GIS defect detection method and system based on X-ray intensity adaptive regulation and control

Abstract

The invention discloses a GIS defect detection method and system based on X-ray intensity self-adaptive regulation and control, and the method comprises the steps: starting a closed-loop self-adaptive regulation and control process through inputting equipment parameters and a safety threshold value; the process preferentially guarantees radiation safety, and intelligently switches to an enhanced excitation mode or an optimization adjustment mode according to the partial discharge signal intensity detected in real time so as to dynamically update the X-ray intensity, thereby effectively exciting latent defects or finding an optimal working point. And when the signal and the intensity change meet the convergence condition, the regulation and control are completed. And then automatically realizing judgment of defect types and positions based on the discharge signal excited by the optimal intensity. According to the method, the limitation of a traditional fixed dose method is overcome, the sensitivity of early defect detection is remarkably improved, and the optimal balance between radiation safety and detection efficiency is realized.

Description

Technical Field

[0001] This invention relates to the field of GIS inspection technology, specifically to a GIS defect detection method and system based on adaptive X-ray intensity control. Background Technology

[0002] Gas-insulated metal-enclosed switchgear (GIS) is one of the key and core pieces of equipment in the power grid. Existing partial discharge detection technologies have low sensitivity and a high false negative rate for latent defects such as tiny air bubbles inside the insulation components of GIS equipment.

[0003] Against the backdrop of the accelerated development of global power systems towards ultra-high voltage and large capacity, latent insulation defects that have long existed within GIS (Gas Insulation System) are gradually evolving into systemic risks threatening the safe and stable operation of the entire power grid. With the continuous increase in operating voltage levels and the increasing complexity of equipment structures, the electro-thermal-mechanical multi-field stresses experienced by the internal insulation system of GIS have significantly increased. This dramatically amplifies the danger of micro-defects that are difficult to detect in the early stages, such as microbubbles remaining from the manufacturing process, weak points at the interfaces of solid insulation components, and particulate contaminants introduced during assembly. Although these defects remain latent under long-term operating voltages, the partial discharge activity they trigger gradually degrades insulation performance over time, potentially leading to insulation breakdown and even cascading failures. Micro-defect-induced discharge is one of the main causes of GIS insulation failure. Therefore, understanding the evolution mechanism, early diagnosis, and effective intervention of latent insulation defects in GIS has become a major cutting-edge topic and urgent technical need in the field of ultra-high voltage power grid insulation reliability research.

[0004] X-ray excitation technology injects sufficient high-energy electrons into the insulation defect region through external irradiation, effectively shortening the statistical delay of discharge formation and thus significantly reducing the partial discharge initiation voltage at the macroscopic level. As a non-invasive external excitation method, this technology can effectively excite latent insulation defects while maintaining the existing operating voltage of the GIS, inducing detectable artificial partial discharge signals. This process not only improves the probability of early insulation defect identification but also provides important data for equipment condition assessment and maintenance. However, traditional X-ray excitation methods often use fixed dose output, which cannot be automatically adjusted according to different voltage levels, shell thicknesses, and detection targets of the GIS equipment. This leads to insufficient or excessive doses in some cases, affecting detection sensitivity and increasing radiation risks. Patent publication number CN120539190A discloses a method for detecting internal defects in GIS using X-rays as the excitation source, but this method does not propose a method for adjusting X-ray intensity. In summary, existing GIS internal defect detection methods lack sensitivity, are passive in defect identification, and have weak active control capabilities for defect characteristics, failing to meet the needs of intelligent operation and maintenance for accurate prediction of equipment status. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a GIS defect detection method and system based on adaptive X-ray intensity control, aiming to solve the problems in the background technology.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a GIS defect detection method based on adaptive X-ray intensity control, comprising the following steps: Step S1: Set the initial X-ray intensity; input the safety parameters and safety convergence threshold of the GIS equipment; the safety parameters include the safety radiation dose threshold and the discharge threshold; the safety convergence threshold includes the average discharge convergence tolerance threshold and the X-ray intensity convergence tolerance threshold. Step S2: Set the X-ray intensity update principle, and execute the multi-mode adaptive control process of steps S3-S5 based on the X-ray intensity update principle; Step S3: Perform radiation safety control and monitor the current radiation dose in real time. If the current radiation dose is greater than or equal to the safe radiation dose threshold, immediately reduce the X-ray intensity to the minimum allowable value and repeat step S3; otherwise, proceed to step S4. Step S4: Based on the comparison between the average discharge quantity of the real-time detected partial discharge signal and the discharge quantity threshold, different control strategies are executed to update the X-ray intensity; Step S5: Check whether the changes in average discharge quantity and X-ray intensity before and after iteration are less than the convergence tolerance thresholds for average discharge quantity and X-ray intensity. If both are satisfied, it is determined that the system has converged to the optimal operating point, the adaptive control is completed, and the process proceeds to step S6; otherwise, return to step S2 to continue iterative control. Step S6: Output the final X-ray intensity that converges to the optimal working point after adaptive adjustment. Use this intensity to continuously excite the GIS equipment to induce partial discharge signals. Extract the characteristic parameters of the partial discharge signals and use the time difference method to automatically identify the defect location.

