High frequency spark machine calibration method, device and computer storage medium

Abstract

The present invention provides a high-frequency spark machine calibration method, device, equipment and computer storage medium, which relates to the technical field of spark machine calibration, including obtaining first information and second information, wherein the first information is the operation video information of the high-frequency spark machine, and the second information is the operation data information corresponding to the operation video information, sending the first information and the second information to an abnormality detection module for processing to obtain abnormal information of the high-frequency spark machine; sending the abnormal information to an adjusted calibration data prediction model for processing to obtain calibration data of the high-frequency spark machine; calibrating the high-frequency spark machine based on the calibration data of the high-frequency spark machine to obtain a calibrated high-frequency spark machine. The present invention can monitor the operation of the high-frequency spark machine at all times, and then calibrate the abnormal data of the high-frequency spark machine, reduce the manpower and material resources required for calibration, and ensure the operation of the high-frequency spark machine.

Description

Technical Field

[0001] The invention relates to the technical field of spark machine calibration, and in particular to a high-frequency spark machine calibration method, device, equipment and computer storage medium. Background Art

[0002] The existing high-frequency spark machines often have certain basic errors during use, and will produce certain wear and tear during use. Therefore, the high-frequency spark machines need to be calibrated every year. However, since the calibration errors require professional personnel to operate, a lot of time and manpower costs are required. In addition, the calibration time and calibration method further limit the calibration frequency, and calibration cannot be performed at any time, which in turn has a certain impact on the use effect of the high-frequency spark machine. Therefore, a method and device that can quickly calibrate the high-frequency spark machine are needed. Summary of the invention

[0003] The purpose of the present invention is to provide a high-frequency spark machine calibration method, device, equipment and computer storage medium to improve the above problems. In order to achieve the above purpose, the technical solution adopted by the present invention is as follows:

[0004] In a first aspect, the present application provides a high-frequency spark machine calibration method, comprising:

[0005] Acquire first information and second information, wherein the first information is operation video information of the high-frequency spark machine, and the second information is operation data information corresponding to the operation video information, wherein the operation data includes voltage value, sensitivity and stability of the high-frequency spark machine when it is running;

[0006] Sending the first information and the second information to an abnormality detection module for processing to obtain third information, wherein the third information includes abnormality information of the high-frequency spark machine;

[0007] Sending the third information to the adjusted calibration data prediction model for processing to obtain calibration data of the high-frequency spark machine;

[0008] The high-frequency spark machine is calibrated based on the calibration data of the high-frequency spark machine to obtain a calibrated high-frequency spark machine.

[0009] In a second aspect, the present application also provides a high-frequency spark machine calibration device, comprising:

[0010] An acquisition unit, used to acquire first information and second information, wherein the first information is operation video information of the high-frequency spark machine, and the second information is operation data information corresponding to the operation video information, wherein the operation data includes voltage value, sensitivity and stability of the high-frequency spark machine when it is running;

[0011] A first processing unit, used for sending the first information and the second information to an abnormality detection module for processing to obtain third information, wherein the third information includes abnormality information of the high-frequency spark machine;

[0012] A second processing unit is used to send the third information to the adjusted calibration data prediction model for processing to obtain calibration data of the high-frequency spark machine;

[0013] The third processing unit is used to calibrate the high-frequency spark machine based on the calibration data of the high-frequency spark machine to obtain a calibrated high-frequency spark machine.

[0014] In a third aspect, the present application further provides a computer storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of the above-mentioned high-frequency spark machine calibration method are implemented.

[0015] The beneficial effects of the present invention are:

[0016] The present invention determines the abnormal data by judging the operation video information of the high-frequency spark machine and the corresponding parameter information during the operation video, and then calculates the calibration data for the high-frequency spark machine that generates the abnormal data, and then calibrates the high-frequency spark machine that generates the abnormality based on the calculated calibration data. In this way, the operation status of the high-frequency spark machine can be monitored at all times, and then the abnormal data of the high-frequency spark machine can be calibrated, thereby reducing the manpower and material resources required for the calibration, and can ensure the operation status of the high-frequency spark machine.

