Arc image morphology detection method for multi-chamber arc extinguishing device

Abstract

The invention provides an arc image morphology detection method for a multi-chamber arc extinguishing device, belonging to the technical field of impulse arc image morphology detection; the arc image morphology detection method for the multi-chamber arc extinguishing device is based on a test circuit composed of an impulse current generator, a capacitive voltage divider, a Rogowski coil, a digital storage oscilloscope, a high-speed camera and a data processor; the arc image morphology detection method for the multi-chamber arc extinguishing device extracts the image morphology characteristics of the arc and the distribution of the cut-off arc, and determines the moment when the arc is completely ejected from the chamber according to the total amount of grayscale of the arc. The arc cut-off moment and the position distribution of the cut-off arc are of great significance to studying the dissipation characteristics of the arc, and play a guiding role in optimizing the layout of the chamber nozzle.

Description

Technical Field

[0001] The present invention relates to the technical field of impulse arc image morphology detection, and in particular to an arc image morphology detection method for a multi-chamber arc extinguishing device. Background Art

[0002] Frequent lightning accidents will pose a great threat to the safe operation of the power grid. With the rapid increase in the scale of the power grid and the continuous improvement of reliability requirements, it is imperative to enhance the lightning protection level of the line. During the discharge of thunderclouds to the ground, the rapid change of the electromagnetic field in the space around the discharge channel will generate induced overvoltage on the nearby overhead lines. Thunderclouds may even directly hit the distribution overhead lines, generating extremely high lightning overvoltages. When the lightning overvoltage exceeds a certain value, it will cause the distribution lines to flash over or even break, causing damage to the power supply network, which will directly affect the reliability of the power supply of the distribution system.

[0003] Multi-chamber arc extinguishing devices are used in transmission lines to quench arcs generated by lightning strikes, reducing power equipment failures caused by lightning strikes. In the laboratory, arc quenching tests of multi-chamber arc extinguishing devices are often carried out to study the dissipation characteristics of arcs. During the test, the most important part is to shoot the discharge area, arc development and dissipation process. However, in actual tests, the test personnel mainly rely on visual observation of the entire process from arc generation to dissipation. Since air gap discharge is usually in the microsecond level, the test personnel cannot capture the arc shape at a certain moment in the discharge process with the naked eye. To address this problem, Wuhan University provides a method of using a single-lens reflex viewfinder camera to shoot the arc path instead of visual observation by the human eye. However, due to the insufficient performance of the camera, this method cannot shoot the entire process of arc generation and dissipation at high speed. The shooting time needs to be set manually before shooting, and the interval between each shooting time is long, so it is impossible to start shooting synchronously at the moment of arc appearance. Mei Hongwei from Tsinghua University used the mirror method to shoot the discharge path of the air gap with a high-speed camera. In this method, the distance from the camera to the virtual image in the mirror is twice that of the actual arc, which can easily cause the image to shrink and can only roughly judge the arc path. The shooting methods proposed by the above two scholars lack the technology to extract the arc morphology from the captured arc image, and fail to deeply explore the characteristic quantities of the arc image, such as area, grayscale, number of arcs, etc., and conduct in-depth research on the diffusion of the arc. Summary of the invention

[0004] In order to overcome the shortcomings of the prior art in extracting arc feature quantities for multiple arcs during a multi-chamber arc extinguishing process, the present invention provides an arc image morphology detection method for a multi-chamber arc extinguishing device.

[0005] The technical solution adopted by the present invention is:

[0006] A method for detecting arc image morphology of a multi-chamber arc extinguishing device is based on a test circuit composed of an impulse current generator, a capacitive voltage divider, a Rogowski coil, a digital storage oscilloscope, a high-speed camera and a data processor, wherein the impulse current generator includes an intelligent control system, a voltage regulator, a step-up transformer, a silicon stack, a wave modulation resistor R, a wave modulation inductor L, a pulse capacitor group and an ignition ball gap; the method specifically includes the following steps:

[0007] Step 1: Determine the test circuit component parameters, that is, determine the wave head time and wave tail time of the impulse current waveform;

[0008] Step 2: directly connect the two ends of the test sample to the test circuit;

[0009] Step 3: Check whether the impulse current waveform meets the requirements;

[0010] Step 4: Adjust the shooting parameters of the high-speed camera;

[0011] Step 5: Repeat the test several times to adjust the frame rate of the control software;

[0012] Step 6: Image processing: grayscale processing, grayscale image filtering processing and arc halo elimination processing are performed on the multi-chamber arc image;

[0013] Step 7: Selection of different arc regions;

[0014] Step 8: Calculate the total grayscale of the arc area and extract the arc morphology characteristics.