[0007] Furthermore, step S4 includes: Step S4.1: If the average discharge quantity of the partial discharge signal is lower than the discharge quantity threshold and the current radiation dose is lower than the safe radiation dose threshold, then enter the enhanced excitation mode; the enhanced excitation mode calculates the preliminary intensity correction amount through the proportional-integral controller based on the error between the current discharge signal and the target value, and integrates the fourth-order Runge-Kutta method and the Adams prediction-correction method to improve the control accuracy and dynamically enhance the X-ray intensity. Step S4.2: If the average discharge amount of the partial discharge signal is higher than or equal to the discharge amount threshold, then enter the optimization adjustment mode. The optimization adjustment mode constructs a multi-objective optimization function that comprehensively balances signal quality, radiation safety and equipment energy consumption, and uses an intelligent optimization algorithm based on gradient descent to iteratively optimize and gradually adjust the X-ray intensity to the lowest effective level that can reliably identify defects.

[0008] Furthermore, the enhanced incentive model specifically includes: Step S4.11: Calculate the discharge quantity error at the current moment based on the average discharge quantity and discharge quantity threshold of the partial discharge signal, and calculate the preliminary X-ray intensity correction based on the proportional-integral controller; Step S4.12: Treat the initial X-ray intensity correction output by the proportional-integral controller as a function of X-ray intensity with respect to time; Step S4.13: Apply the fourth-order Runge-Kutta method to calculate four intermediate slope values ​​in sequence based on the current X-ray intensity, time step, and X-ray intensity as a function of time; Step S4.14: Calculate the weighted average of the slopes to obtain the predicted value of the X-ray intensity; Step S4.15: Based on the predicted X-ray intensity, the Adams prediction-correction method is then used to calculate the corrected control output; Step S4.16: Based on the corrected control output, estimate the error of the slope in the last step at the current step size; Step S4.17: Adaptively adjust the time step based on the relationship between the error of the slope in the last step under the current step size and the preset tolerance threshold; Step S4.18: Update the X-ray intensity using the adaptively adjusted time step and the corrected control output to complete this iteration.

[0009] Furthermore, the optimization and adjustment mode specifically includes: Step S4.21: Establish a multi-objective function that balances partial discharge signal quality, radiation safety, and equipment energy consumption; Step S4.22: Calculate the gradient of the multi-objective function and apply the fourth-order Runge-Kutta method to calculate four intermediate gradient step values ​​based on the current X-ray intensity, learning rate and time step. Step S4.23: Take a weighted average of the four intermediate gradient step values ​​to obtain the predicted value of X-ray intensity; Step S4.24: Determine whether the multi-objective function value is smaller than before the iteration. If yes, use the predicted value of X-ray intensity from step S4.23; otherwise, keep the original X-ray intensity.

[0010] Furthermore, applying the fourth-order Runge-Kutta method, based on the current X-ray intensity, time step, and the function of X-ray intensity with respect to time, four intermediate slope values ​​are calculated sequentially, expressed as: ; ; ; ; In the formula, This represents the first intermediate slope value; This represents the second intermediate slope value; This represents the third intermediate slope value; This represents the 4th intermediate slope value; Indicates the time step of numerical integration; Indicates the first X-ray intensity after the next iteration Regarding time The function; The gradient of the multi-objective function is calculated, and the fourth-order Runge-Kutta method is applied to calculate four intermediate gradient step values ​​based on the current X-ray intensity, learning rate, and time step, expressed as: ; ; ; ; In the formula, This represents the first intermediate gradient step value; This represents the second intermediate gradient step value; This represents the third intermediate gradient step value; This represents the 4th intermediate gradient step value; Indicates the learning rate; This represents the gradient of a multi-objective function.

