained after denoising is closer to the actual state, providing a more solid foundation for subsequent analysis and processing.

[0088] At the same time, the introduction of the signal-to-noise ratio evaluation coefficient as a quantitative indicator of data quality can effectively evaluate the quality of the data after denoising. The evaluation coefficient comprehensively considers multiple factors such as the change in the signal-to-noise ratio before and after denoising, the data standard deviation, the maximum and minimum values, and can fully reflect the quality of the data. In addition, by comparing the signal-to-noise ratio evaluation coefficient with the preset evaluation coefficient threshold, it is automatically determined whether further denoising is required. If the evaluation coefficient is lower than the threshold, denoising continues until the quality requirements are met; if the evaluation coefficient is not lower than the threshold, the next step is entered. This automated process reduces manual intervention and improves processing efficiency.

[0089] When data distortion is too large and the data quality is lower than the preset requirements, an abnormal data quality alarm is issued. This helps to promptly discover and deal with data quality issues and avoid erroneous analysis and decision-making based on low-quality data. The entire technical solution enhances the reliability and stability of the circuit breaker mechanical condition monitoring system by optimizing the data processing process and improving data quality. Accurate data helps to more accurately evaluate the operating status of the circuit breaker and promptly detect potential faults, thereby ensuring the safe and stable operation of the power system. The above technical solution significantly improves the accuracy and reliability of the real-time monitoring data of the circuit breaker mechanical state through a series of scientific data processing and analysis methods, providing a strong guarantee for the safe and stable operation of the power system.

[0090] Specifically, when the signal-to-noise ratio evaluation coefficient is not lower than a preset evaluation coefficient threshold, a data distortion determination is performed, and when the data distortion is too large and the data quality is lower than a preset data quality requirement, a data quality abnormality alarm is performed, including:

[0091] When the signal-to-noise ratio evaluation coefficient is lower than a preset evaluation coefficient threshold, extracting the data gain ratio of the multi-dimensional feature data of the mechanical state of the circuit breaker, the transmission mechanism data and the energy storage mechanism data;

[0092] Obtaining a distortion determination parameter by combining the multidimensional characteristic data of the mechanical state of the circuit breaker, the transmission mechanism data and the energy storage mechanism data after denoising using the data gain ratio of the multidimensional characteristic data of the mechanical state of the circuit breaker, the transmission mechanism data and the energy storage mechanism data;

[0093] The distortion determination parameter is obtained by the following formula:

[0094]

[0095] Wherein, Z represents the distortion determination parameter; n represents the number of data contained in the multi-dimensional characteristic data of the mechanical state of the circuit breaker, the transmission mechanism data and the energy storage mechanism data after the communication interference, white noise interference and random pulse interference in the data are removed; S hi represents the signal-to-noise ratio of the i-th data after denoising; S xi represents the signal-to-noise ratio of the i-th data before denoising; Z i Indicates the data gain ratio after the ith data is denoised; Z c Indicates the preset data gain ratio reference value; S hb represents the signal-to-noise ratio standard deviation corresponding to n data of the multidimensional characteristic data of the mechanical state of the circuit breaker, the transmission mechanism data and the energy storage mechanism data after denoising; Z b represents the data gain ratio standard deviation corresponding to n data of the multi-dimensional characteristic data of the mechanical state of the circuit breaker, the transmission mechanism data and the energy storage mechanism data after denoising; Z min represents the minimum value of the data gain ratio corresponding to n data of the multi-dimensional characteristic data of the mechanical state of the circuit breaker, the transmission mechanism data and the energy storage mechanism data after denoising; S zx Indicates the signal-to-noise ratio change ratio of the data corresponding to the minimum value of the data gain ratio before and after the noise removal processing;

[0096] Comparing the distortion determination parameter with a preset parameter threshold;

[0097] When the distortion determination parameter exceeds a preset parameter threshold, it is determined that the data distortion is too large, resulting in the data quality being lower than a preset data quality requirement;

[0098] When data distortion is too large and the data quality is lower than the preset data quality requirements, a data quality abnormality alarm will be issued.

