A method for detecting mechanical and insulation status of transformer windings based on underdamped oscillation characteristics

Abstract

The present invention discloses a transformer winding mechanical and insulation state detection method based on underdamped oscillation characteristics. The underdamped oscillation wave data of the transformer winding is obtained by an underdamped oscillation curve test system. First, the oscillation fluctuation factor W is used to ij Determine the damped oscillation characteristic analysis band, and then calculate the multi-dimensional characteristic relationship WC to diagnose whether the winding is faulty. After determining that the winding is faulty, perform image matrix reconstruction on the damped oscillation characteristic analysis band, calculate the average roughness WM of the reconstructed image, and distinguish between mechanical faults and insulation faults in the winding.

Description

Technical Field

[0001] The present application relates to the technical field of fault diagnosis of electric power equipment, and in particular to a method for detecting the mechanical and insulation status of transformer windings based on underdamped oscillation characteristics. Background Art

[0002] As a key equipment in the power system, the safe and stable operation of the transformer is particularly important for maintaining the order of national production and life. However, since the transformer winding is susceptible to mechanical stress or short-circuit impact, the mechanical structure and insulation state of the winding deteriorate, which gradually evolves into winding mechanical failure and insulation failure under the cumulative effect, further affecting the normal operation of the transformer. In severe cases, it can cause fire and explosion in the substation. Through research, it was found that more than 40% of transformer failures are caused by winding failures. Therefore, effective detection of the mechanical and insulation state of the transformer winding can greatly reduce serious accidents of the transformer.

[0003] The transformer winding can be regarded as a lumped circuit parameter equivalent network composed of inductance, resistance, capacitance and other parameters. When the mechanical and insulation state of the transformer changes, the inductance, resistance and capacitance parameters in the winding equivalent network will change accordingly, so that the winding fault condition can be judged by the underdamped oscillation wave characteristics of the winding. However, the current technology related to the winding oscillation wave cannot accurately distinguish between the mechanical fault and insulation fault of the winding. To solve this problem, the present invention introduces a transformer winding mechanical and insulation state detection method based on underdamped oscillation characteristics, which has the advantages of accurate detection of the winding mechanical and insulation state and clear process, and is of great significance for ensuring the operational reliability of the power system. Summary of the invention

[0004] The present application provides a method for detecting the mechanical and insulation status of transformer windings based on underdamped oscillation characteristics. According to the proposed method, the mechanical and insulation status of transformer windings can be accurately and effectively determined.

[0005] The present application provides a method for detecting mechanical and insulation states of transformer windings based on underdamped oscillation characteristics, the method comprising:

[0006] (1) Through the underdamped oscillation curve test system, the test time range is set to 0~0.1ms, and the time series is obtained: The voltage data sequence of the winding underdamped oscillation curve under , N represents the total number of sampling points of the underdamped oscillation curve;

[0007] (2) Calculate the oscillation fluctuation factor of the underdamped oscillation curve:

[0008] (1)

[0009] Among them, W ij is the oscillation fluctuation factor between the oscillation stable point combination i and j of the underdamped oscillation curve, i=0, 1,2,…, j=i+1; U b is the attenuation limit stability value of the underdamped oscillation curve; U fmax is the voltage maximum value in the first oscillation stable point combination; U fmin is the voltage minimum value in the second oscillation stable point combination; k1 and k2 are oscillation limit compensation constant factors related to the transformer voltage level and capacity; when When the under-damped oscillation is judged to have not terminated, let i=i+1 and repeatedly calculate the oscillation fluctuation factor of the under-damped oscillation curve; when When the under-damped oscillation is terminated, the oscillation wave curve band containing the first j voltage trough values ​​is selected as the damped oscillation characteristic analysis band; It is a constant related to the transformer grade and capacity;

[0010] (3) Based on the selected damped oscillation characteristic analysis band, calculate the multi-dimensional characteristic relationship WC:

[0011] (2)

[0012] Where, q is the total number of oscillation stable points in the damped oscillation characteristic analysis band, p=q-1, i=0, 1, 2,…, j=i+1; N t is the number of sampling points of the curve in the damped oscillation characteristic analysis band, U nt is the voltage value of the nth data in the damped oscillation characteristic analysis band; when When the transformer winding to be tested is normal, It is judged that there is a fault in the transformer winding; It is a normalization constant related to the voltage level and capacity of the transformer;

[0013] (4) After determining that the transformer winding is faulty, the damped oscillation characteristic analysis band is subjected to image matrix reconstruction, as shown in the following formula:

[0014] (3)

[0015] (4)

