Transformer operation state analysis system and method for transformer substation noise

Abstract

The invention relates to the field of intelligent analysis, and discloses a transformer operation state analysis system and method for transformer substation noise, and the method comprises the steps: collecting the sound signal and noise intensity of a transformer in each region during operation at a set frequency; performing feature extraction on the sound signal of each area to obtain corresponding waterfall spectrogram feature information; inputting the waterfall spectrogram feature information into a pre-trained convolutional neural network model, and outputting a transformer operation state of each region; the transformer operation state of each area in the transformer operation state analysis module is compared with the noise intensity, if the transformer operation state is matched with the noise intensity, it is judged that the current transformer operation state is accurate, and if the transformer operation state is not matched with the noise intensity, it is judged that the current transformer operation state is not accurate, all operation parameters of the transformer are obtained and then analyzed, and transformer operation coefficients are obtained; and updating the current transformer operation state according to the transformer operation coefficient.

Description

Technical Field

[0001] The present invention relates to the field of intelligent analysis, and in particular to a transformer operation status analysis system and method for substation noise. Background Art

[0002] The transformer operation status analysis system of substation noise is a technical system that uses noise signals to monitor the operation status of transformers. This system usually includes the collection, transmission, processing and analysis of noise data, and can evaluate the health status and potential faults of transformers by analyzing the noise characteristics generated during operation.

[0003] However, the existing transformer operating status analysis still has certain defects. For example, for the noise-based transformer status analysis system, the transformer operating status may still be misjudged due to the error of single noise sampling and calculation error. Summary of the invention

[0004] The purpose of the present invention is to provide a transformer operation status analysis system and method for substation noise to solve the above technical problems.

[0005] The purpose of the present invention can be achieved through the following technical solutions: A transformer operation status analysis system for substation noise, comprising: The noise information collection module divides the substation into a corresponding number of areas according to the number of transformers. Each area is equipped with a voiceprint collector and a noise sensor to collect the acoustic signals and noise intensity of the transformers in each area during operation at a set frequency. The noise information processing module extracts the features of the acoustic signal in each area and obtains the corresponding waterfall spectrogram feature information; The transformer operation status analysis module is used to input the waterfall spectrogram feature information into a pre-trained convolutional neural network model and output the transformer operation status of each area; the transformer operation status is normal, overload and partial discharge; The corresponding verification system compares the transformer operating status of each area in the transformer operating status analysis module with the noise intensity. If they match, the current transformer operating status is judged to be accurate. If they do not match, the current transformer operating status is judged to be inaccurate. After obtaining various operating parameters of the transformer, they are analyzed to obtain the transformer operating coefficient, and then the current transformer operating status is updated according to the transformer operating coefficient.

[0006] As a further technical solution, the process of judging whether the transformer operation status of each area matches the noise intensity is as follows: Collect the noise intensity values ​​of each transformer in normal operation, partial discharge and overload during the previous monitoring cycle , , ; Calculate the fluctuation value when the transformer is in normal operation , Fluctuation value during partial discharge , Fluctuation value when overload ; The theoretical noise intensity range when the transformer is in normal operation is determined to be , the theoretical noise intensity range of partial discharge in transformer operation state is , the theoretical noise intensity range when the transformer is in overload operation is ; in, , , They are the mean values ​​of noise intensity when the transformer is in normal operation, partial discharge, and overload conditions respectively; The actual noise intensity value collected when the output transformer is in operation Compare with the three theoretical noise intensity ranges mentioned above; If the actual noise intensity value If it falls within the corresponding theoretical noise intensity range, it is considered a match; Otherwise, it is judged as mismatch.