[0015] Furthermore, the arc image morphology detection method of the multi-chamber arc extinguishing device can be used to study the arc dissipation characteristics and optimize the chamber nozzle layout of the multi-chamber arc extinguishing device.

[0016] Furthermore, the weighted average method is used in the grayscale processing of the image. The specific grayscale value calculation formula of each pixel is as follows:

[0017] g(i, j)=0.2989R+0.5870G+0.1140B

[0018] Where g(i, j) is the grayscale value of the pixel with coordinates (i, j), and R, G, and B represent the three primary color values ​​of each pixel.

[0019] Furthermore, bilateral filtering is used in grayscale image filtering processing; specifically, the neighborhood to be used for filtering is segmented or classified according to the pixel value, and then a relatively high weight is given to the category to which the point belongs, and then the neighborhood weighted sum is performed to obtain the final result.

[0020] Furthermore, the kernel function of bilateral filtering is a Gaussian kernel function related to spatial distance multiplied by a Gaussian function related to grayscale distance;

[0021] Among them, the calculation formula of the Gaussian kernel function related to spatial distance is:

[0022]

[0023] Where (k, l) is the coordinate of the center pixel of the processing window; (i, j) is the coordinate of other pixels in the processing window; σ d is the standard deviation of the Gaussian function;

[0024] The calculation formula of the Gaussian function related to the grayscale distance is:

[0025]

[0026] Where f(i, j) is the grayscale value of the pixel at the coordinate (i, j), f(k, l) is the grayscale value of the pixel at the coordinate (k, l), σ r is the standard deviation of the Gaussian function with respect to grayscale;

[0027] The formula for calculating the product of the two is:

[0028]

[0029] The bilateral filter kernel function calculates the grayscale value of each pixel as follows:

[0030]

[0031] g(i, j) is the grayscale value of the pixel at coordinate (i, j).

[0032] Furthermore, an iterative algorithm is used to obtain the optimal threshold T in the arc halo elimination process to eliminate the halo around the arc; specifically, a threshold is first assumed, and then the foreground and background center values ​​under the threshold are calculated. When the average value of the foreground and background center values ​​and the assumed threshold are equal, if the iterative result converges, the iteration is terminated and binarization is performed using this value.

[0033] Furthermore, the iterative algorithm first obtains an initial threshold, then divides the image into foreground and background according to the initial threshold, and respectively obtains the average grayscale value of the foreground and background; and finally iterates and solves until convergence;

[0034] The formula for calculating the initial threshold is:

[0035]

[0036] Where g l is the maximum gray value of the image, g u is the minimum gray value of the image;

[0037] The formula for calculating the average gray value of the foreground and background is:

[0038]

[0039]

[0040] Where g represents the pixel gray value, h(g) is the weight corresponding to different gray values, and A b is the average gray value of the foreground, A f is the average gray value of the background;

[0041] The formula for iterative solution is:

[0042]

[0043] Where T k Get the grayscale threshold for the Kth iteration;

[0044] The expression formula until convergence is:

[0045] T k+1 -T k <ε

[0046] Where ε is the iterative calculation accuracy.

[0047] Furthermore, in step 7, different arc regions are selected using a two-pass scanning algorithm to scan the image from left to right and from top to bottom, and adjacent pixels with the same pixel values ​​are connected to form a connected domain; the arc regions are extracted and marked, and the number of arc regions and the boundary coordinates of the connected domain are recorded.