[0017] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or be understood by implementing the embodiments of the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description, claims, and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0019] Figure 1 It is a schematic flow chart of the high-frequency spark machine calibration method described in an embodiment of the present invention;

[0020] Figure 2 It is a schematic structural diagram of the high-frequency spark machine calibration device described in an embodiment of the present invention.

[0021] Markings in the figure: 701, acquisition unit; 702, first processing unit; 703, second processing unit; 704, third processing unit; 7021, first processing subunit; 7022, first comparison subunit; 7023, first judgment subunit; 7024, second processing subunit; 70211, second comparison subunit; 70212, third processing subunit; 70213, first calculation subunit; 70214, second judgment subunit; 70231, acquisition subunit; 70232, fourth processing subunit; 70233, third judgment subunit; 70234, third comparison subunit; 7031, fifth processing subunit; 7032, second calculation subunit; 7033, sixth processing subunit. DETAILED DESCRIPTION

[0022] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. The components of the embodiments of the present invention generally described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed invention, but merely represents selected embodiments of the present invention. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0023] It should be noted that similar reference numerals and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in the subsequent drawings. At the same time, in the description of the present invention, the terms "first", "second", etc. are only used to distinguish the description and cannot be understood as indicating or implying relative importance.

[0024] Embodiment 1:

[0025] like Figure 1 As shown, this embodiment provides a high-frequency spark machine calibration method, which includes step S1, step S2, step S3 and step S4.

[0026] Step S1, obtaining first information and second information, wherein the first information is the operation video information of the high-frequency spark machine, and the second information is the operation data information corresponding to the operation video information, and the operation data includes the voltage value, sensitivity and stability of the high-frequency spark machine when it is running;

[0027] It can be understood that in this step, the camera equipment captures the operation video of the high-frequency spark machine, including the discharge segment of the high-frequency spark machine, and then judges whether the high-frequency spark machine can operate normally by judging the discharge image of the high-frequency spark machine, and obtains the operation parameters corresponding to the discharge image, and then judges whether the abnormal data is caused by the parameters.

[0028] Step S2, sending the first information and the second information to an abnormality detection module for processing to obtain third information, wherein the third information includes abnormality information of the high-frequency spark machine;

[0029] It can be understood that the above steps determine the abnormal image of the discharge image of the high-frequency spark machine through image recognition, and then judge whether the discharge angle and discharge frequency of the high-frequency spark machine are correct, and then judge whether the high-frequency spark machine is abnormal. In addition, the above steps also judge whether the operating parameters of the high-frequency spark machine are abnormal through the voltage value, sensitivity and stability of the high-frequency spark machine during operation, and then obtain the abnormal information of the high-frequency spark machine.

[0030] It can be understood that the voltage value of the high-frequency spark machine during operation is measured by a voltmeter and uploaded and stored. The sensitivity refers to the detection by placing a cable with a preset breakdown number on the high-frequency spark machine, and then judging whether the breakdown number is consistent with the preset breakdown number, and then determining the sensitivity. The stability refers to judging whether the high-frequency spark machine can discharge normally when detecting a cable without defects. If it can discharge normally, its stability is 1, and if it cannot discharge normally, its stability is 0.

[0031] Step S3, sending the third information to the adjusted calibration data prediction model for processing to obtain calibration data of the high-frequency spark machine;

[0032] It can be understood that the present invention performs regression analysis through the XGBoost algorithm to determine the calibration data of the high-frequency spark machine, and then can quickly calibrate the discharge angle, discharge frequency, voltage value, sensitivity and stability of the high-frequency spark machine to ensure the normal operation of the high-frequency spark machine.

[0033] Step S4: calibrating the high-frequency spark machine based on the calibration data of the high-frequency spark machine to obtain a calibrated high-frequency spark machine.

[0034] It can be understood that the above steps determine the abnormal data by judging the operating video information of the high-frequency spark machine and the corresponding parameter information during the operating video, and then calculate the calibration data for the high-frequency spark machine that generates the abnormal data, and then calibrate the high-frequency spark machine that generates the abnormality based on the calculated calibration data. In this way, the operating status of the high-frequency spark machine can be monitored at all times, and then the abnormal data of the high-frequency spark machine can be calibrated, thereby reducing the manpower and material resources required for calibration and ensuring the operating status of the high-frequency spark machine.