[0007] As a further technical solution, calculate the fluctuation value when the transformer is in normal operation , Fluctuation value during partial discharge , Fluctuation value when overload The process is: By formula group: ; ; ; The fluctuation values ​​when the transformer is in normal operation are obtained respectively , Fluctuation value of partial discharge in transformer operation state , Fluctuation value when the transformer is in overload operation ; in, is the total number of collections in the previous monitoring cycle, For the Second-rate, , , They are the maximum values ​​of noise intensity of normal transformer operation, partial discharge, and overload in the previous monitoring cycle, respectively. , , They are the minimum values ​​of noise intensity of normal transformer operation, partial discharge, and overload in the previous monitoring cycle, respectively. , , is the preset scale factor.

[0008] As a further technical solution, the process of obtaining the transformer operating coefficient by analyzing the various operating parameters of the transformer is as follows: Obtain various operating parameters of the transformer during a collection process, including the overall temperature rise , output current , output voltage , oil temperature , partial discharge ; By formula:

[0009] Calculate the first coefficient of the current transformer ; By formula:

[0010] Calculate the second coefficient of the current transformer ; By formula:

[0011] Calculate the current transformer operation coefficient ; in, for The curve of partial discharge amount changing with time during the detection period, is a reference curve obtained based on historical data analysis. is the reference value, is the actual partial discharge amount, is the overall temperature of the transformer With ambient temperature The ratio of is expressed as: , is the reference coefficient, is the judgment function about the overall temperature rise, , is the preset weight coefficient.

[0012] As a further technical solution, the judgment function about the overall temperature rise is The acquisition process is: Draw the curve of the overall temperature of the current transformer changing with time ; Get the current time The overall temperature of the transformer under ; make , find the curve The extreme point closest to the current moment , the overall temperature of the transformer corresponding to this extreme point and Make comparisons; like ,but ; like ,but 1.

[0013] As a further technical solution, the process of updating the current transformer operating state according to the transformer operating coefficient is: The calculated transformer operating factor Compare with preset transformer operation threshold Compare; like , it is judged that the current transformer operation state is abnormal, and further analyzed that the actual operation state of the current transformer is partial discharge or overload; Otherwise, it is determined that the current transformer operating state is normal, and the current transformer operating state is updated to normal.

[0014] As a further technical solution, the process of further analyzing the actual operating state of the current transformer as partial discharge or overload is as follows: like , then it is judged that the actual operating state of the current transformer is overload; like , then it is judged that the actual operating state of the current transformer is partial discharge; like and , then it is judged that the actual operating state of the current transformer is partial discharge and overload.

[0015] A transformer operation status analysis method for substation noise, the specific steps are as follows: S1. Divide the substation into a corresponding number of areas according to the number of transformers. Arrange a voiceprint collector and a noise sensor in each area to collect the acoustic signal and noise intensity of the transformer in each area during operation at a set frequency; S2, extracting features from the acoustic signal of each area to obtain corresponding waterfall spectrogram feature information; S3, inputting the waterfall spectrogram feature information into a pre-trained convolutional neural network model, and outputting the transformer operating status of each area; the transformer operating status is normal, overload and partial discharge; S4. Compare the transformer operating status of each area in the transformer operating status analysis module with the noise intensity. If they match, the current transformer operating status is judged to be accurate. If they do not match, the current transformer operating status is judged to be inaccurate. After obtaining various operating parameters of the transformer, they are analyzed to obtain the transformer operating coefficient, and then the current transformer operating status is updated according to the transformer operating coefficient.

