Method for monitoring the thermal fault condition of a circuit breaker

Abstract

The application relates to the technical field of power equipment, and is a kind of thermal fault state monitoring method of circuit breaker, which acquires the infrared thermal image of the power distribution cabinet through multiple acquisitions, acquires the circuit breaker area, generates an RGB three-channel image through three-primary color separation of the image of the circuit breaker area, takes the average value of each channel image pixel, takes the average value as the characteristic value of the corresponding color channel of the frame image, and generates the digital signals of the three channels of the preprocessed green, blue and red; after processing the digital signals, the required signals are selected using the correlation analysis method based on the green channel signal, the change of the corresponding ratio is obtained according to the selected signals, and the thermal fault state type is judged according to the temperature change rate. The application can realize the identification of the thermal fault state of the circuit breaker in the power distribution cabinet, improve the detection efficiency and the accuracy of the detection result, ensure the normal operation of the circuit breaker, and thus the stability of the power grid operation can be improved.

Description

Technical Field

[0001] This invention relates to the field of power equipment technology, and is a method for monitoring the thermal fault status of circuit breakers. Background Technology

[0002] Circuit breakers are used to distribute electrical energy and provide a certain degree of protection for power lines. High-voltage circuit breakers can not only interrupt or close no-load current and load current in high-voltage circuits, but also, when a system fault occurs, interrupt overload current and short-circuit current through the action of relay protection devices. They have a fairly complete arc-extinguishing structure and sufficient breaking capacity. Therefore, ensuring the normal operation of circuit breakers is an important guarantee for the stable operation of the power grid.

[0003] Currently, the thermal fault status of circuit breakers in distribution cabinets is detected manually. This method is inefficient, relies on manual judgment of results, requires highly skilled personnel, and suffers from poor accuracy in interpreting the results. This affects the normal operation of circuit breakers and consequently impacts the stability of the power grid. Therefore, accurate monitoring of the thermal fault status of circuit breakers in distribution cabinets is of paramount importance. Summary of the Invention

[0004] This invention provides a method for monitoring the thermal fault status of circuit breakers, which overcomes the shortcomings of the prior art. It can effectively solve the problems of low detection efficiency, high requirements for the professional level of the testing personnel, and poor accuracy of the test results when manually testing circuit breakers in distribution cabinets.

[0005] The technical solution of the present invention is achieved through the following measures: a method for monitoring the thermal fault status of a circuit breaker, comprising the following steps:

[0006] Infrared images of the circuit breaker under different thermal fault states are collected, so that each thermal fault state corresponds to a type of thermal fault state of the circuit breaker.

[0007] Infrared thermal images of the distribution cabinet are acquired and preprocessed to extract the circuit breaker area from each frame of the preprocessed image.

[0008] The image of the circuit breaker area is subjected to three primary color separation to generate an RGB three-channel image. The average value of the pixels in each channel image is taken and the average value is used as the feature value of the frame image in the corresponding color channel to generate the preprocessed digital signals xR(t), xG(t) and xB(t) for the green, blue and red channels.

[0009] The processed digital signals xR(t), xG(t), and xB(t) are standardized to obtain standardized signals. The standardized signals are then used as input signals for blind source separation to generate three new signals, denoted as ya(t), yb(t), and yc(t).

[0010] Signal identification and filtering are performed on ya(t), yb(t), and yc(t), and the required signal is selected using correlation analysis. The change of the corresponding ratio is obtained based on the selected signal. The change of the ratio is the rate of temperature change. The thermal fault state type is determined based on the rate of temperature change.

[0011] The following are further optimizations and / or improvements to the above-mentioned technical solution:

[0012] The above preprocessing can be performed by numerically normalizing the image and then mapping it to the corresponding color pixels to generate a data sequence S. The original value x of the data sequence S is then normalized by Min-max and mapped to a value x* in the interval [0,1], with the formula: x*=(x-Smin) / (Smax-Smin); where Smin and Smax are the minimum and maximum values ​​of the data sequence S, respectively.

[0013] When extracting the circuit breaker region from each frame of the image after the above preprocessing, the specific steps may include: using the K-means unsupervised clustering algorithm to segment the image, labeling the segmented binarized image using the four-neighbor region growing method, and extracting the area and coordinate attributes of the connected components, thus obtaining the circuit breaker region and its coordinates.