[0035] In a specific implementation of the present disclosure, step S2 includes step S21, step S22, step S23 and step S24.

[0036] Step S21, extracting the operation video information of the high-frequency spark machine according to the operation time of the high-frequency spark machine to obtain a key image of the operation video of the high-frequency spark machine, wherein the key image is at least one frame of discharge image of the high-frequency spark machine obtained based on image recognition processing;

[0037] It can be understood that this step is to identify the discharge segments of the high-frequency spark machine through image recognition, and then obtain the discharge image of each frame of the high-frequency spark machine, thereby reducing the need to judge the image information in the future, thereby reducing the amount of calculation.

[0038] Step S22, comparing the key image of the operation video of the high-frequency spark machine with the preset discharge image of the spark machine to obtain abnormal discharge image information of the high-frequency spark machine;

[0039] It can be understood that this step determines the discharge angle and discharge frequency by comparing the key image with the preset discharge image, wherein the discharge frequency is determined by the interval time of the discharge image, thereby quickly judging whether the discharge angle of the high-frequency spark machine is correct and whether it is necessary to adjust the angle and frequency of the discharge head of the spark machine.

[0040] Step S23, sending the second information to the trained abnormal data judgment model for abnormal judgment, to obtain abnormal operation data of the high-frequency spark machine;

[0041] Step S24, obtaining abnormal information of the high-frequency spark machine based on the abnormal discharge image information of the high-frequency spark machine and the abnormal discharge image information of the high-frequency spark machine.

[0042] It can be understood that the above steps determine the abnormal information of the high-frequency spark machine, including abnormal operation data and abnormal discharge images, by combining the abnormal discharge image information of the high-frequency spark machine and the abnormal discharge image of the high-frequency spark machine, and then calibrate the data for all the data to prevent the occurrence of inaccurate calibration data due to separate calibration.

[0043] In a specific implementation of the present disclosure, step S21 includes step S211, step S212, step S213 and step S214.

[0044] Step S211, input each frame of the key image of the operation video of the high-frequency spark machine into the image matching degree calculation model, and compare them with the preset discharge images of the high-frequency spark machine respectively, to obtain at least two matching degree values ​​of the key image of the operation video of the high-frequency spark machine and the preset discharge image of the spark machine;

[0045] It can be understood that the steps are performed by comparing the key images of the operation videos of all high-frequency spark machines with the preset discharge images of the high-frequency spark machines, wherein by comparing the pixel points of the discharge head and connecting the pixel points into lines, it is determined whether the angles of the lines are consistent, and the present invention also obtains the discharge frequency by comparing the discharge time corresponding to the discharge image, and performs dimensionless quantization processing on the discharge frequency and the discharge angle, and then performs correlation analysis based on the discharge frequency and the discharge angle to determine the correlation between the key images of the operation videos of the high-frequency spark machine and the preset discharge images of the high-frequency spark machine, and then uses the correlation value as the matching value.

[0046] Step S212: Summarize the matching values ​​of the key images of the operation videos of all high-frequency spark machines and the preset discharge images of the spark machines to obtain a matching value set;

[0047] Step S213, performing root mean square calculation on all data in the matching value set, and using the obtained root mean square value as a threshold for determining whether the key image of the operation video of the high-frequency spark machine is an abnormal discharge image;

[0048] Step S214: based on the threshold for determining whether the key image of the operation video of the high-frequency spark machine is an abnormal discharge image, obtain abnormal discharge image information of the high-frequency spark machine.

[0049] It can be understood that the above steps determine the matching values ​​of the key images of the operating videos of all high-frequency spark machines and the preset discharge images of the spark machines, then perform root mean square calculation on all the matching values, and use the key images corresponding to the matching values ​​less than the root mean square value as the abnormal discharge image information of the high-frequency spark machine. This can increase the accuracy of judging abnormal data in the present invention, so as to quickly and accurately judge the high-frequency spark machines that need to be calibrated.

[0050] In a specific implementation of the present disclosure, step S23 includes step S231, step S232, step S233 and step S234.