[0011] Furthermore, the specific process of step S6 is as follows: Step S6.1: Under the continuous excitation of the final X-ray intensity, collect the partial discharge signal induced by the GIS equipment and extract the feature parameters for positioning, including the spatial coordinates of multiple sensors deployed around the GIS equipment, the time when each sensor receives the partial discharge signal, and the propagation speed of the partial discharge signal in the GIS equipment. Step S6.2: Defect spatial localization is performed using the time difference method on the extracted feature parameters. First, the first sensor is selected as the reference sensor, and the time when the reference sensor receives the partial discharge signal is obtained as the reference time. Then, for each of the remaining sensors, the time difference between the time when it receives the partial discharge signal and the reference time is calculated. Based on the characteristic of the uniform propagation of the partial discharge signal in the GIS equipment, the distance difference between the discharge source location and the spatial coordinates of each sensor satisfies the relationship that the distance difference is equal to the product of the propagation speed and the time difference, and a hyperbolic equation system is established. Finally, the hyperbolic equation system is solved by the least squares method to obtain the final spatial coordinates of the discharge location, thereby realizing the automatic identification of the defect location.

[0012] Furthermore, the principle for updating X-ray intensity is: the next X-ray intensity equals the current X-ray intensity plus a correction amount.

[0013] A GIS defect detection system based on adaptive X-ray intensity control includes: The GIS parameter input module is used to execute: The initial X-ray intensity is set based on the input electrical parameters and the measured structural parameters; Input the safety parameters and safety convergence threshold of the GIS equipment; The adaptive control module is used to execute: Establish X-ray intensity update principles, and execute multi-mode adaptive control procedures based on X-ray intensity update principles; Radiation safety control is implemented, and the current radiation dose is monitored in real time. If the current radiation dose is greater than or equal to the safe radiation dose threshold, the X-ray intensity is immediately reduced to the minimum allowable value, and radiation safety control is re-implemented. Otherwise, different control strategies are implemented to update the X-ray intensity based on the comparison between the average discharge amount of the real-time detected partial discharge signal and the discharge amount threshold. Check whether the changes in average discharge quantity and X-ray intensity before and after iteration are less than the convergence tolerance thresholds for average discharge quantity and X-ray intensity. If both are satisfied, it is determined that the system has converged to the optimal operating point and the adaptive control is completed; otherwise, iterative control continues. The partial discharge detection module is used to perform: The final X-ray intensity, which converges to the optimal operating point after adaptive adjustment, is output. This intensity is used to continuously excite the GIS equipment, inducing partial discharge signals. The characteristic parameters of the partial discharge signals are extracted, and the time difference method is used to automatically identify the defect location.

[0014] An electronic device includes a processor, a memory, and a bus, wherein the processor and the memory are connected via the bus, wherein the memory is used to store a set of program code, and the processor is used to call the program code stored in the memory to execute a GIS defect detection method based on adaptive X-ray intensity control.

[0015] A non-volatile computer storage medium storing computer-executable instructions that execute a GIS defect detection method based on adaptive X-ray intensity control.

[0016] Compared with existing technologies, the present invention has the following advantages:

[0017] (1) This invention achieves dynamic and precise adjustment of X-ray intensity by constructing a closed-loop adaptive control process and integrating a multi-modal intelligent switching strategy of radiation safety priority control, enhanced excitation mode and optimized adjustment mode. This combined mechanism can actively excite latent defects and automatically converge to the optimal operating point based on the real-time feedback of partial discharge signals, under the premise of ensuring radiation safety. It overcomes the technical limitations of the traditional fixed dose method in balancing detection sensitivity and safety, and significantly improves the ability to identify early insulation defects of GIS equipment.

[0018] (2) In the enhanced excitation mode, the present invention introduces a high-precision numerical control method that integrates a proportional-integral controller with the fourth-order Runge-Kutta method and the Adams prediction-correction method. Through slope weighted averaging, error estimation and adaptive adjustment of time step, the precise and stable control of the X-ray intensity enhancement process is achieved. In the optimization adjustment mode, a multi-objective optimization function that comprehensively balances signal quality, radiation safety and equipment energy consumption is constructed. Iterative optimization is carried out by combining gradient descent and the fourth-order Runge-Kutta method, which effectively ensures the convergence stability of the control process and the accuracy of the operating point optimization.