[0099] The technical effect of the above technical solution is: after the signal-to-noise ratio evaluation coefficient meets the basic requirements (i.e. not less than the preset evaluation coefficient threshold), a distortion judgment parameter is further introduced to evaluate the authenticity and accuracy of the data. This dual evaluation mechanism can more comprehensively reflect the quality status of the data and ensure the reliability of subsequent analysis and decision-making. The distortion judgment parameter is calculated by extracting the data gain ratio of the multi-dimensional feature data of the mechanical state of the circuit breaker, the transmission mechanism data, and the energy storage mechanism data, and combining the denoising data. The data gain ratio reflects the degree of change of the data during the denoising process. Incorporating it into the distortion evaluation can more accurately reflect the effect of data processing and the true state of the data.

[0100] The calculation formula of the distortion judgment parameter comprehensively considers multiple factors such as the signal-to-noise ratio before and after denoising, the data gain ratio and its standard deviation, the minimum value, and the signal-to-noise ratio change ratio of the data corresponding to the minimum value of the data gain ratio before and after denoising. This comprehensive evaluation method can more scientifically judge the degree of data distortion and avoid misjudgment caused by a single indicator. When the distortion judgment parameter exceeds the preset parameter threshold, the system can quickly determine that the data distortion is too large, resulting in data quality lower than the preset requirements. At this time, the system can automatically trigger the data quality abnormality alarm to remind relevant personnel to pay attention to and deal with data quality issues in a timely manner, and avoid erroneous analysis and decision-making based on low-quality data. At the same time, by introducing the distortion judgment and data quality abnormality alarm mechanism, this technical solution can significantly improve the overall performance of the circuit breaker mechanical condition monitoring system. It can not only ensure the accuracy and reliability of the data, but also timely discover and deal with data quality issues, providing a more solid guarantee for the safe and stable operation of the power system.

[0101] The EMD algorithm is used to establish a data cleaning model, which includes:

[0102] Improve the EMD algorithm and receive the processed real-time monitoring data of the mechanical status of the circuit breaker;

[0103] The real-time monitoring data of the mechanical state of the circuit breaker is decomposed into multiple intrinsic mode functions through the improved EMD algorithm for screening, the intrinsic mode functions containing noise are removed, and the intrinsic mode functions containing useful information are retained. The screened intrinsic mode functions are superimposed and trained to obtain the data cleaning model.

[0104] By using the improved EMD algorithm to establish a data cleaning model, the characteristics of the real-time monitored circuit breaker status data can be better extracted, and the complex time series real-time monitored circuit breaker status data can be decomposed into a series of intrinsic mode functions. Each intrinsic mode function corresponds to the characteristics on different time scales. This decomposition method helps to capture the local characteristics of the real-time monitored circuit breaker status data, especially when dealing with complex and changeable real-time monitored circuit breaker status data, it can provide more detailed analysis.

[0105] Improvements to the EMD algorithm include:

[0106] The improved wavelet denoising method is used to denoise the real-time monitored circuit breaker state data, and the denoised real-time monitored circuit breaker state data is adaptively decomposed into a series of intrinsic mode function components;

[0107] Select appropriate prediction models for components of different frequencies, and finally superimpose the prediction values ​​of each component to obtain the final prediction value;

[0108] The EMD algorithm is improved by adding Gaussian noise and cubic spline interpolation sampling, and removing the first intrinsic mode function.

[0109] By improving the EMD algorithm, the decomposition accuracy of the data cleaning model is improved, and more accurate decomposition results can be obtained when cleaning the circuit breaker state monitoring data. The improved wavelet denoising method is used to denoise the real-time monitored circuit breaker state data, which can make the spectral peak at the fault characteristic frequency more obvious, thereby improving the accuracy of circuit breaker state monitoring data analysis.

[0110] Use the data cleaning model to decompose all monitoring data one by one, including:

[0111] Column renaming: If there are two column names with the same name or the same meaning in the monitoring data column, rename the name of one of the data columns;

[0112] Delete duplicate values: If there are duplicate data values ​​in the monitoring data column, delete the duplicate data values ​​and keep the first record of the duplicate data when deleting;

[0113] Missing value processing: If there are missing values ​​in the monitoring data column, statistical methods are used to fill in the missing values ​​in the data. By determining the data type, missing value situation, data distribution, etc. in the monitoring data column, the data quality is improved. When processing missing values, different strategies are adopted according to the number of missing values ​​and attribute importance. For example, when the missing rate is low and the attribute importance is low, the mean or median can be used to fill in. When the missing rate is high and the attribute importance is high, interpolation or modeling may be required for processing.