[0051] Step S231, acquiring historical operation data information, screening abnormal data of the historical operation data information, and calibrating the abnormal data of the historical operation data information into abnormal types to obtain calibrated abnormal data information;

[0052] Step S232: Process the abnormal data information of the historical operation data information based on the CART algorithm to obtain a CART decision tree, and perform random pruning on the CART decision tree to obtain a decision tree model for judging the abnormal data information;

[0053] Step S233: sending the historical operation data information to the decision tree model for judgment to obtain abnormal data information of the historical operation data information;

[0054] Step S234, comparing the abnormal data information based on the historical operation data information with the calibrated abnormal data information, and adjusting the judgment parameters in the decision tree model based on the comparison result, until the comparison result is the same as the preset comparison result, to obtain a trained decision tree model.

[0055] It can be understood that the present invention performs abnormal calibration on historical data, then processes it through the CART algorithm to establish a decision tree model, and optimizes the parameters of the decision tree model to increase the judgment accuracy of the decision tree model, reduce labor costs and improve judgment efficiency.

[0056] In a specific implementation of the present disclosure, step S3 includes step S31, step S32 and step S33.

[0057] Step S31, performing regression analysis on the preset historical abnormal data of the high-frequency spark machine based on the XGBoost algorithm to obtain calibration data of the historical abnormal data of the high-frequency spark machine;

[0058] It can be understood that this step selects the CART tree as the regression tree of the model, and then performs ensemble learning based on the XGBoost algorithm, where the objective function of the XGBoost algorithm is as follows. The objective function is parameterized and the tree structure is introduced into the objective function, and then trees are continuously added for optimization. Each additional tree adds a new function.

[0059] The formula of the objective function is as follows:

[0060]

[0061] Where obj represents the objective function of the above XGBoost algorithm, L(y i , y) represents the loss function, y irepresents the predicted value of the calibration data, y represents the input high-frequency spark machine historical abnormal data, and the square loss function is selected, that is, L(y i , y)=(y i -y) 2 , k represents a total of k trees, f k represents the function model of the kth tree, where Ω(f k ) is positive, and

[0062]

[0063] Among them, γ and λ are constant coefficients, T represents the number of leaf nodes in each tree, and ω is the set of scores of leaf nodes in each tree.

[0064] The expression of the optimization model is as follows:

[0065]

[0066] Among them, f t (x) represents the function model of the t-th tree, is the optimized function model of the t-th tree, is the optimized function model of the t-1th tree.

[0067] Step S32, calculating the calibration data of the historical abnormal data of the high-frequency spark machine based on a preset evaluation formula to obtain an evaluation value of the calibration data of the historical abnormal data of the high-frequency spark machine;

[0068] It can be understood that this step evaluates the calibration data through the mean absolute error model, wherein the predicted value of the calibration data and the mean absolute error of the historical abnormal data are calculated by the mean absolute error calculation formula, wherein the smaller the MAE value, the more accurate the calibration data.

[0069] Step S33: adjusting the constant coefficient in the XGBoost algorithm based on the evaluation value until the evaluation value is less than a preset evaluation threshold, thereby obtaining an adjusted calibration data prediction model.

[0070] It can be understood that the present invention repeatedly iteratively adjusts the constant coefficients in the XGBoost algorithm until the predicted calibration data is calibration data that meets the requirements, that is, the average absolute error between the predicted calibration data and the historical abnormal data is less than a threshold, and then the iteration stops to obtain a prediction model for the calibration data with adjusted parameters.

[0071] Embodiment 2:

[0072] like Figure 2As shown, this embodiment provides a device for designing the position of a lamp bead, and the device includes an acquisition unit 701, a first processing unit 702, a second processing unit 703 and a third processing unit 704.

[0073] An acquisition unit 701 is used to acquire first information and second information, wherein the first information is operation video information of the high-frequency spark machine, and the second information is operation data information corresponding to the operation video information, wherein the operation data includes voltage value, sensitivity and stability of the high-frequency spark machine when it is running;

[0074] A first processing unit 702 is used to send the first information and the second information to an abnormality detection module for processing to obtain third information, wherein the third information includes abnormality information of the high-frequency spark machine;

[0075] The second processing unit 703 is used to send the third information to the adjusted calibration data prediction model for processing to obtain calibration data of the high-frequency spark machine;

[0076] The third processing unit 704 is used to calibrate the high-frequency spark machine based on the calibration data of the high-frequency spark machine to obtain a calibrated high-frequency spark machine.