[0048] Further, step 8, obtaining the total grayscale of the arc area: according to the arc area determined in step 7, extracting the corresponding arc area in the grayscale image using the recorded coordinate data, calculating the total grayscale value of the arc area, obtaining the grayscale features of the arc at different development moments, and reflecting the moment when the arc is completely ejected from the chamber according to the grayscale change;

[0049] Among them, the formula for calculating the total grayscale amount in the arc area is:

[0050]

[0051] The beneficial effects of the present invention are:

[0052] The arc image morphology detection method for a multi-chamber arc extinguishing device is based on a test circuit composed of an impulse current generator, a capacitive voltage divider, a Rogowski coil, a digital storage oscilloscope, a high-speed camera and a data processor; the arc image morphology detection method for a multi-chamber arc extinguishing device extracts the image morphology characteristics of the arc and the distribution of the cut-off arc, and determines the moment when the arc is completely ejected from the chamber according to the total amount of grayscale of the arc. The arc cut-off moment and the position distribution of the cut-off arc are of great significance to the study of the dissipation characteristics of the arc, and play a guiding role in optimizing the layout of the chamber nozzle. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 It is a flow chart of the arc image morphology detection method of the multi-chamber arc extinguishing device;

[0054] Figure 2 This is the connection diagram of the test circuit;

[0055] Figure 3 It is a logo diagram of the arc area during the arc dissipation process;

[0056] Figure 4 These are pictures of arc shapes at different times after the halo is removed;

[0057] Figure 5 It is the histogram of the total grayscale of the image at each moment during the arc dissipation process;

[0058] Figure 2 Among them, 1- impulse current generator, 2- capacitor voltage divider, 3- digital storage oscilloscope, 4- Rogowski coil, 5- high-speed camera, 6- data processor, 7- test sample. DETAILED DESCRIPTION

[0059] The following will be combined with the accompanying drawings of 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.

[0060] The impact test device and impact arc image detection method of the multi-chamber arc extinguishing chamber are constructed by building an impact current discharge test platform of the self-energy multi-chamber arc extinguishing chamber, using a high-speed camera to observe the movement of the impact arc after it is ejected from the self-energy multi-chamber arc extinguishing chamber to the extinguishing process, improving the morphological detection method of the impact arc image, and determining the moment when the arc is completely ejected from the chamber by extracting the grayscale features of the arc image; the whole process simplifies the test steps, has high imaging efficiency and good effect, obtains parameters such as the airflow velocity in the chamber, and is used to study its change law, effectively improving the experimental effect of the arc dissipation characteristics in the self-energy multi-chamber arc extinguishing chamber.

[0061] The arc image morphology detection method of the multi-chamber arc extinguishing device is based on a test circuit composed of an impulse current generator 1, a capacitive voltage divider 2, a Rogowski coil 4, a digital storage oscilloscope 3, a high-speed camera 5 and a data processor 6. The impulse current generator 1 (ICG) can generate a double exponential current wave with an adjustable amplitude of 8 to 200 kA, a variable wavefront time of 1.2 to 20 μs, and a variable wave tail time of 20 to 1000 μs. The impulse current generator 1 mainly includes: an intelligent control system, a voltage regulator, a step-up transformer, a silicon stack, a wave modulation resistor R, a wave modulation inductor L, a pulse capacitor group, and an ignition ball gap. In the test, the 380V AC power supply is connected to the 380V industrial frequency power supply through the input end of the step-up and voltage regulator through a wire, and the output end of the voltage regulator is connected to the primary side of the step-up transformer through a wire; the secondary side of the step-up transformer is connected to the silicon stack with a wire, and the silicon stack rectifier charges the pulse capacitor group. When the charging voltage reaches the preset value, the ignition ball gap applies an ignition pulse through the pulse ignition device to achieve breakdown discharge, thereby forming an impact current injected into the test sample 7. By changing the wave modulation inductance and wave modulation resistance in the test circuit, the impact current waveform with different wave front time and half peak time can be obtained; by controlling the charging voltage of the capacitor group, the impact current with different peak values ​​can be obtained.