[0114] Consistency processing: If there are inconsistent data or naming rules in the data column, use the split column function to split the data values ​​in the inconsistent data column;

[0115] Data sorting and processing: filter and sort the data in the data column, and identify and process outliers in the data to ensure the accuracy and reliability of the data;

[0116] Clustering of all monitoring data decomposition results, including:

[0117] Multi-scale entropy is used to extract the multi-dimensional feature data of the circuit breaker mechanical state in the monitoring data column, and features useful for cluster analysis are extracted, which helps to reduce the dimension of the data while ensuring that the extracted features can best represent the essence of the data.

[0118] The features of different dimensions or magnitudes in the multidimensional feature data of the mechanical state of the circuit breaker are converted to the same scale. Through standardization, the features of different dimensions or magnitudes can be converted to the same scale, thereby ensuring that the clustering algorithm is more fair and effective in processing these features;

[0119] The K-Means clustering algorithm is used to cluster the feature information in the multidimensional feature data of the mechanical state of the circuit breaker after feature transformation. K-Means is a common clustering algorithm that attempts to divide the data into K clusters so that the data in each cluster is as similar as possible, while the data in different clusters are as different as possible. Through iterative optimization, the K-Means algorithm attempts to minimize the intra-cluster variance, thereby achieving the purpose of clustering.

[0120] The advantages of using a data cleaning model to cluster data decomposition results mainly include improving data quality and accuracy, improving the effectiveness and reliability of data analysis, and saving time and resources. Through data cleaning, invalid, duplicate and erroneous data can be removed to make the data more accurate and reliable, which not only improves the overall quality of the data, but also provides a more reliable foundation for subsequent data analysis, making the cleaned data more in line with the actual situation, which is also more accurate and reliable for the results of data analysis, thereby ensuring the reliability and effectiveness of data analysis, allowing clustering algorithms to more accurately identify patterns and structures in the data, thereby improving the quality and effectiveness of clustering. The cleaned data is cleaner and tidier, which can save a lot of unnecessary work, such as manual checking and correcting erroneous data, thereby saving time and resources and making the data analysis process more efficient.

[0121] The multi-dimensional feature data of the mechanical state of the circuit breaker in the monitoring data column is extracted by using multi-scale entropy, including:

[0122] Inputting the multi-dimensional characteristic data of the mechanical state of the circuit breaker into the empirical mode decomposition algorithm, and initializing the parameters of the empirical mode decomposition algorithm to obtain multiple target decomposition parameters;

[0123] Performing modal decomposition on the multi-dimensional characteristic data of the mechanical state of the circuit breaker according to multiple target decomposition parameters to obtain multiple intrinsic modal component data;

[0124] The multi-scale entropy algorithm is used to analyze the characteristics of multiple intrinsic mode component data respectively, and the multi-scale entropy value characteristics of each intrinsic mode component data are obtained;

[0125] The multi-scale entropy value features of each intrinsic mode component data are classified by a support vector machine model to obtain the feature classification result of each intrinsic mode component data;

[0126] Determine multiple inherent mode functions of multidimensional characteristic data of the mechanical state of the circuit breaker according to the characteristic classification result of each intrinsic mode component data; iteratively analyze the multiple inherent mode functions to obtain multiple components of the multidimensional characteristic data of the mechanical state of the circuit breaker;

[0127] State features are extracted from components of a plurality of multi-dimensional feature data of mechanical states of circuit breakers to obtain state features of each of the plurality of multi-dimensional feature data of mechanical states of circuit breakers.

[0128] By using multi-scale entropy to analyze the real-time monitored circuit breaker status data at different time scales, the behavioral characteristics of the circuit breaker at different time scales can be captured. For example, the circuit breaker may experience frequent operations in a short period of time, but may exhibit stable performance on a long time scale. At the same time, since the circuit breaker may be affected by many factors during operation, including environmental factors, frequency of use, etc., multi-scale entropy can more accurately reflect the impact of these factors on the performance of the circuit breaker, thereby improving the accuracy of fault diagnosis. Through multi-scale entropy analysis, potential problems of the circuit breaker can be discovered earlier, so that maintenance measures can be taken in advance to avoid equipment failures, extend equipment service life, reduce maintenance costs, effectively improve data quality, ensure that the data retained after data cleaning better reflects the operating status of the circuit breaker, and improve the accuracy of circuit breaker fault diagnosis.