[0019] (3) After the adaptive control converges, the present invention outputs the optimal X-ray intensity to continuously excite the GIS equipment. It combines the multi-sensor time difference method and the least squares method to construct a hyperbola equation system for defect spatial positioning, realizing a complete detection chain from intensity adaptive control to automatic identification of defect type and location. The whole process does not require manual intervention and can minimize radiation dose and equipment energy consumption while ensuring detection reliability. It provides a technical solution for GIS equipment insulation status assessment that is intelligent, highly sensitive and highly safe. Attached Figure Description

[0020] Figure 1 This is a flowchart of the method of the present invention.

[0021] Figure 2 This is a system structure diagram of the present invention. Detailed Implementation

[0022] like Figure 1 As shown, the present invention provides a technical solution: a GIS defect detection method based on adaptive X-ray intensity control, comprising the following steps: Step S1: Input the safety parameters and safety convergence threshold of the GIS equipment; the safety parameters include the safety radiation dose threshold. Discharge threshold The safe convergence threshold includes the average discharge quantity convergence tolerance threshold. and X-ray intensity convergence tolerance threshold .

[0023] Step S2: Set the X-ray intensity update principle (i.e., the next X-ray intensity equals the current X-ray intensity plus the correction amount), and execute the multi-mode adaptive control process of steps S3-S5 based on the X-ray intensity update principle.

[0024] The principle for updating X-ray intensity is expressed as follows: ; Indicates the first X-ray intensity after the next iteration; Indicates the first The X-ray intensity after the next iteration, i.e., the current X-ray intensity; This represents the X-ray intensity correction amount, which is an adjustment value for the current X-ray intensity.

[0025] Step S3: Implement radiation safety control and monitor the current radiation dose in real time. If the current radiation dose ≥Safe radiation dose threshold Then immediately reduce the X-ray intensity to the minimum permissible value. , that is to say If necessary, repeat step S3; otherwise, proceed to step S4.

[0026] Step S4: Average discharge quantity based on the real-time detected partial discharge signal With discharge threshold By comparing the results, different control strategies were implemented to update the X-ray intensity: Step S4.1: If the average discharge quantity of the partial discharge signal... Below the discharge threshold And the current radiation dose Below the safe radiation dose threshold Then, it enters the enhanced excitation mode. The enhanced excitation mode calculates the initial intensity correction based on the error between the current discharge signal and the target value using a proportional-integral controller, and integrates the fourth-order Runge-Kutta method and the Adams predictive-correction method to improve control accuracy, dynamically enhancing X-ray intensity to effectively activate the discharge signal of latent defects. The enhanced excitation mode specifically includes: Step S4.11: Calculate the current time. Discharge error The initial X-ray intensity correction was calculated based on a proportional-integral (PI) controller. ; This represents the proportionality coefficient. Represents the integral coefficient. express Error in discharge quantity at any given moment; Step S4.12: Input the initial X-ray intensity correction from the proportional-integral (PI) controller. Consider it as a function of X-ray intensity with respect to time. ; Indicates X-ray intensity; Step S4.13: Apply the fourth-order Runge-Kutta method (RK4) to calculate four intermediate slope values ​​sequentially based on the current X-ray intensity, time step, and the function of X-ray intensity with respect to time, in order to improve the iteration accuracy. This is expressed as: ; ; ; ; In the formula, This represents the first intermediate slope value; This represents the second intermediate slope value; This represents the third intermediate slope value; This represents the 4th intermediate slope value; Indicates the time step of numerical integration; Indicates the first X-ray intensity after the next iteration Regarding time The function; Step S4.14: Calculate the predicted X-ray intensity by weighting the slopes. , is represented as: ; Step S4.15: Apply the Adams predictive-correction method again to calculate the corrected control output. : ; In the formula, The intensity function represents the predicted state; Step S4.16: Estimate the error of the slope in the last step at the current step size. ; Step S4.17: Based on the error With respect to the preset tolerance threshold The relationship allows for adaptive adjustment of the time step. ; Step S4.18: Use the adaptively adjusted time step and the corrected control output Update X-ray intensity to complete this iteration: .