[0016] Where X and Y are the time and voltage values ​​of the data points in the damped oscillation characteristic analysis band, respectively, and Y=f(X) is the relationship between the reconstructed curve. i ,y i and m are the time value, voltage value and total number of minimum points corresponding to the i-th minimum point in the damped oscillation characteristic analysis band, c iis the i-th segment coefficient of the reconstructed curve, k is the constant of the minimum point number; A is the image matrix surrounded by the damped oscillation characteristic analysis band and the reconstructed curve, B is the structural element, E M is an M-dimensional space, d and v are element variables, a and b are sub-element units;

[0017] (5) Calculate the average roughness WM of the reconstructed damped oscillation feature analysis band image:

[0018] (5)

[0019] (6)

[0020] (7)

[0021] Where k is the pixel number exponential factor of the active window of the image matrix, H (i, j) is the grayscale value of the pixel located at (i, j) in the image matrix, x, w and y, l are the width variable, effective width value and height variable, effective height value of the reconstructed image respectively; when When the transformer winding is judged to have only mechanical fault; when When the transformer winding has mechanical and insulation faults at the same time, it is judged that It is judged that there is only insulation fault in the transformer winding; , It is a normalization constant related to the voltage level and capacity of the transformer.

[0022] The beneficial effect of the present invention is that after obtaining the under-damped oscillation wave data of the transformer winding through the under-damped oscillation curve test system, the oscillation fluctuation factor W is firstly used to ij Determine the damped oscillation characteristic analysis band, and then calculate the multi-dimensional characteristic relationship WC to diagnose whether the winding is faulty; after judging that the winding is faulty, perform image matrix reconstruction on the damped oscillation characteristic analysis band, calculate the average roughness WM of the reconstructed image, and distinguish between mechanical faults and insulation faults of the winding; provide a reference for the evaluation of the mechanical and insulation status of the transformer winding, and improve the detection accuracy of the winding status. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 is a flow chart of the method of the present invention;

[0024] Figure 2 The invention relates to an underdamped oscillation curve testing system of the method of the invention. DETAILED DESCRIPTION

[0025] The present invention is further described in detail below in conjunction with the accompanying drawings:

[0026] like Figure 1As shown, a method for detecting mechanical and insulation status of transformer windings based on underdamped oscillation characteristics, the method comprising:

[0027] (1) Build Figure 2 The underdamped oscillation curve test system shown in the figure mainly comprises a transformer excitation end bushing (1), a transformer response end bushing (2), a winding input port (3), a winding output port (4), a box shell (5) and a PC terminal (6); when the system is working, the underdamped oscillation wave generator generates an excitation signal, and a high-frequency pulse signal is injected from the winding input port (3) through the transformer excitation end bushing (1); the filtering system collects the oscillation wave signal of the winding output port (4) from the transformer response end bushing (2), transmits the data to the oscillation wave data acquisition device after filtering and denoising, and processes and analyzes it through the PC terminal (6); through the underdamped oscillation curve test system, the test time range is set to 0~0.1ms, and the time series is obtained as follows: The voltage data sequence of the winding underdamped oscillation curve under , N represents the total number of sampling points of the underdamped oscillation curve;

[0028] (2) Calculate the oscillation fluctuation factor of the underdamped oscillation curve:

[0029] (1)

[0030] Among them, W ij is the oscillation fluctuation factor between the oscillation stable point combination i and j of the underdamped oscillation curve, i=0, 1,2,…, j=i+1; U b is the attenuation limit stability value of the underdamped oscillation curve; U fmax is the voltage maximum value in the first oscillation stable point combination; U fmin is the voltage minimum value in the second oscillation stable point combination; k1 and k2 are oscillation limit compensation constant factors related to the transformer voltage level and capacity; when When the under-damped oscillation is judged to have not terminated, let i=i+1 and repeatedly calculate the oscillation fluctuation factor of the under-damped oscillation curve; when When the under-damped oscillation is terminated, the oscillation wave curve band containing the first j voltage trough values ​​is selected as the damped oscillation characteristic analysis band; It is a constant related to the transformer grade and capacity;

[0031] (3) Based on the selected damped oscillation characteristic analysis band, calculate the multi-dimensional characteristic relationship WC:

[0032] (2)

[0033] Where, q is the total number of oscillation stable points in the damped oscillation characteristic analysis band, p=q-1, i=0, 1, 2,…, j=i+1; N t is the number of sampling points of the curve in the damped oscillation characteristic analysis band, U nt is the voltage value of the nth data in the damped oscillation characteristic analysis band; when When the transformer winding to be tested is normal, It is judged that there is a fault in the transformer winding; It is a normalization constant related to the voltage level and capacity of the transformer;