[0129] A circuit breaker mechanical state monitoring data cleaning method of the present invention removes interference from circuit breaker state monitoring data collected in real time, so that the cleaned data is cleaner and neater, and a lot of unnecessary work can be saved, thereby saving time and resources. By using the improved EMD algorithm to establish a data cleaning model, the characteristics of the circuit breaker state data monitored in real time can be better extracted, and the circuit breaker state data monitored in real time in a complex time series can be decomposed into a series of intrinsic mode functions, each intrinsic mode function corresponds to characteristics on different time scales. This decomposition method is helpful to capture the local characteristics of the circuit breaker state data monitored in real time, especially when processing complex and changeable circuit breaker state data monitored in real time, it can provide more detailed analysis, and through multi-scale entropy analysis, potential problems of the circuit breaker can be discovered earlier, so that maintenance measures can be taken in advance, equipment failures can be avoided, equipment service life can be extended, maintenance costs can be reduced, data quality can be effectively improved, and it is ensured that the data retained after data cleaning better reflects the operating status of the circuit breaker, and the accuracy of circuit breaker fault diagnosis is improved.

[0130] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0131] While the embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that many changes, modifications, substitutions and variations can be made to the embodiments without departing from the principles and spirit of the invention.

Claims

1. A method for cleaning data of mechanical condition monitoring of a circuit breaker, characterized in that: The following steps are involved: Step 1: Obtain historical circuit breaker mechanical condition monitoring data, and establish a data cleaning model using the EMD algorithm based on the acquired historical circuit breaker mechanical condition monitoring data; Step 2: Acquire the real-time monitoring data of the mechanical state of the circuit breaker, process the acquired monitoring data, and store the monitoring data after the processing to form a monitoring data column; Step 3: Input the monitoring data column into the data cleaning model, and use the data cleaning model to decompose all monitoring data one by one; Step 4: Save all monitoring data decomposition results, and cluster all monitoring data decomposition results to obtain multiple clustered data sets; Step 5: Remove the noise point data outside the multiple clustered data sets to obtain the cleaned circuit breaker mechanical condition monitoring data.

2. A circuit breaker mechanical condition monitoring data cleaning method according to claim 1, characterized in that: The circuit breaker mechanical state monitoring data includes multi-dimensional characteristic data of the circuit breaker mechanical state, transmission mechanism data and energy storage mechanism data. The multi-dimensional characteristic data of the circuit breaker mechanical state includes opening and closing data, travel and time characteristics, circuit breaker state data of real-time vibration monitoring, coil current and speed. The collection of opening and closing data specifically includes: using a timer to collect the opening and closing time of the circuit breaker, using a photoelectric travel sensor to collect the opening and closing displacement of the circuit breaker, and using an audio collector to collect the opening and closing sound data of the circuit breaker.

3. A circuit breaker mechanical condition monitoring data cleaning method according to claim 2, characterized in that: The step of obtaining real-time monitoring data of the mechanical state of the circuit breaker and processing the obtained monitoring data specifically includes: Real-time collection of multi-dimensional characteristic data of the mechanical state of the circuit breaker, transmission mechanism data and energy storage mechanism data of the circuit breaker; The communication interference, white noise interference and random pulse interference in the collected data are removed; the data after the interference is removed is enhanced so that the characteristic parameters in the data after the interference is removed are enhanced.

