Vehicle-mounted insulator pollution grade evaluation method based on leakage current and optimized SPD image

Abstract

The invention discloses a vehicle-mounted insulator pollution grade evaluation method based on leakage current and an optimized SPD image. The method is mainly suitable for rapid and accurate evaluation of the pollution grade of a vehicle-mounted insulator of a motor train unit. The method comprises the following specific steps: (1) manufacturing insulator samples under different pollution grades; (2) measuring a leakage current time domain signal of the insulator sample and carrying out normalization processing; (3) carrying out variational mode decomposition on the leakage current to obtain six different intrinsic mode components; (4) obtaining optimized SPD images of different insulator samples; (5) obtaining an optimized SPD image of the insulator to be detected and calculating a pollution correlation coefficient; and (6) judging the pollution grade of the vehicle-mounted insulator according to the pollution correlation coefficient. The method is simple and efficient, can be used for accurately and quickly evaluating the pollution grade of the vehicle-mounted insulator under the condition of no power failure, and has important significance for reducing the train risk caused by the overhigh pollution grade of the insulator.

Description

Technical Field

[0001] The invention belongs to the field of insulator pollution level detection and evaluation, and in particular relates to a vehicle-mounted insulator pollution level evaluation method based on leakage current and optimized SPD image. Background Art

[0002] On-board insulators are key components of high-voltage equipment in EMUs, mainly providing support and insulation for the lines, effectively ensuring the safe and stable operation of the EMUs. Compared with fixed insulators in the power grid, due to the high-speed movement and complex operation routes of the EMUs, on-board insulators in EMUs are more susceptible to contamination by various pollutants, resulting in a continuous decline in their insulation performance. As the pollution state increases, severely polluted insulators may induce surface flashover in humidity, light rain or dense fog. Therefore, keeping the surface of on-board insulators clean or at a low level of contamination is essential to ensure the safe and stable operation of the EMUs.

[0003] Regarding the pollution level of insulators, the national standard GB / T16434-1996 provides a clear classification standard based on equivalent salt deposition density (ESDD). Unfortunately, the measurement of ESDD is a complex and time-consuming process that is mainly applicable to laboratory studies, which makes it difficult to evaluate the pollution level of insulators on site, especially considering the limited maintenance time of EMUs and the large number of on-board insulators that need to be inspected and evaluated. Due to the difficulty in determining the pollution level of on-board insulators, the current method involves routine wiping work after the train enters the depot. Generally speaking, the railway department performs this task once a day in bad weather and once every three days or a week in normal weather. This is a passive measure to ensure the safe operation of on-board insulators, resulting in huge material and labor costs.

[0004] Therefore, exploring online intelligent detection and evaluation methods for the contamination status of on-board insulators will help staff to quickly formulate maintenance strategies, while minimizing the risks to trains caused by excessive insulator contamination levels. Summary of the invention

[0005] The vehicle insulator has a close relationship with its leakage current. It has been proven that the degree of insulator contamination can be effectively analyzed by extracting the characteristic parameters of the leakage current. These commonly used parameters include the ratio of the third harmonic content to the fifth harmonic content of the leakage current, the trend fluctuation analysis of the leakage current, and the harmonic component. However, the leakage current of the insulator is a one-dimensional time domain signal in the Cartesian coordinate system. Due to the randomness and non-stationarity of the leakage current signal, it is difficult to fully reflect the contamination state of the insulator by only analyzing its time domain or frequency domain information.

[0006] In order to more quickly and accurately evaluate the pollution level of vehicle-mounted insulators, the present invention provides a method for evaluating the pollution level of vehicle-mounted insulators based on leakage current and optimized SPD images.

[0007] The technical solution of the present invention is as follows:

[0008] Step 1: Prepare insulator samples at different pollution levels

[0009] In the laboratory, the "solid layer method" was used to conduct artificial pollution experiments on insulators, and insulator samples with different equivalent salt deposition densities were made, denoted as S1~S5. The equivalent salt deposition densities were set to 0, 0.03, 0.06, 0.10 and 0.25, corresponding to five situations: no pollution, light pollution, moderate pollution, heavy pollution and severe pollution.

[0010] Step 2: Measure the leakage current time domain signal of the insulator sample and perform normalization

[0011] The step-by-step voltage boost method is used to apply voltage on the high-voltage side of the insulator until the working voltage reaches 25 kV. A high-precision current transformer is embedded in the insulator base, and its output is connected to an oscilloscope to measure the leakage current signal of the insulator sample containing N sampling points, which is recorded as y = (y1, y2, ..., y i ,…,y N ), and normalized according to the following formula:

[0012]

[0013] Where y' is the normalized leakage current signal, y max With y min are the maximum and minimum values ​​of the timing signal y respectively;

[0014] Step 3: Perform variational mode decomposition on the leakage current to obtain six different eigenmode components