[0016] Beneficial effects of the present invention: (1) The transformer operating status is analyzed through the noise information processing module and the transformer operating status analysis module, and divided into three types: normal, overload and partial discharge. The transformer operating status obtained by the analysis is matched with the noise intensity through the corresponding verification coefficient, so that the accuracy of the transformer operating status analysis can be accurately judged. When the transformer operating status is not accurately judged, the transformer operating coefficient is calculated according to the transformer operating parameters, and the judgment of the transformer operating status is adjusted according to the transformer operating coefficient. This can not only achieve the purpose of quickly judging the transformer operating status, but also provide a verification method through the corresponding verification module to quickly judge whether the transformer operating status is accurate. When it is inaccurate, the actual transformer operating status is updated in time. At the same time, it can avoid the computing power occupation caused by the continuous calculation of operating parameters and improve the operation efficiency. (2) By analyzing the noise intensity data under each operating state of the transformer, the corresponding theoretical noise intensity range under each operating state is obtained. By comprehensively analyzing and calculating the noise intensity fluctuations under the three states in each historical monitoring period through standard deviation and range, the fluctuation value of the noise intensity under each operating state is obtained, so as to realize real-time updating of the theoretical noise intensity range corresponding to each operating state, which provides conditions for the accurate matching of the subsequent noise intensity and the transformer operating state analysis module, thereby ensuring the accurate output of the transformer analysis results. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The present invention will be further described below in conjunction with the accompanying drawings.

[0018] Figure 1 is a system structure block diagram of the present invention; Figure 2 It is a diagram of the method steps of the present invention. DETAILED DESCRIPTION

[0019] 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.

[0020] See also Figure 1 As shown, the present invention is a transformer operation status analysis system for substation noise, comprising: The noise information collection module divides the substation into a corresponding number of areas according to the number of transformers. Each area is equipped with a voiceprint collector and a noise sensor to collect the acoustic signals and noise intensity of the transformers in each area during operation at a set frequency. The noise information processing module extracts the characteristics of the acoustic signal in each area to obtain the corresponding waterfall sound spectrogram characteristic information; the extraction process of the waterfall sound spectrogram characteristic information is an existing technology and will not be described in detail here; The transformer operation status analysis module is used to input the waterfall spectrogram feature information into the pre-trained convolutional neural network model, and output the transformer operation status of each area; the transformer operation status is normal, overload and partial discharge; the convolutional neural network model training and establishment process adopts existing technology and does not need to be described in detail; The corresponding verification system compares the transformer operating status of each area in the transformer operating status analysis module with the noise intensity. If they match, the current transformer operating status is judged to be accurate. If they do not match, the current transformer operating status is judged to be inaccurate. After obtaining various operating parameters of the transformer, they are analyzed to obtain the transformer operating coefficient, and then the current transformer operating status is updated according to the transformer operating coefficient.

[0021] In this embodiment, the analysis of the transformer operating status is realized through the noise information processing module and the transformer operating status analysis module, and is divided into three types: normal, overload and partial discharge. The analyzed transformer operating status is matched with the noise intensity through the corresponding verification coefficient, so that the accuracy of the transformer operating status analysis can be accurately judged. When the transformer operating status is judged inaccurately, the transformer operating coefficient is calculated according to the transformer operating parameters, and the judgment of the transformer operating status is adjusted according to the transformer operating coefficient. It can not only achieve the purpose of quickly judging the transformer operating status, but also provide a verification method through the corresponding verification module, so as to quickly judge whether the transformer operating status is accurate, and timely update the actual transformer operating status when it is inaccurate. At the same time, it can avoid the computing power occupation caused by the continuous calculation of operating parameters, thereby improving the operating efficiency.

[0022] The process of judging whether the transformer operating status in each area matches the noise intensity is as follows: Collect the noise intensity values ​​of each transformer in normal operation, partial discharge and overload during the previous monitoring cycle , , ; Calculate the fluctuation value when the transformer is in normal operation , Fluctuation value during partial discharge , Fluctuation value when overload ; The theoretical noise intensity range when the transformer is in normal operation is determined to be , the theoretical noise intensity range of partial discharge in transformer operation state is , the theoretical noise intensity range when the transformer is in overload operation is ; in, , , They are the mean values ​​of noise intensity when the transformer is in normal operation, partial discharge, and overload conditions respectively; The actual noise intensity value collected when the output transformer is in operation Compare with the three theoretical noise intensity ranges mentioned above; If the actual noise intensity value If it falls within the corresponding theoretical noise intensity range, it is considered a match; Otherwise, it is judged as mismatch.