[0062] In order to achieve precise measurement of the impact current and impact current value and provide measurement data for triggering the high-speed camera 5 to shoot, a through-core current sensor is used to form a current measurement system with a digital storage oscilloscope 3. The through-core current sensor is composed of a non-magnetic skeleton, a copper coil, an integration circuit, a signal transmission interface and a polymer insulation shell. The non-magnetic skeleton is a ring with an inner diameter of 2 to 10 cm, an outer diameter of 2.5 to 12 cm, a cross-sectional diameter of 1 to 4 cm, and a material of a non-magnetic polymer. Its function is to fix the copper coil and prevent the sensor core from saturating when measuring a large impact current. The copper coil is made of copper enameled wire with a wire diameter of 0.44 to 1.67 mm, which is evenly wound on the annular non-magnetic skeleton. The number of turns is 50 to 1000 turns. The lead wires at both ends of the copper coil are connected to the input end of the integration circuit to integrate the induced electromotive force, thereby obtaining the magnitude of the current changing with time t. The capacitive voltage divider 2 and the digital storage oscilloscope 3 form a voltage measurement system, with a voltage divider ratio of 1000:1, a voltage signal with an amplitude of -400 to 400 kV and a frequency of 0 to 1 MHz, and the measured signal does not produce attenuation and deformation. The oscilloscope is powered by an independent power supply, and simultaneously collects voltage signals with an amplitude of -400 to 400 V and a frequency of 0 to 100 MHz in two signal acquisition channels, with a sampling frequency of 0 to 10 GS / s and a storage capacity of 0 to 100 MB, which can ensure the accuracy and length of the collected signals. Once the air gap in the multi-chamber arc extinguishing device is broken down, at the same time, the oscilloscope sends a trigger signal to trigger the high-speed camera 5 to start shooting the arc appearance, development and dissipation process to obtain the arc image. The high-speed camera 5 has a 12-bit monochrome chip (36-bit RGB color), a pixel of 20 μm, and a frame rate that can be adjusted according to the pixel, up to 1,000,000 fps, and the shooting parameters are controlled by software.

[0063] The arc image morphology detection method of the multi-chamber arc extinguishing device is as follows: Figure 1 As shown, the specific steps include:

[0064] Step 1: Determine the test circuit component parameters

[0065] First, determine the test circuit component parameters, that is, determine the wave crest time and wave tail time of the impulse current waveform. Determine it according to the wave crest time and wave tail time of the actual lightning current and the simulation ratio n. That is, first divide the wave crest time and wave tail time of the actual lightning current by the simulation ratio n, calculate the wave crest time and wave tail time of the impulse current of the simulation test, and then change the wave modulation resistor R and wave modulation inductor L in the impulse current generator 1 circuit to achieve the calculated wave crest time and wave tail time of the impulse current.

[0066] Step 2: Figure 2 As shown, connect each test device and directly connect the two ends of the test sample 7 to wire the test circuit.

[0067] Use braided copper tape as a connecting wire to ground the negative pole of the pulse capacitor group, and use braided copper tape as a connecting wire to connect the current output end of the impulse current generator 1 to the modulation inductor and modulation resistor. The braided copper tape needs to pass through the Rogowski coil 4 and connect to the test piece 7. Connect the high-voltage end of the test piece 7 to the capacitor voltage divider 2 for voltage measurement. At the same time, short-circuit the two ends of the test piece 7 first to check whether the waveform is correct later. Use a coaxial shielded cable to connect the Rogowski coil 4 and the capacitor voltage divider 2 to the oscilloscope input channel, and connect the oscilloscope trigger signal output end to the high-speed camera 5 to ensure that the high-speed camera 5 takes pictures immediately when the arc appears. During the test line connection process, pay attention to avoid contact between the signal line and the braided copper tape and keep an appropriate distance. Finally, check whether the line is correct and the connection is good. When all the wiring connections are good, proceed to the next step.

[0068] Step 3: Check whether the impulse current waveform meets the requirements

[0069] Under the short-circuit condition, the impulse current generator 1 is started to output the impulse current, and the waveform displayed by the oscilloscope is used to determine whether the current waveform meets the requirements.