[0015] The normalized insulator leakage current signal is subjected to variational mode decomposition, where the mode number K is set to 6 and the quadratic penalty factor α is set to 1000. Then, six intrinsic mode components that can characterize the insulator contamination state are obtained, which are denoted as IMF1 to IMF6. The decomposed leakage current can be expressed as:

[0016]

[0017] Where R(t) is the residual component of the signal after VMD decomposition;

[0018] Step 4: Obtain optimized SPD images for different insulator samples

[0019] Under the parameter configurations of starting angle, drawing gain and time lag coefficient set to 60°, 18° and 0.1*N respectively, the point symmetry transformation method is used to transform IMF1~IMF6 into a polar coordinate system. The number of mirror symmetry planes in the polar coordinate system is 6, corresponding to the data of 6 different modal components respectively. This operation is implemented for each insulator sample to obtain the optimized SPD images under 5 different pollution levels, which are recorded as images A1~A5 respectively.

[0020] Step 5: Obtain the optimized SPD image of the insulator to be tested and calculate the pollution correlation coefficient

[0021] For an insulator to be tested, repeat the operations from step 2 to step 4 to obtain the optimized SPD image of the insulator, recorded as image B, and calculate the pollution correlation coefficient between image B and images A1 to A5, recorded as CC1-CC5. The calculation formula of the pollution correlation coefficient is:

[0022]

[0023] Where CC(A,B) represents the contamination correlation coefficient between image A and image B; m and n are the number of rows and columns of the optimized SDP image matrix respectively; A mn and B mn are the elements in the binary images of image A and image B respectively; and are the means of all elements in the image matrices A and B respectively; the size of CC(A,B) reflects the similarity between the value to be analyzed and the reference value, and its absolute value is between 0 and 1. The larger the CC(A,B) value, the more similar images A and B are;

[0024] Step 6: Determine the pollution level of the vehicle insulator based on the pollution correlation coefficient

[0025] Analyze the similarity between the measured value and the reference value, and find the maximum value among CC1-CC5, that is, determine that the pollution level of the insulator to be tested is consistent with the pollution level of the sample.

[0026] The beneficial effect of the present invention is that by performing variational modal decomposition on the insulator leakage current and converting it into an optimized SPD image, and then analyzing the similarity between the insulator to be tested and the reference value, the pollution level of the on-board insulator is accurately identified. Different from the conventional pollution level assessment method that requires ESDD measurement in the laboratory, the method in this paper is simple and efficient, and can accurately assess the pollution level without power outages, which is conducive to the railway management department to grasp the pollution status of the on-board insulator in a timely manner, and is of great significance for ensuring the safe and stable operation of the EMU. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1It is a flow chart of the present invention. DETAILED DESCRIPTION

[0028] The present invention will be further described below in conjunction with the accompanying drawings:

[0029] Figure 1 The figure shows the flow chart of the method for evaluating the pollution level of vehicle insulators based on leakage current and optimized SPD image. Figure 1 It can be seen that the method flow includes the following steps:

[0030] 1. A method for evaluating the pollution level of vehicle-mounted insulators based on leakage current and optimized SPD image, characterized in that: the leakage current of the insulator under test is measured and converted into an optimized SPD image, and the pollution level of the vehicle-mounted insulator is evaluated through the following steps:

[0031] Step 1: Prepare insulator samples at different pollution levels

[0032] In the laboratory, the "solid layer method" was used to conduct artificial pollution experiments on insulators, and insulator samples with different equivalent salt deposition densities were made, denoted as S1~S5. The equivalent salt deposition densities were set to 0, 0.03, 0.06, 0.10 and 0.25, corresponding to five situations: no pollution, light pollution, moderate pollution, heavy pollution and severe pollution.

[0033] Step 2: Measure the leakage current time domain signal of the insulator sample and perform normalization

[0034] The step-by-step voltage boost method is used to apply voltage on the high-voltage side of the insulator until the working voltage reaches 25 kV. A high-precision current transformer is embedded in the insulator base, and its output is connected to an oscilloscope to measure the leakage current signal of the insulator sample containing N sampling points, which is recorded as y = (y1, y2, ..., y i ,…,y N ), and normalized according to the following formula:

[0035]

[0036] Where y' is the normalized leakage current signal, y max With y min are the maximum and minimum values ​​of the timing signal y respectively;

[0037] Step 3: Perform variational mode decomposition on the leakage current to obtain six different eigenmode components

[0038] The normalized insulator leakage current signal is subjected to variational mode decomposition, where the mode number K is set to 6 and the quadratic penalty factor α is set to 1000. Then, six intrinsic mode components that can characterize the insulator contamination state are obtained, which are denoted as IMF1 to IMF6. The decomposed leakage current can be expressed as:

[0039]

[0040] Where R(t) is the residual component of the signal after VMD decomposition;

[0041] Step 4: Obtain optimized SPD images for different insulator samples

[0042] Under the parameter configurations of starting angle, drawing gain and time lag coefficient set to 60°, 18° and 0.1*N respectively, the point symmetry transformation method is used to transform IMF1~IMF6 into a polar coordinate system. The number of mirror symmetry planes in the polar coordinate system is 6, corresponding to the data of 6 different modal components respectively. This operation is implemented for each insulator sample to obtain the optimized SPD images under 5 different pollution levels, which are recorded as images A1~A5 respectively.