[0014] When standardizing the preprocessed original signal, the specific steps may include: assuming the original signal is a real random variable with a mean of zero, the original signal is zero-mean processed using the following formula: x'i(t)=(xi(t)-μi) / σi, where i=1,2,3, μi and σi are the mean and standard deviation of x(t) respectively, and the mean of x'(t) after standardization is zero and the variance is 1.

[0015] When averaging the pixel values ​​of the above-mentioned channels, the following steps may be included: constructing a spatial average sequence Y = [yiR, yiG, yiB]T, where yiR, yiG, and yiB are the R, G, and B channel signals of the pixels in region i after spatial averaging, and T is the time point corresponding to the acquisition of each channel image.

[0016] Infrared images from different locations can be dynamically acquired using a mobile method, with a set overlap rate of 30% to 55% between two adjacent acquired images.

[0017] Images from different locations can be acquired sequentially in a "Z" pattern.

[0018] This invention, based on infrared thermography and image processing technologies, enables the identification of thermal fault states in circuit breakers within distribution cabinets. It eliminates the need for manual inspection of circuit breakers, improving inspection efficiency. The inspection results can be intelligently determined by a computer, requiring less expertise from inspectors and increasing the accuracy of the results. This ensures the normal operation of circuit breakers and thus improves the stability of the power grid. This invention facilitates long-term monitoring of circuit breakers, reflecting the safety status of the distribution cabinet. Personnel can promptly perform maintenance before a major disaster occurs, providing a solid foundation for online monitoring, dynamic prediction, and control of the safety and health of circuit breakers within distribution cabinets. Attached Figure Description

[0019] Appendix Figure 1 This is a flowchart of an embodiment of the present invention. Detailed Implementation

[0020] The present invention is not limited to the following embodiments, and the specific implementation can be determined according to the technical solution of the present invention and the actual situation.

[0021] In this invention, for ease of description, the relative positional relationships of each component are described according to the layout of the accompanying drawings. For example, the positional relationships of front, back, top, bottom, left, right, etc., are determined according to the layout direction of the accompanying drawings.

[0022] The present invention will be further described below with reference to embodiments and accompanying drawings:

[0023] Example 1: As shown in the attached document Figure 1 As shown, the method for monitoring the thermal fault status of the circuit breaker includes the following steps:

[0024] S1. Collect infrared images of the circuit breaker under different thermal fault states, so that each thermal fault state corresponds to a thermal fault state type of the circuit breaker; it should be noted that different thermal fault state types correspond to corresponding infrared images, that is, the temperature change rate corresponding to different thermal fault state types is preset.

[0025] Infrared thermal images of the power distribution cabinet are collected multiple times according to preset time points and preprocessed. The preprocessing involves numerically normalizing the images and then mapping them to the corresponding color pixels to generate a data sequence S. An original value x of the data sequence S is mapped to a value x* in the interval [0,1] through Min-max normalization, as follows: x*=(x-Smin) / (Smax-Smin), where Smin and Smax are the minimum and maximum values ​​of the data sequence S, respectively.

[0026] Extracting the circuit breaker region from each frame of the preprocessed image; specifically including the following steps: using the K-means unsupervised clustering algorithm for image segmentation, labeling the segmented binarized image using the four-neighbor region growing method, and extracting the area and coordinate attributes of the connected components, thus obtaining the circuit breaker region and its coordinates.

[0027] The image of the circuit breaker area is subjected to three-primary-color separation to generate an RGB three-channel image, and the average value of the pixels in each channel is taken. Specifically, a spatial average sequence Y = [yiR, yiG, yiB]T is constructed, where yiR, yiG, and yiB are the R, G, and B channel signals of the pixels in region i after spatial averaging, and T is the time point corresponding to the acquisition of each channel image. The average value is then used as the feature value of the corresponding color channel of the frame image to generate the preprocessed digital signals xR(t), xG(t), and xB(t) for the green, blue, and red channels.

[0028] The processed original digital signals xR(t), xG(t), and xB(t) are standardized to obtain standardized signals. The specific steps include: Assuming the original signals are real random variables with zero mean, the original signals are zero-mean processed using the following formula: x'i(t) = (xi(t) - μi) / σi, where i = 1, 2, 3, μi and σi are the mean and standard deviation of x(t), respectively. The standardized x'(t) has a mean of zero and a variance of 1. The obtained standardized signals are then used as input signals for blind source separation, generating three new signals, denoted as ya(t), yb(t), and yc(t).