[0070] Step 4: Adjust the shooting parameters of high-speed camera 5

[0071] The high-speed camera 5 is a Photron FASTCAM SA5, which has a 12-bit monochrome chip (36-bit RGB color), a pixel of 20 μm, and a frame rate adjustable according to the pixel, up to 1,000,000 fps, and is controlled by Photron's Fastcam Viewer (PFV) software. The frame rate is set to 15,000 fps, the trigger mode is set to center point trigger, and the ISO sensitivity and other required parameters are set by the control software. A focusing lens is not added to the high-speed camera 5, and the focal length is adjusted to make the sample 7 clear in the imaging area.

[0072] Step 5: Repeat the test several times and adjust the frame rate of the control software

[0073] Remove the short-circuit wiring at both ends of the test piece 7, set the charging voltage value of the impulse current generator 1, and set the charging time according to the impulse voltage value to start charging; after charging is completed, trigger the impulse current generator 1 to check whether the synchronous trigger sends a trigger signal. The high-speed camera 5 captures the arc image. According to the captured image, adjust the control software frame rate, etc., and repeat the test multiple times.

[0074] Step 6: Image processing. The processing steps are as follows:

[0075] Image grayscale: The arc image obtained by the high-speed camera 5 is an RGB image. The RGB image cannot reflect the morphological characteristics of the image. It only adjusts the color based on the optical principle. What actually reflects the image characteristics is the brightness of the image. At the same time, in order to reduce the matrix storage capacity in the computer calculation process and improve the image processing speed, the weighted average method is used to grayscale the original image.

[0076] The gray value calculation formula for each pixel is as follows:

[0077] g(i, j)=0.2989R+0.5870G+0.1140B

[0078] Where g(i, j) is the grayscale value of the pixel with coordinates (i, j), and R, G, and B represent the three primary color values ​​of each pixel.

[0079] Grayscale image filtering: Bilateral filtering can preserve the edge features of the arc area. The principle is to first segment or classify the neighborhood to be used for filtering according to the pixel value, then give the category to which the point belongs a relatively high weight, and then perform weighted summation of the neighborhood to obtain the final result. The kernel function of bilateral filtering is a Gaussian kernel function related to spatial distance multiplied by a Gaussian function related to grayscale distance. Bilateral filtering adds weights to grayscale information, that is, within the neighborhood, the closer the grayscale value is to the grayscale value of the center point, the greater the weight, and the smaller the weight of the point with a large difference in grayscale value. The size of the weight is determined by the range Gaussian function. The weight coefficients related to spatial distance and grayscale value are multiplied to obtain the final kernel function.

[0080] The calculation formula of the Gaussian kernel function related to spatial distance is:

[0081]

[0082] Where (k, l) is the coordinate of the center pixel of the processing window; (i, j) is the coordinate of other pixels in the processing window; σ d is the standard deviation of the Gaussian function.

[0083] The calculation formula of the Gaussian function related to the grayscale distance is:

[0084]

[0085] Where f(i, j) is the grayscale value of the pixel at the coordinate (i, j), f(k, l) is the grayscale value of the pixel at the coordinate (k, l), σ r is the standard deviation of the Gaussian function with respect to grayscale.

[0086] The formula for calculating the product of the two is:

[0087]

[0088] The bilateral filter kernel function calculates the grayscale value of each pixel as follows:

[0089]

[0090] Where g(i, j) represents the pixel grayscale value after calculation of the bilateral filter kernel function.

[0091] The arc generation and dissipation process after the original image is grayed and bilaterally filtered Figure 3 As shown, it can be seen from the picture that the edge of the arc is clear and obvious, and the image distortion is not obvious.

[0092] Eliminate arc halo: The arc is accompanied by strong gas ionization during its formation and dissipation. Gas ionization will generate a large number of photons that escape around the arc. In the captured arc image, a halo can be seen around the arc. This paper uses an iterative algorithm to find the optimal threshold T to eliminate the halo around the arc. First, assume a threshold, then calculate the foreground and background center values ​​under the threshold. When the average value of the foreground and background center values ​​and the assumed threshold, if the iterative result converges, the iteration is terminated and binarization is performed with this value.