[0043] Step 5: Obtain the optimized SPD image of the insulator to be tested and calculate the pollution correlation coefficient

[0044] For an insulator to be tested, repeat the operations from step 2 to step 4 to obtain the optimized SPD image of the insulator, recorded as image B, and calculate the pollution correlation coefficient between image B and images A1 to A5, recorded as CC1-CC5. The calculation formula of the pollution correlation coefficient is:

[0045]

[0046] Where CC(A,B) represents the contamination correlation coefficient between image A and image B; m and n are the number of rows and columns of the optimized SDP image matrix respectively; A mn and B mn are the elements in the binary images of image A and image B respectively; and are the means of all elements in the image matrices A and B respectively; the size of CC(A,B) reflects the similarity between the value to be analyzed and the reference value, and its absolute value is between 0 and 1. The larger the CC(A,B) value, the more similar images A and B are;

[0047] Step 6: Determine the pollution level of the vehicle insulator based on the pollution correlation coefficient

[0048] Analyze the similarity between the measured value and the reference value, and find the maximum value among CC1-CC5, that is, determine that the pollution level of the insulator to be tested is consistent with the pollution level of the sample.

Claims

1. A method for evaluating the pollution level of vehicle-mounted insulators based on leakage current and optimized SPD image, characterized in that: The leakage current of the insulator under test is measured and converted into an optimized SPD image to evaluate the pollution level of the on-board insulator through the following steps: Step 1: Prepare insulator samples at different pollution levels In the laboratory, the "solid layer method" was used to conduct artificial pollution experiments on insulators, and insulator samples with different equivalent salt deposition densities were made, denoted as S1~S5. The equivalent salt deposition densities were set to 0, 0.03, 0.06, 0.10 and 0.25, corresponding to five situations: no pollution, light pollution, moderate pollution, heavy pollution and severe pollution. Step 2: Measure the leakage current time domain signal of the insulator sample and perform normalization The step-by-step voltage boost method is used to apply voltage on the high-voltage side of the insulator until the working voltage reaches 25 kV. A high-precision current transformer is embedded in the insulator base, and its output is connected to an oscilloscope to measure the leakage current signal of the insulator sample containing N sampling points, which is recorded as y = (y1, y2, ..., y i ,…,y N ), and normalized according to the following formula: Where y' is the normalized leakage current signal, y max With y min are the maximum and minimum values ​​of the timing signal y respectively; Step 3: Perform variational mode decomposition on the leakage current to obtain six different eigenmode components The normalized insulator leakage current signal is subjected to variational mode decomposition, where the mode number K is set to 6 and the quadratic penalty factor α is set to 1000. Then, six intrinsic mode components that can characterize the insulator contamination state are obtained, which are denoted as IMF1 to IMF6. The decomposed leakage current can be expressed as: Where R(t) is the residual component of the signal after VMD decomposition; Step 4: Obtain optimized SPD images for different insulator samples Under the parameter configurations of starting angle, drawing gain and time lag coefficient set to 60°, 18° and 0.1*N respectively, the point symmetry transformation method is used to transform IMF1~IMF6 into a polar coordinate system. The number of mirror symmetry planes in the polar coordinate system is 6, corresponding to the data of 6 different modal components respectively. This operation is implemented for each insulator sample to obtain the optimized SPD images under 5 different pollution levels, which are recorded as images A1~A5 respectively. Step 5: Obtain the optimized SPD image of the insulator to be tested and calculate the pollution correlation coefficient For an insulator to be tested, repeat the operations from step 2 to step 4 to obtain the optimized SPD image of the insulator, recorded as image B, and calculate the pollution correlation coefficient between image B and images A1 to A5, recorded as CC1-CC5. The calculation formula of the pollution correlation coefficient is: Where CC(A,B) represents the contamination correlation coefficient between image A and image B; m and n are the number of rows and columns of the optimized SDP image matrix respectively; A mn and B mn are the elements in the binary images of image A and image B respectively; and are the means of all elements in the image matrices A and B respectively; the size of CC(A,B) reflects the similarity between the value to be analyzed and the reference value, and its absolute value is between 0 and 1. The larger the CC(A,B) value, the more similar images A and B are; Step 6: Determine the pollution level of the vehicle insulator based on the pollution correlation coefficient Analyze the similarity between the measured value and the reference value, and find the maximum value among CC1-CC5, that is, determine that the pollution level of the insulator to be tested is consistent with the pollution level of the sample.

Images