[0077] In a specific implementation of the present disclosure, the first processing unit 702 includes a first processing subunit 7021 , a first comparing subunit 7022 , a first judging subunit 7023 , and a second processing subunit 7024 .

[0078] The first processing subunit 7021 is used to extract the operation video information of the high-frequency spark machine according to the operation time of the high-frequency spark machine to obtain a key image of the operation video of the high-frequency spark machine, wherein the key image is at least one frame of discharge image of the high-frequency spark machine obtained based on image recognition processing;

[0079] The first comparison subunit 7022 is used to compare the key image of the operation video of the high-frequency spark machine with the preset discharge image of the spark machine to obtain abnormal discharge image information of the high-frequency spark machine;

[0080] The first judgment subunit 7023 is used to send the second information to the trained abnormal data judgment model for abnormal judgment to obtain abnormal operation data of the high-frequency spark machine;

[0081] The second processing subunit 7024 is used to obtain abnormal information of the high-frequency spark machine based on the abnormal discharge image information of the high-frequency spark machine and the abnormal discharge image information of the high-frequency spark machine.

[0082] In a specific implementation of the present disclosure, the first processing subunit 7021 includes a second comparing subunit 70211 , a third processing subunit 70212 , a first calculating subunit 70213 and a second judging subunit 70214 .

[0083] The second comparison subunit 70211 is used to input each frame of the key image of the operation video of the high-frequency spark machine into the image matching degree calculation model, and compare and process each frame with the preset discharge image of the high-frequency spark machine to obtain at least two matching values ​​of the key image of the operation video of the high-frequency spark machine and the preset discharge image of the spark machine;

[0084] The third processing subunit 70212 is used to aggregate the matching values ​​of the key images of the operation videos of all high-frequency spark machines and the preset discharge images of the spark machines to obtain a matching value set;

[0085] The first calculation subunit 70213 is used to perform root mean square calculation on all data in the matching value set, and use the obtained root mean square value as a threshold for determining whether a key image of the operation video of the high-frequency spark machine is an abnormal discharge image;

[0086] The second judgment subunit 70214 is used to obtain abnormal discharge image information of the high-frequency spark machine based on the threshold for judging whether the key image of the operation video of the high-frequency spark machine is an abnormal discharge image.

[0087] In a specific implementation of the present disclosure, the first judgment subunit 7023 includes an acquisition subunit 70231 , a fourth processing subunit 70232 , a third judgment subunit 70233 and a third comparison subunit 70234 .

[0088] The acquisition subunit 70231 is used to acquire historical operation data information, filter abnormal data of the historical operation data information, and perform abnormal type calibration on the abnormal data of the historical operation data information to obtain the calibrated abnormal data information;

[0089] The fourth processing subunit 70232 is used to process the abnormal data information of the historical operation data information based on the CART algorithm to obtain a CART decision tree, and perform random pruning on the CART decision tree to obtain a decision tree model for judging the abnormal data information;

[0090] The third judgment subunit 70233 is used to send the historical operation data information to the decision tree model for judgment to obtain abnormal data information of the historical operation data information;

[0091] The third comparison subunit 70234 is used to compare the abnormal data information based on the historical operation data information with the calibrated abnormal data information, and adjust the judgment parameters in the decision tree model based on the comparison result until the comparison result is the same as the preset comparison result, thereby obtaining a trained decision tree model.

[0092] In a specific implementation of the present disclosure, the second processing unit 703 includes a fifth processing subunit 7031 , a second calculation subunit 7032 and a sixth processing subunit 7033 .

[0093] The fifth processing subunit 7031 is used to perform regression analysis on the preset historical abnormal data of the high-frequency spark machine based on the XGBoost algorithm to obtain calibration data of the historical abnormal data of the high-frequency spark machine;

[0094] The second calculation subunit 7032 is used to calculate the calibration data of the historical abnormal data of the high-frequency spark machine based on a preset evaluation formula to obtain an evaluation value of the calibration data of the historical abnormal data of the high-frequency spark machine;

[0095] The sixth processing subunit 7033 is used to adjust the constant coefficient in the XGBoost algorithm based on the evaluation value until the evaluation value is less than a preset evaluation threshold, thereby obtaining an adjusted calibration data prediction model.