[0093] Find the maximum and minimum grayscale values ​​of the image, denoted as g l and g u , the initial threshold is:

[0094]

[0095] Where g l is the maximum gray value of the image, g u is the minimum grayscale value of the image.

[0096] According to the initial threshold, the image is divided into foreground and background, and the average gray value of the two is calculated respectively:

[0097]

[0098]

[0099] Where g represents the pixel grayscale value, and h(g) is the weight corresponding to different grayscale values. b is the average gray value of the foreground, A f is the average gray value of the background.

[0100] Iterative solution Until convergence, T k+1 -T k <ε.

[0101] Where T k The grayscale threshold is obtained for the Kth iteration, and ε is the calculation accuracy of the iteration.

[0102] Step 7: Selection of different arc areas

[0103] Selection of different arc regions: From the figure, it can be seen that in the early stage of arc quenching, the arcs ejected from multiple nozzles of the multi-chamber arc extinguishing device are connected to each other, presenting a gradually expanding plasma group. When the arc develops and begins to dissipate, the arc is cut into several small arc regions. Each arc region corresponds to a connected domain in the image. After eliminating the aperture in the previous step, the region is composed of foreground pixels with the same pixel value and adjacent positions. A two-pass scanning algorithm is used to scan the image from left to right and from top to bottom, and pixels with the same pixel value and adjacent to each other are connected to form a connected domain. The arc regions are extracted and marked, and the number of arc regions and the boundary coordinates of the connected domain are recorded. Figure 3 and Figure 4 It can be found that as the arc dissipates, the arc ejected by the multi-chamber arc extinguishing device does not dissipate uniformly as a whole. The arc in the middle dissipates faster, while the arc at the upper and lower ends is not easy to dissipate. Therefore, the arc cutoff moment and the position distribution of the cutoff arc are of great significance to the study of the arc dissipation characteristics, and play a guiding role in optimizing the layout of the chamber nozzle.

[0104] Step 8: Calculate the total grayscale value in the arc area

[0105] Obtain the total amount of grayscale in the arc area: The grayscale value of the original image after grayscale processing reflects the brightness of the arc. The process from the generation to the dissipation of the arc is accompanied by the ionization of the gas to produce photons. As the charged particles inside the arc continue to recombine and diffuse, the deionization effect is enhanced, and the brightness gradually weakens with the development of the arc. Therefore, the moment when the brightness of the arc area reaches the maximum is also the moment when the arc is completely ejected from the chamber. According to the boundary coordinates of the multiple truncated arcs recorded in the previous step, the arc area is determined in the grayscale image, that is, the calculation interval is determined. The pixel points in the area are traversed in turn and their grayscale values ​​are accumulated. According to the trend of the arc grayscale total amount increasing rapidly and then gradually decreasing, the moment when the arc is completely ejected from the nozzle and begins to extinguish is determined.

[0106] Among them, the formula for calculating the total grayscale amount in the arc area is:

[0107]

[0108] Depend on Figure 5 From the grayscale total amount histogram, it can be seen that at 160μs, the total grayscale amount of the image reaches the maximum, indicating that the arc brightness is the highest at this moment and the arc reaction releases a large number of photons. At time , the total grayscale amount of the arc image begins to decrease, indicating that the arc has completely ejected from the chamber, causing the arc to be cut off and begin to dissipate with the release of energy.