[0029] The separated signals ya(t), yb(t), and yc(t) are identified and filtered, and the desired signal is selected using a correlation analysis method based on the green channel signal. Specifically, a correlation curve can be plotted for visual display. In this embodiment, the correlation analysis method is used to perform correlation analysis between the three independent signals after blind source separation and the green channel signal in the source signal. That is, the similarity between the two signals is studied, and the most relevant independent signal is found as the desired signal, thus completing the signal filtering. In this embodiment, a correlation analysis method based on the red or blue channel signals can also be used to select the desired signal, which will not be elaborated here.

[0030] The change in the corresponding ratio is obtained based on the selected signal. The ratio is the normal temperature value of the circuit breaker divided by the fault temperature value. The change in this ratio represents the rate of temperature change, and the type of thermal fault is determined based on this rate of temperature change. Specifically, a rate of temperature change less than 80% indicates a general fault; 80% ≤ rate of temperature change ≤ 95% indicates a severe fault; and a rate of temperature change greater than 95% indicates a critical fault. In this embodiment, the `corrcoef` function in Matlab can be used to calculate and find the ICA signal corresponding to the largest absolute value of the correlation coefficient, which is the selected signal. Analysis using the `corrcoef` function reveals that the first ICA signal is most correlated with the green channel source signal. The change in this ratio is the required rate of temperature change, which is used to determine the thermal fault status of the circuit breaker. This achieves the monitoring of thermal fault status.

[0031] This invention acquires multiple infrared thermal images of the distribution cabinet to obtain the circuit breaker area. The images of the circuit breaker area undergo three-primary-color separation to generate RGB three-channel images. The average value of pixels in each channel is then used as the feature value of the corresponding color channel for that frame, generating pre-processed digital signals for green, blue, and red channels. After standardization and blind source separation of these digital signals, a correlation analysis method based on the green channel signal is used to select the desired signal. The change in the corresponding ratio is obtained based on the selected signal; this change represents the rate of temperature change. The type of thermal fault state is determined based on the rate of temperature change. Infrared thermal imaging (IRT) technology has advantages such as non-contact detection, no electromagnetic interference, safety, reliability, and a large monitoring range. Compared to temperature sensors, thermal imaging cameras can acquire more intuitive two-dimensional thermal distribution images. Pre-processing using image processing techniques allows for further analysis of hot spots, i.e., using the relationship between infrared image temperature and grayscale to perceive abnormal temperature points in the circuit breaker, thereby determining the thermal fault state of the circuit breaker. Infrared thermal imaging technology enables the monitoring of electrical equipment faults without shutdown.

[0032] This invention, based on infrared thermography and image processing technologies, enables the identification of thermal fault states in circuit breakers within distribution cabinets. It eliminates the need for manual inspection of the circuit breakers, improving inspection efficiency. The inspection results can be intelligently determined by a computer, requiring less expertise from the inspectors and increasing the accuracy of the results. This ensures the normal operation of the circuit breakers and thus improves the stability of the power grid. This invention also facilitates long-term monitoring of circuit breakers, reflecting the safety status of the distribution cabinet. Workers can promptly complete maintenance before a major disaster occurs, providing a solid foundation for online monitoring, dynamic prediction, and control of the safety and health of circuit breakers within distribution cabinets.

[0033] Example 2: As shown in the attached document Figure 1 As shown, the thermal fault status monitoring method for this circuit breaker can also utilize mobile dynamic acquisition of infrared images from different locations, with a set overlap rate between adjacent acquisitions. This set overlap rate refers to the probability that temperatures at different locations of the circuit breaker will overlap. In this embodiment, the set overlap rate can be 30% to 55%. Specifically, images from different locations can be acquired sequentially in a "Z" shaped order, which can be achieved using a thermal imager equipped with an infrared lens. The dual-view thermal imager sequentially acquires local images of the GIS shell in a "Z" shaped order and converts the information contained therein into corresponding image visual and data statistical information. This further improves the accuracy of the detection.

[0034] The above technical features constitute the embodiments of the present invention, which have strong adaptability and implementation effect. Unnecessary technical features can be added or removed according to actual needs to meet the needs of different situations.