4. A circuit breaker mechanical condition monitoring data cleaning method according to claim 3, characterized in that: The step of obtaining real-time monitoring data of the mechanical state of the circuit breaker and processing the obtained monitoring data further includes: Extract the data information after communication interference, white noise interference and random pulse interference are removed from the data; Retrieving the corresponding signal-to-noise ratio parameters in the multi-dimensional characteristic data of the mechanical state of the circuit breaker, the transmission mechanism data and the energy storage mechanism data after the communication interference, white noise interference and random pulse interference in the completion data are removed; The signal-to-noise ratio evaluation coefficient is obtained by using the corresponding signal-to-noise ratio parameters in the multi-dimensional characteristic data of the mechanical state of the circuit breaker, the transmission mechanism data and the energy storage mechanism data; wherein the signal-to-noise ratio evaluation coefficient is obtained by the following formula: Wherein, S represents the signal-to-noise ratio evaluation coefficient; n represents the number of data contained in the multidimensional characteristic data of the mechanical state of the circuit breaker, the transmission mechanism data and the energy storage mechanism data after the communication interference, white noise interference and random pulse interference in the data are removed; S hi represents the signal-to-noise ratio of the i-th data after denoising; S xi represents the signal-to-noise ratio of the i-th data before denoising; S xb represents the standard deviation of the signal-to-noise ratio corresponding to n data of the multidimensional characteristic data of the mechanical state of the circuit breaker, the transmission mechanism data and the energy storage mechanism data before denoising; S hb represents the signal-to-noise ratio standard deviation corresponding to n data of the multidimensional characteristic data of the mechanical state of the circuit breaker, the transmission mechanism data and the energy storage mechanism data after denoising; S hmax and S hmin represents the maximum value of the signal-to-noise ratio and the minimum value of the signal-to-noise ratio corresponding to n data after denoising of the multi-dimensional characteristic data of the mechanical state of the circuit breaker, the transmission mechanism data and the energy storage mechanism data; S xmax and S xmin Indicates the maximum value of the signal-to-noise ratio and the minimum value of the signal-to-noise ratio corresponding to n data of the multidimensional characteristic data of the mechanical state of the circuit breaker, the transmission mechanism data and the energy storage mechanism data before denoising; Comparing the signal-to-noise ratio evaluation coefficient with a preset evaluation coefficient threshold; When the signal-to-noise ratio evaluation coefficient is lower than a preset evaluation coefficient threshold, denoising is performed on the multidimensional feature data of the mechanical state of the circuit breaker, the transmission mechanism data and the energy storage mechanism data after the communication interference, the white noise interference and the random pulse interference in the completed data until the signal-to-noise ratio evaluation coefficient corresponding to the multidimensional feature data of the mechanical state of the circuit breaker, the transmission mechanism data and the energy storage mechanism data after the denoising exceeds the preset evaluation coefficient threshold; When the signal-to-noise ratio evaluation coefficient is not lower than a preset evaluation coefficient threshold, a data distortion determination is performed, and a data quality abnormality alarm is issued when the data distortion is too large and the data quality is lower than a preset data quality requirement.

5. A circuit breaker mechanical condition monitoring data cleaning method according to claim 4, characterized in that: When the signal-to-noise ratio evaluation coefficient is not lower than the preset evaluation coefficient threshold, a data distortion determination is performed, and when the data distortion is too large and the data quality is lower than the preset data quality requirement, a data quality abnormality alarm is issued, including: When the signal-to-noise ratio evaluation coefficient is lower than a preset evaluation coefficient threshold, extracting the data gain ratio of the multi-dimensional feature data of the mechanical state of the circuit breaker, the transmission mechanism data and the energy storage mechanism data; Obtaining a distortion determination parameter by combining the multidimensional characteristic data of the mechanical state of the circuit breaker, the transmission mechanism data and the energy storage mechanism data after denoising using the data gain ratio of the multidimensional characteristic data of the mechanical state of the circuit breaker, the transmission mechanism data and the energy storage mechanism data; The distortion determination parameter is obtained by the following formula: Wherein, Z represents the distortion determination parameter; n represents the number of data contained in the multi-dimensional characteristic data of the mechanical state of the circuit breaker, the transmission mechanism data and the energy storage mechanism data after the communication interference, white noise interference and random pulse interference in the data are removed; S hi represents the signal-to-noise ratio of the i-th data after denoising; S xi represents the signal-to-noise ratio of the i-th data before denoising; Z i Indicates the data gain ratio after the ith data is denoised; Z c Indicates the preset data gain ratio reference value; S hb represents the standard deviation of the signal-to-noise ratio corresponding to n data of the multidimensional characteristic data of the mechanical state of the circuit breaker, the transmission mechanism data and the energy storage mechanism data after denoising; Z b represents the data gain ratio standard deviation corresponding to n data of the multi-dimensional characteristic data of the mechanical state of the circuit breaker, the transmission mechanism data and the energy storage mechanism data after denoising; Z min represents the minimum value of the data gain ratio corresponding to n data of the multi-dimensional characteristic data of the mechanical state of the circuit breaker, the transmission mechanism data and the energy storage mechanism data after denoising; S zx Indicates the signal-to-noise ratio change ratio of the data corresponding to the minimum value of the data gain ratio before and after the noise removal processing; Comparing the distortion determination parameter with a preset parameter threshold; When the distortion determination parameter exceeds a preset parameter threshold, it is determined that the data distortion is too large, resulting in the data quality being lower than a preset data quality requirement; When data distortion is too large and the data quality is lower than the preset data quality requirements, a data quality abnormality alarm will be issued.