[0023] Calculate the fluctuation value when the transformer is in normal operation , Fluctuation value during partial discharge , Fluctuation value when overload The process is: By formula group: ; ; ; The fluctuation values ​​when the transformer is in normal operation are obtained respectively , Fluctuation value of partial discharge in transformer operation state , Fluctuation value when the transformer is in overload operation ; in, is the total number of collections in the previous monitoring cycle, For the Second-rate, , , They are the maximum values ​​of noise intensity of normal transformer operation, partial discharge, and overload in the previous monitoring cycle, respectively. , , They are the minimum values ​​of noise intensity of normal transformer operation, partial discharge, and overload in the previous monitoring cycle, respectively. , , It is a preset proportionality factor, which is determined based on analysis of historical and empirical data.

[0024] In this embodiment, the noise intensity data under each operating state of the transformer is analyzed to obtain the corresponding theoretical noise intensity range under each operating state, and the noise intensity fluctuations under the three states in each historical monitoring period are comprehensively analyzed and calculated by combining the standard deviation with the range to obtain the fluctuation value of the noise intensity under each operating state, so as to realize real-time updating of the theoretical noise intensity range corresponding to each operating state, which provides conditions for the accurate matching of the subsequent noise intensity and the transformer operating state analysis module, thereby ensuring the accurate output of the transformer analysis results.

[0025] After obtaining the various operating parameters of the transformer and analyzing them, the process of obtaining the transformer operating coefficient is as follows: Obtain various operating parameters of the transformer during a collection process, including the overall temperature rise , output current , output voltage , oil temperature , partial discharge ; By formula:

[0026] Calculate the first coefficient of the current transformer ; By formula:

[0027] Calculate the second coefficient of the current transformer ; By formula:

[0028] Calculate the current transformer operation coefficient ; in, for The curve of partial discharge amount changing with time during the detection period, is a reference curve obtained based on historical data analysis. is the reference value, is the actual partial discharge amount, is the overall temperature of the transformer With ambient temperature The ratio of is expressed as: , is the reference coefficient, is the judgment function about the overall temperature rise, , To preset weight coefficients, it is formulated based on historical data and experimental data; The judgment function about the overall temperature rise The acquisition process is: Draw the curve of the overall temperature of the current transformer changing with time ; Get the current time The overall temperature of the transformer under ; make , find the curve The extreme point closest to the current moment , the overall temperature of the transformer corresponding to this extreme point and Make comparisons; like ,but ; like ,but 1.

[0029] In this embodiment, a method for obtaining a transformer operating coefficient is provided. Specifically, by integrating multivariate data during transformer operation, accurate calculation of the transformer operating coefficient is achieved, thereby ensuring the accuracy of accurate determination of the actual operation of the transformer from the direction of operating parameters, so that the transformer operating status can be updated in real time; specifically, by Calculate the first coefficient of the current transformer , when the transformer is overloaded, its output current will exceed the rated current, causing the current to increase. The bigger, The larger the capacity, the lower the secondary output voltage of the transformer will be. In severe cases, the voltage may drop to zero. Therefore, the output voltage The smaller, The larger the transformer is, the greater the copper loss will be. Overload operation of the transformer will increase the copper loss, which will in turn cause the overall temperature of the transformer to rise. The bigger, The larger the load, the higher the oil temperature of the transformer. The bigger, The bigger; is the overall temperature of the transformer With ambient temperature The ratio of is expressed as: Obviously, the larger the ratio is, the lower the ambient temperature is than the overall temperature of the transformer, so the influence of the ambient temperature on the overall temperature of the transformer is smaller.