Claims

1. A method for monitoring the thermal fault status of a circuit breaker, characterized in that... Includes the following steps: Infrared images of the circuit breaker under different thermal fault states are collected, so that each thermal fault state corresponds to a type of thermal fault state of the circuit breaker. Infrared thermal images of the distribution cabinet are acquired and preprocessed to extract the circuit breaker area from each frame of the preprocessed image. The image of the circuit breaker area is subjected to three primary color separation to generate an RGB three-channel image. The average value of the pixels in each channel image is taken and the average value is used as the feature value of the frame image in the corresponding color channel to generate the preprocessed digital signals xR(t), xG(t) and xB(t) for the green, blue and red channels. The processed digital signals xR(t), xG(t), and xB(t) are standardized to obtain standardized signals. The standardized signals are then used as input signals for blind source separation to generate three new signals, denoted as ya(t), yb(t), and yc(t). Correlation analysis is performed on any digital signal among xR(t), xG(t), and xB(t) with ya(t), yb(t), and yc(t), respectively. The signal most correlated with ya(t), yb(t), and yc(t) is selected as the required signal. The change of the corresponding ratio is obtained based on the selected signal, where the ratio is the normal value of the circuit breaker temperature / the fault value of the temperature. The thermal fault state type is determined based on the change of the ratio.

2. The method for monitoring the thermal fault status of a circuit breaker according to claim 1, characterized in that... The preprocessing involves numerically normalizing the image and then mapping it to the corresponding color pixels to generate a data sequence S. An original value x of the data sequence S is then normalized using Min-max and mapped to a value x* in the interval [0,1], with the formula: x*=(x-Smin) / (Smax-Smin); where Smin and Smax are the minimum and maximum values ​​of the data sequence S, respectively.

3. The method for monitoring the thermal fault status of a circuit breaker according to claim 1 or 2, characterized in that... When extracting the circuit breaker region from each frame of the preprocessed image, the specific steps include: using the K-means unsupervised clustering algorithm for image segmentation, labeling the segmented binarized image using the four-neighborhood region growing method, and extracting the area and coordinate attributes of the connected components to obtain the circuit breaker region and its coordinates.

4. The method for monitoring the thermal fault status of a circuit breaker according to claim 1 or 2, characterized in that... The standardization process for the preprocessed original signal includes the following steps: Assuming the original signal is a real random variable with a mean of zero, the original signal is zero-mean processed using the following formula: x'i(t)=(xi(t)-μi) / σi, where i=1,2,3, μi and σi are the mean and standard deviation of x(t) respectively, and the mean of x'(t) after standardization is zero and the variance is 1.

5. The method for monitoring the thermal fault status of a circuit breaker according to claim 3, characterized in that... The standardization process for the preprocessed original signal includes the following steps: Assuming the original signal is a real random variable with a mean of zero, the original signal is zero-mean processed using the following formula: x'i(t)=(xi(t)-μi) / σi, where i=1,2,3, μi and σi are the mean and standard deviation of x(t) respectively, and the mean of x'(t) after standardization is zero and the variance is 1.

6. The method for monitoring the thermal fault status of a circuit breaker according to claim 1, 2, or 5, characterized in that... The process of averaging the pixels of each channel image includes the following steps: constructing a spatial averaging sequence Y = [yiR, yiG, yiB]T, where yiR, yiG, and yiB are the R, G, and B channel signals of the pixels in region i after spatial averaging, and T is the time point corresponding to the acquisition of each channel image.

7. The method for monitoring the thermal fault status of a circuit breaker according to claim 4, characterized in that... The process of averaging the pixels of each channel image includes the following steps: constructing a spatial averaging sequence Y = [yiR, yiG, yiB]T, where yiR, yiG, and yiB are the R, G, and B channel signals of the pixels in region i after spatial averaging, and T is the time point corresponding to the acquisition of each channel image.

8. The method for monitoring the thermal fault status of a circuit breaker according to claim 1, 2, 5, or 7, characterized in that... Infrared images from different locations are acquired dynamically using a mobile method, with a set overlap rate of 30% to 55% between two adjacent acquired images.

9. The method for monitoring the thermal fault status of a circuit breaker according to claim 8, characterized in that... Images from different locations were acquired sequentially in a "Z" pattern.

Images