Claims

1. A method for detecting arc image morphology of a multi-chamber arc extinguishing device, characterized in that: The arc image morphology detection method of the multi-chamber arc extinguishing device is based on a test circuit composed of an impulse current generator, a capacitive voltage divider, a Rogowski coil, a digital storage oscilloscope, a high-speed camera and a data processor, wherein the impulse current generator includes an intelligent control system, a voltage regulator, a step-up transformer, a silicon stack, a wave modulation resistor R, a wave modulation inductor L, a pulse capacitor group, and an ignition ball gap; specifically, the following steps are included: Step 1: Determine the test circuit component parameters, that is, determine the wave head time and wave tail time of the impulse current waveform; Step 2: directly connect the two ends of the test sample to the test circuit; Step 3: Check whether the impulse current waveform meets the requirements; Step 4: Adjust the shooting parameters of the high-speed camera; Step 5: Repeat the test several times to adjust the frame rate of the control software; Step 6: Image processing: grayscale processing, grayscale image filtering processing and arc halo elimination processing are performed on the multi-chamber arc image; The weighted average method is used in the image grayscale processing. The specific grayscale value calculation formula of each pixel is as follows: g(i,j)=0.2989R+0.5870G+0.1140B Where g(i,j) is the gray value of the pixel with coordinates (i,j), and R, G, and B represent the three primary color values ​​of each pixel; Bilateral filtering is used in grayscale image filtering. Specifically, the neighborhood to be used for filtering is segmented or classified according to the pixel value, and then a relatively high weight is given to the category to which the pixel point to be filtered belongs, and then the neighborhood weighted sum is performed to obtain the final result. The kernel function of bilateral filtering is a Gaussian kernel function related to spatial distance multiplied by a Gaussian function related to grayscale distance; Among them, the calculation formula of the Gaussian kernel function related to spatial distance is: Where (k, l) is the coordinate of the center pixel of the processing window; (i, j) is the coordinate of other pixels in the processing window; σ d is the standard deviation of the Gaussian function; The calculation formula of the Gaussian function related to the grayscale distance is: Where f(i,j) is the grayscale value of the pixel at the coordinate (i,j), f(k,l) is the grayscale value of the pixel at the coordinate (k,l), σ r is the standard deviation of the Gaussian function with respect to grayscale; The formula for calculating the product of the two is: The bilateral filter kernel function calculates the grayscale value of each pixel as follows: g(i,j) is the gray value of the pixel at coordinate (i,j); In the process of eliminating arc halo, an iterative algorithm is used to find the optimal threshold T to eliminate the halo around the arc; specifically, a threshold is assumed first, and then the foreground and background center values ​​under the threshold are calculated. When the average value of the foreground and background center values ​​and the assumed threshold are equal, if the iterative result converges, the iteration is terminated and binarization is performed with this value; The iterative algorithm first obtains an initial threshold, then divides the image into foreground and background according to the initial threshold, and respectively obtains the average grayscale value of the foreground and background; and finally iterates and solves until convergence; The formula for calculating the initial threshold is: Where g l is the maximum gray value of the image, g u is the minimum gray value of the image; The formula for calculating the average gray value of the foreground and background is: Where g represents the pixel gray value, h(g) is the weight corresponding to different gray values, and A b is the average gray value of the foreground, A f is the average gray value of the background; The formula for iterative solution is: Where T k Get the grayscale threshold for the Kth iteration; The expression formula until convergence is: T k+1 -T k <ε Where ε is the iterative calculation accuracy; Step 7: Selection of different arc regions; Step 8, obtaining the total grayscale of the arc area and extracting the arc morphology characteristics; Obtain the total grayscale of the arc area. According to the arc area determined in step 7, extract the corresponding arc area in the grayscale image using the recorded coordinate data, calculate the total grayscale value of the arc area, and obtain the grayscale characteristics of the arc at different development moments. The grayscale value after the image is grayed reflects the brightness of the arc. When the brightness of the arc area reaches the maximum, it reflects the moment when the arc is completely ejected from the chamber. Among them, the formula for calculating the total grayscale amount in the arc area is: G=∑ i,j∈S g(i,j)。 2. The arc image morphology detection method of a multi-chamber arc extinguishing device according to claim 1 is characterized in that: The arc image morphology detection method of the multi-chamber arc extinguishing device can be used to study the arc dissipation characteristics and optimize the chamber nozzle layout of the multi-chamber arc extinguishing device.

3. The arc image morphology detection method of a multi-chamber arc extinguishing device according to claim 1, characterized in that: Step 7: Different arc regions are selected using a two-pass scanning algorithm. The image is scanned from left to right and from top to bottom, and adjacent pixels with the same pixel values ​​are connected to form a connected domain. The arc regions are extracted and marked, and the number of arc regions and the boundary coordinates of the connected domain are recorded.

Images