[0030] The process of updating the current transformer operating status according to the transformer operating coefficient is: The calculated transformer operating factor Compare with preset transformer operation threshold Compare; like , it is judged that the current transformer operation state is abnormal, and further analysis is made on the actual operation state of the current transformer as partial discharge or overload; the process of further analyzing the actual operation state of the current transformer as partial discharge or overload is as follows: like , then it is judged that the actual operating state of the current transformer is overload; like , then it is judged that the actual operating state of the current transformer is partial discharge; like and , then it is judged that the actual operating state of the current transformer is partial discharge and overload.

[0031] Otherwise, it is determined that the current transformer operating state is normal, and the current transformer operating state is updated to normal.

[0032] In this embodiment, a method for updating the actual operating status of the transformer is provided. First, the transformer operating coefficient Compare with preset transformer operation threshold By comparison, it can quickly determine whether the actual operating status of the transformer is normal. If normal, the transformer status is immediately updated to normal, so there is no need to carry out the next analysis process, which saves time and improves efficiency. In the case of abnormality, the abnormal status is further divided to accurately update the operating status of the transformer.

[0033] See also Figure 2 As shown, a transformer operation status analysis method for substation noise, the specific steps are as follows: S1. Divide the substation into a corresponding number of areas according to the number of transformers. Arrange a voiceprint collector and a noise sensor in each area to collect the acoustic signal and noise intensity of the transformer in each area during operation at a set frequency; S2, extracting features from the acoustic signal of each area to obtain corresponding waterfall spectrogram feature information; S3, inputting the waterfall spectrogram feature information into a pre-trained convolutional neural network model, and outputting the transformer operating status of each area; the transformer operating status is normal, overload and partial discharge; S4. Compare the transformer operating status of each area in the transformer operating status analysis module with the noise intensity. If they match, the current transformer operating status is judged to be accurate. If they do not match, the current transformer operating status is judged to be inaccurate. After obtaining various operating parameters of the transformer, they are analyzed to obtain the transformer operating coefficient, and then the current transformer operating status is updated according to the transformer operating coefficient.

[0034] It should be noted that the calculation formulas and various parameters involved in the calculation in the present invention have been dimensionally processed in advance, and the process of dimensionless processing is well known in the industry and will not be described here.

[0035] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A transformer operation status analysis system for substation noise, characterized in that: include: The noise information collection module divides the substation into a corresponding number of areas according to the number of transformers. Each area is equipped with a voiceprint collector and a noise sensor to collect the acoustic signals and noise intensity of the transformers in each area during operation at a set frequency. The noise information processing module extracts the features of the acoustic signal in each area and obtains the corresponding waterfall spectrogram feature information; The transformer operation status analysis module is used to input the waterfall spectrogram feature information into a pre-trained convolutional neural network model and output the transformer operation status of each area; the transformer operation status is normal, overload and partial discharge; The corresponding verification system compares the transformer operating status of each area in the transformer operating status analysis module with the noise intensity. If they match, the current transformer operating status is judged to be accurate. If they do not match, the current transformer operating status is judged to be inaccurate. After obtaining various operating parameters of the transformer, they are analyzed to obtain the transformer operating coefficient, and then the current transformer operating status is updated according to the transformer operating coefficient.

2. The transformer operation status analysis system for substation noise according to claim 1, characterized in that: The process of judging whether the transformer operating status in each area matches the noise intensity is as follows: Collect the noise intensity values ​​of each transformer in normal operation, partial discharge and overload during the previous monitoring cycle , , ; Calculate the fluctuation value when the transformer is in normal operation , Fluctuation value during partial discharge , Fluctuation value when overload ; The theoretical noise intensity range when the transformer is in normal operation is determined to be , the theoretical noise intensity range of partial discharge in transformer operation state is , the theoretical noise intensity range when the transformer is in overload operation is ; in, , , They are the mean values ​​of noise intensity when the transformer is in normal operation, partial discharge, and overload conditions respectively; The actual noise intensity value collected when the output transformer is in operation Compare with the three theoretical noise intensity ranges mentioned above; If the actual noise intensity value If it falls within the corresponding theoretical noise intensity range, it is considered a match; Otherwise, it is judged as mismatch.