[0096] It should be noted that, regarding the device in the above embodiment, the specific manner in which each module performs the operation has been described in detail in the embodiment of the method, and will not be elaborated here.

[0097] Embodiment 3:

[0098] Corresponding to the above method embodiment, a computer storage medium is also provided in this embodiment. The computer storage medium described below and the high-frequency spark machine calibration method described above can refer to each other.

[0099] A computer storage medium stores a computer program, which, when executed by a processor, implements the steps of the high-frequency spark machine calibration method of the above method embodiment.

[0100] The computer storage medium may specifically be a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, or other computer storage medium that can store program codes.

[0101] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

[0102] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A high frequency spark machine calibration method, characterized in that: include: Acquire first information and second information, wherein the first information is operation video information of the high-frequency spark machine, including a discharge segment of the high-frequency spark machine; The second information is the operation data information corresponding to the operation video information, and the operation data includes the voltage value, sensitivity and stability of the high-frequency spark machine when it is running; Sending the first information and the second information to an abnormality detection module for processing to obtain third information, wherein the third information includes abnormality information of the high-frequency spark machine; Sending the third information to the adjusted calibration data prediction model for processing to obtain calibration data of the high-frequency spark machine; The high-frequency spark machine is calibrated based on the calibration data of the high-frequency spark machine to obtain a calibrated high-frequency spark machine.

2. The high-frequency spark machine calibration method according to claim 1, characterized in that , sending the first information and the second information to the abnormality detection module for processing to obtain the third information, including: Extracting the operation video information of the high-frequency spark machine according to the operation time of the high-frequency spark machine to obtain a key image of the operation video of the high-frequency spark machine, wherein the key image is at least one frame of discharge image of the high-frequency spark machine obtained based on image recognition processing; Comparing the key image of the operation video of the high-frequency spark machine with the preset discharge image of the spark machine to obtain abnormal discharge image information of the high-frequency spark machine; Sending the second information to the trained abnormal data judgment model for abnormal judgment to obtain abnormal operation data of the high-frequency spark machine; Based on the abnormal discharge image information of the high-frequency spark machine and the abnormal operation data of the high-frequency spark machine, the abnormal information of the high-frequency spark machine is obtained.

3. The high-frequency spark machine calibration method according to claim 2, characterized in that , compare the key image of the operation video of the high-frequency spark machine with the preset discharge image of the spark machine to obtain abnormal discharge image information of the high-frequency spark machine, including: Input each frame of the key image of the operation video of the high-frequency spark machine into the image matching degree calculation model, and compare them with the preset discharge image of the spark machine respectively, to obtain at least two matching degree values ​​of the key image of the operation video of the high-frequency spark machine and the preset discharge image of the spark machine; Summarize the matching values ​​of the key images of the operation videos of all high-frequency spark machines and the preset discharge images of the spark machines to obtain a matching value set; Calculate the root mean square of all data in the matching value set, and use the obtained root mean square value as a threshold for determining whether a key image of the operation video of the high-frequency spark machine is an abnormal discharge image; Based on the threshold for judging whether the key image of the operation video of the high-frequency spark machine is an abnormal discharge image, the abnormal discharge image information of the high-frequency spark machine is obtained.

4. The high-frequency spark machine calibration method according to claim 2, characterized in that ,The method for constructing the trained abnormal data judgment model includes: Acquire historical operation data information, filter abnormal data of the historical operation data information, and calibrate the abnormal data of the historical operation data information to obtain abnormal type information after calibration; Processing the abnormal data information of the historical operation data information based on the CART algorithm to obtain a CART decision tree, and performing random pruning on the CART decision tree to obtain a decision tree model for judging the abnormal data information; Sending the historical operation data information to the decision tree model for judgment to obtain abnormal data information of the historical operation data information; The abnormal data information based on the historical operation data information is compared with the calibrated abnormal data information, and the judgment parameters in the decision tree model are adjusted based on the comparison result until the comparison result is the same as the preset comparison result, thereby obtaining a trained decision tree model.