[0034] (4) After determining that the transformer winding is faulty, the damped oscillation characteristic analysis band is subjected to image matrix reconstruction, as shown in the following formula:

[0035] (3)

[0036] (4)

[0037] Where X and Y are the time and voltage values ​​of the data points in the damped oscillation characteristic analysis band, respectively, and Y=f(X) is the relationship between the reconstructed curve. i ,y i and m are the time value, voltage value and total number of minimum points corresponding to the i-th minimum point in the damped oscillation characteristic analysis band, c i is the i-th segment coefficient of the reconstructed curve, k is the constant of the minimum point number; A is the image matrix surrounded by the damped oscillation characteristic analysis band and the reconstructed curve, B is the structural element, E M is an M-dimensional space, d and v are element variables, a and b are sub-element units;

[0038] (5) Calculate the average roughness WM of the reconstructed damped oscillation feature analysis band image:

[0039] (5)

[0040] (6)

[0041] (7)

[0042] Where k is the pixel number exponential factor of the active window of the image matrix, H (i, j) is the grayscale value of the pixel located at (i, j) in the image matrix, x, w and y, l are the width variable, effective width value and height variable, effective height value of the reconstructed image respectively; when When the transformer winding is judged to have only mechanical fault; when When the transformer winding has mechanical and insulation faults at the same time, it is judged that It is judged that there is only insulation fault in the transformer winding; , It is a normalization constant related to the voltage level and capacity of the transformer.

Claims

1. A method for detecting mechanical and insulation status of transformer windings based on underdamped oscillation characteristics, characterized in that: (1) Through the underdamped oscillation curve test system, the test time range is set to 0~0.1ms, and the time series is obtained: The voltage data sequence of the winding underdamped oscillation curve under , N represents the total number of sampling points of the underdamped oscillation curve; (2) Calculate the oscillation fluctuation factor of the underdamped oscillation curve: （1） Among them, W ij is the oscillation fluctuation factor between the oscillation stable point combination i and j of the underdamped oscillation curve, i=0, 1, 2,…, j=i+1; U b is the attenuation limit stability value of the underdamped oscillation curve; U fmax is the voltage maximum value in the first oscillation stable point combination; U fmin is the voltage minimum value in the second oscillation stable point combination; k1 and k2 are oscillation limit compensation constant factors related to the transformer voltage level and capacity; when When the under-damped oscillation is judged to have not terminated, let i=i+1 and repeatedly calculate the oscillation fluctuation factor of the under-damped oscillation curve; when When the under-damped oscillation is terminated, the oscillation wave curve band containing the first j voltage trough values ​​is selected as the damped oscillation characteristic analysis band; It is a constant related to the transformer grade and capacity; (3) Based on the selected damped oscillation characteristic analysis band, calculate the multi-dimensional characteristic relationship WC: （2） Where, q is the total number of oscillation stable points in the damped oscillation characteristic analysis band, p=q-1, i=0, 1, 2,…, j=i+1; N t is the number of sampling points of the curve in the damped oscillation characteristic analysis band, U nt is the voltage value of the nth data in the damped oscillation characteristic analysis band; when When the transformer winding to be tested is normal, It is judged that there is a fault in the transformer winding; It is a normalization constant related to the voltage level and capacity of the transformer; (4) After determining that the transformer winding has a fault, the damped oscillation characteristic analysis band is subjected to image matrix reconstruction, as shown in the following formula: （3） （4） Where X and Y are the time and voltage values ​​of the data points in the damped oscillation characteristic analysis band, respectively, and Y=f(X) is the relationship between the reconstructed curve. i ,y i and m are the time value, voltage value and total number of minimum points corresponding to the i-th minimum point in the damped oscillation characteristic analysis band, c i is the i-th segment coefficient of the reconstructed curve, k is the constant of the minimum point number; A is the image matrix surrounded by the damped oscillation characteristic analysis band and the reconstructed curve, B is the structural element, E M is an M-dimensional space, d and v are element variables, a and b are sub-element units; (5) Calculate the average roughness WM of the reconstructed damped oscillation feature analysis band image: （5） （6） （7） Where k is the pixel number exponential factor of the active window of the image matrix, H (i, j) is the grayscale value of the pixel at (i, j) in the image matrix, x, w and y, l are the width variable, effective width value and height variable, effective height value of the reconstructed image respectively; when When the transformer winding is judged to have only mechanical fault; when When the transformer winding has mechanical and insulation faults at the same time, it is judged that It is judged that there is only insulation fault in the transformer winding; , It is a normalization constant related to the voltage level and capacity of the transformer.

Images