6. A circuit breaker mechanical condition monitoring data cleaning method according to claim 1, characterized in that: The data cleaning model is established by using the EMD algorithm, specifically including: Improve the EMD algorithm and receive the processed real-time monitoring data of the mechanical status of the circuit breaker; The real-time monitoring data of the mechanical state of the circuit breaker is decomposed into multiple intrinsic mode functions through the improved EMD algorithm for screening, the intrinsic mode functions containing noise are removed, and the intrinsic mode functions containing useful information are retained; The filtered intrinsic mode functions are superimposed, and the superimposed intrinsic mode functions are superimposed for training to obtain a data cleaning model.

7. A circuit breaker mechanical condition monitoring data cleaning method according to claim 6, characterized in that: The improvement of the EMD algorithm specifically includes: The improved wavelet denoising method is used to denoise the real-time monitored circuit breaker state data, and the denoised real-time monitored circuit breaker state data is adaptively decomposed into a series of intrinsic mode function components; Select appropriate prediction models for components of different frequencies, and finally superimpose the prediction values ​​of each component to obtain the final prediction value; The EMD algorithm is improved by adding Gaussian noise and cubic spline interpolation sampling, and removing the first intrinsic mode function.

8. A circuit breaker mechanical condition monitoring data cleaning method according to claim 1, characterized in that: The data cleaning model is used to decompose all monitoring data one by one, specifically including: Column renaming: If there are two column names with the same name or the same meaning in the monitoring data column, rename the name of one of the data columns; Delete duplicate values: If there are duplicate data values ​​in the monitoring data column, delete the duplicate data values ​​and keep the first record of the duplicate data when deleting; Missing value processing: If there are missing values ​​in the data in the monitoring data column, use statistical methods to fill in the missing values ​​in the data; Consistency processing: If there are inconsistent data or naming rules in the data column, use the split column function to split the data values ​​in the inconsistent data column; Data sorting and processing: filter and sort the data in the data column, and identify and process outliers in the data.

9. A circuit breaker mechanical condition monitoring data cleaning method according to claim 1, characterized in that: The clustering of the decomposition results of all monitoring data specifically includes: The multi-dimensional feature data of the mechanical state of the circuit breaker in the monitoring data column is extracted by using multi-scale entropy to extract features useful for cluster analysis. Convert the features of different dimensions or magnitudes in the multi-dimensional feature data of the mechanical state of the circuit breaker to the same scale; The K-Means clustering algorithm is used to cluster the feature information in the multi-dimensional feature data of the circuit breaker mechanical state after feature transformation.

10. A circuit breaker mechanical condition monitoring data cleaning method according to claim 9, characterized in that: The method of extracting the multi-dimensional feature data of the mechanical state of the circuit breaker in the monitoring data column by using multi-scale entropy specifically includes: Inputting the multi-dimensional characteristic data of the mechanical state of the circuit breaker into the empirical mode decomposition algorithm, and initializing the parameters of the empirical mode decomposition algorithm to obtain multiple target decomposition parameters; Performing modal decomposition on the multi-dimensional characteristic data of the mechanical state of the circuit breaker according to multiple target decomposition parameters to obtain multiple intrinsic modal component data; The multi-scale entropy algorithm is used to analyze the characteristics of multiple intrinsic mode component data respectively, and the multi-scale entropy value characteristics of each intrinsic mode component data are obtained; The multi-scale entropy value features of each intrinsic mode component data are classified by a support vector machine model to obtain the feature classification result of each intrinsic mode component data; Determine multiple inherent mode functions of multidimensional characteristic data of the mechanical state of the circuit breaker according to the characteristic classification result of each intrinsic mode component data; iteratively analyze the multiple inherent mode functions to obtain multiple components of the multidimensional characteristic data of the mechanical state of the circuit breaker; State features are extracted from components of a plurality of multi-dimensional feature data of mechanical states of circuit breakers to obtain state features of each of the plurality of multi-dimensional feature data of mechanical states of circuit breakers.

Images