5. The high-frequency spark machine calibration method according to claim 1, characterized in that ,The method for constructing the adjusted calibration data prediction model includes: Perform regression analysis on the preset historical abnormal data of the spark machine based on the XGBoost algorithm to obtain calibration data of the historical abnormal data of the high-frequency spark machine; Calculating the calibration data of the historical abnormal data of the high-frequency spark machine based on a preset evaluation formula to obtain an evaluation value of the calibration data of the historical abnormal data of the high-frequency spark machine; The constant coefficient in the XGBoost algorithm is adjusted based on the evaluation value until the evaluation value is less than a preset evaluation threshold, thereby obtaining an adjusted calibration data prediction model.

6. A high-frequency spark machine calibration device, characterized in that: include: An acquisition unit, used for acquiring first information and second information, wherein the first information is operation video information of the high-frequency spark machine, including a discharge segment of the high-frequency spark machine; The second information is the operation data information corresponding to the operation video information, and the operation data includes the voltage value, sensitivity and stability of the high-frequency spark machine when it is running; A first processing unit, used for sending the first information and the second information to an abnormality detection module for processing to obtain third information, wherein the third information includes abnormality information of the high-frequency spark machine; A second processing unit is used to send the third information to the adjusted calibration data prediction model for processing to obtain calibration data of the high-frequency spark machine; The third processing unit is used to calibrate the high-frequency spark machine based on the calibration data of the high-frequency spark machine to obtain a calibrated high-frequency spark machine.

7. The high-frequency spark machine calibration device according to claim 6, characterized in that: The first processing unit comprises: A first processing subunit is used to extract the operation video information of the high-frequency spark machine according to the operation time of the high-frequency spark machine to obtain a key image of the operation video of the high-frequency spark machine, wherein the key image is at least one frame of discharge image of the high-frequency spark machine obtained based on image recognition processing; A first comparison subunit is used to compare the key image of the operation video of the high-frequency spark machine with a preset discharge image of the spark machine to obtain abnormal discharge image information of the high-frequency spark machine; A first judgment subunit is used to send the second information to a trained abnormal data judgment model for abnormal judgment to obtain abnormal operation data of the high-frequency spark machine; The second processing subunit is used to obtain abnormal information of the high-frequency spark machine based on the abnormal discharge image information of the high-frequency spark machine and the abnormal operation data of the high-frequency spark machine.

8. The high-frequency spark machine calibration device according to claim 7, characterized in that: The first processing subunit includes: The second comparison subunit is used to input each frame of the key image of the operation video of the high-frequency spark machine into the image matching degree calculation model, and compare and process each frame with the preset discharge image of the spark machine to obtain at least two matching values ​​of the key image of the operation video of the high-frequency spark machine and the preset discharge image of the spark machine; The third processing subunit is used to aggregate the matching values ​​of the key images of the operation videos of all high-frequency spark machines and the preset discharge images of the spark machines to obtain a matching value set; A first calculation subunit is used to perform root mean square calculation on all data in the matching value set, and use the obtained root mean square value as a threshold for judging whether a key image of the operation video of the high-frequency spark machine is an abnormal discharge image; The second judgment subunit is used to obtain abnormal discharge image information of the high-frequency spark machine based on the threshold for judging whether the key image of the operation video of the high-frequency spark machine is an abnormal discharge image.

9. The high-frequency spark machine calibration device according to claim 7, characterized in that: The first judgment subunit includes: An acquisition subunit is used to acquire historical operation data information, filter abnormal data of the historical operation data information, and calibrate the abnormal data of the historical operation data information to obtain abnormal type information after calibration; A fourth processing subunit is used to process the abnormal data information of the historical operation data information based on the CART algorithm to obtain a CART decision tree, and perform random pruning on the CART decision tree to obtain a decision tree model for judging the abnormal data information; A third judgment subunit is used to send the historical operation data information to the decision tree model for judgment to obtain abnormal data information of the historical operation data information; The third comparison subunit is used to compare the abnormal data information based on the historical operation data information with the calibrated abnormal data information, and adjust the judgment parameters in the decision tree model based on the comparison result until the comparison result is the same as the preset comparison result, thereby obtaining a trained decision tree model.

10. A computer storage medium, characterized in that: The computer storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the high-frequency spark machine calibration method according to any one of claims 1 to 5 are implemented.

Images