3. The transformer operation status analysis system for substation noise according to claim 2 is characterized in that: Calculate the fluctuation value when the transformer is in normal operation , Fluctuation value during partial discharge , Fluctuation value when overload The process is: By formula group: ； ； ； The fluctuation values ​​when the transformer is in normal operation are obtained respectively , Fluctuation value of partial discharge in transformer operation state , Fluctuation value when the transformer is in overload operation ; in, is the total number of collections in the previous monitoring cycle, For the Second-rate, , , They are the maximum values ​​of noise intensity of normal transformer operation, partial discharge, and overload in the previous monitoring cycle, respectively. , , They are the minimum values ​​of noise intensity of normal transformer operation, partial discharge, and overload in the previous monitoring cycle, respectively. , , is the preset scale factor.

4. The transformer operation status analysis system for substation noise according to claim 3 is characterized in that: After obtaining the various operating parameters of the transformer and analyzing them, the process of obtaining the transformer operating coefficient is as follows: Obtain various operating parameters of the transformer during a collection process, including the overall temperature rise , output current , output voltage , oil temperature , partial discharge ; By formula: Calculate the first coefficient of the current transformer ; By formula: Calculate the second coefficient of the current transformer ; By formula: Calculate the current transformer operation coefficient ; in, for The curve of partial discharge amount changing with time during the detection period, is a reference curve obtained based on historical data analysis. is the reference value, is the actual partial discharge amount, is the overall temperature of the transformer With ambient temperature The ratio of is expressed as: , is the reference coefficient, is the judgment function about the overall temperature rise, , is the preset weight coefficient.

5. The transformer operation status analysis system for substation noise according to claim 4, characterized in that: The judgment function about the overall temperature rise The acquisition process is: Draw the curve of the overall temperature of the current transformer changing with time ; Get the current time The overall temperature of the transformer under ; make , find the curve The extreme point closest to the current moment , the overall temperature of the transformer corresponding to this extreme point and Make comparisons; like ,but ; like ,but 1.

6. The transformer operation status analysis system for substation noise according to claim 1, characterized in that: The process of updating the current transformer operating status according to the transformer operating coefficient is: The calculated transformer operating factor Compare with preset transformer operation threshold Compare; like , it is judged that the current transformer operation state is abnormal, and further analyzed that the actual operation state of the current transformer is partial discharge or overload; Otherwise, it is determined that the current transformer operating state is normal, and the current transformer operating state is updated to normal.

7. The transformer operation status analysis system for substation noise according to claim 6, characterized in that: Further analysis shows that the actual operating state of the current transformer is partial discharge or overload process: like , then it is judged that the actual operating state of the current transformer is overload; like , then it is judged that the actual operating state of the current transformer is partial discharge; like and , then it is judged that the actual operating state of the current transformer is partial discharge and overload.

8. A transformer operation status analysis method for substation noise, the method being applicable to the transformer operation status analysis system for substation noise according to any one of claims 1 to 7, characterized in that: The specific steps are as follows: S1. Divide the substation into a corresponding number of areas according to the number of transformers. Arrange a voiceprint collector and a noise sensor in each area to collect the acoustic signal and noise intensity of the transformer in each area during operation at a set frequency; S2, extracting features from the acoustic signal of each area to obtain the corresponding waterfall spectrogram feature information; S3, inputting the waterfall spectrogram feature information into a pre-trained convolutional neural network model, and outputting the transformer operating status of each area; the transformer operating status is normal, overload and partial discharge; S4. Compare the transformer operating status of each area in the transformer operating status analysis module with the noise intensity. If they match, the current transformer operating status is judged to be accurate. If they do not match, the current transformer operating status is judged to be inaccurate. After obtaining various operating parameters of the transformer, they are analyzed to obtain the transformer operating coefficient, and then the current transformer operating status is updated according to the transformer operating coefficient.

Images