[0027] Step S4.2: If the average discharge quantity of the partial discharge signal... Higher than or equal to the discharge threshold Then, it enters the optimization and adjustment mode. The optimization and adjustment mode constructs a multi-objective optimization function that comprehensively balances signal quality, radiation safety, and equipment energy consumption, and uses a gradient descent-based intelligent optimization algorithm for iterative optimization, gradually adjusting the X-ray intensity to the lowest effective level that can reliably identify defects. The optimization and adjustment mode specifically includes: Step S4.21: Establish a multi-objective function that balances partial discharge signal quality, radiation safety, and equipment energy consumption. : ; In the formula, This represents the partial discharge signal quality weighting coefficient, used to adjust the weight of the average discharge quantity in the multi-objective function, reflecting the degree of importance attached to the reliability of defect identification; This represents the radiation safety weighting coefficient, used to adjust the weight of radiation dose in the objective function, reflecting the degree of importance attached to radiation protection safety. This represents the weighting coefficient for usage efficiency (equipment energy consumption), used to adjust the weight of X-ray intensity in the objective function, reflecting the degree of importance attached to equipment energy consumption and usage efficiency. Indicates X-ray intensity The average discharge quantity of the partial discharge signal measured below; Indicates X-ray intensity The current radiation dose measured below; Step S4.22: Calculate the gradient of the multi-objective function and apply the fourth-order Runge-Kutta method (RK4) to calculate four intermediate gradient step values ​​based on the current X-ray intensity, learning rate, and time step, expressed as: ; ; ; ; In the formula, This represents the first intermediate gradient step value; This represents the second intermediate gradient step value; This represents the third intermediate gradient step value; This represents the 4th intermediate gradient step value; Indicates the learning rate; Represents the gradient of a multi-objective function; Step S4.23: Calculate the weighted average of the four intermediate gradient step values ​​to obtain the predicted X-ray intensity, expressed as: ; Step S4.24: Determine if the multi-objective function value is smaller than before iteration. If yes, then use the predicted X-ray intensity value from step S4.23, i.e., let... If not, then maintain the original X-ray intensity.

[0028] In the multi-objective function, the first term is the signal quality term, which represents the average discharge quantity and discharge quantity threshold of the discharge signal measured under the current X-ray intensity. The ratio. This value needs to be large enough to ensure the signal is clearly distinguishable. This represents the importance of signal quality in the overall objective. The larger the value, the higher the requirement for detection sensitivity; The second item is the safety penalty, which represents the difference between the radiation dose produced at the current X-ray intensity and the safe radiation dose threshold. The ratio. This value needs to be as small as possible to ensure safety. This represents the weight of radiation safety in the overall objective. The larger the value, the stricter the requirements for radiation safety; The third item is the efficiency cost item, which represents the current X-ray intensity and the maximum allowable value of X-ray intensity. The ratio. In practice, it is desirable to work with the lowest possible X-ray intensity, which saves energy, extends equipment life, and further reduces potential risks. The negative sign at the beginning means that the X-ray intensity itself is a cost that is to be minimized. The larger the value, the higher the requirement for energy conservation and emission reduction; The above three weighting coefficients , , The three numbers are independent non-negative real numbers, and their sum must be 1.

[0029] Step S5: Check the change in average discharge quantity before and after iteration. and the change in X-ray intensity itself Is it less than the average discharge convergence tolerance threshold? and X-ray intensity convergence tolerance threshold ( , They represent the first , If the average discharge after each iteration is satisfied, it is determined that the system has converged to the optimal operating point, the adaptive control is completed, and the process proceeds to step S6; otherwise, the process returns to step S2 to continue iterative control.

[0030] Step S6: Output the final X-ray intensity that converges to the optimal working point after adaptive adjustment. Use this intensity to continuously excite the GIS equipment to induce partial discharge signals. Extract the characteristic parameters of the partial discharge signals and use the time difference method to automatically identify the defect location.

[0031] The specific process of step S6 is as follows: Step S6.1: Under continuous excitation by the final X-ray intensity, acquire the partial discharge signal induced by the GIS equipment and extract the feature parameters for positioning, including the spatial coordinates of multiple sensors deployed around the GIS equipment. ( , , They represent the first (x, y, z axis coordinates of each sensor in three-dimensional space), and the time when each sensor receives the partial discharge signal. And the propagation speed of partial discharge signals in GIS equipment. The partial discharge signal is composed of electromagnetic waves excited by partial discharge at the defect. Step S6.2: Use the time difference method to locate the extracted feature parameters in the defect space. First, select the first sensor as the reference sensor and obtain the time when the reference sensor receives the partial discharge signal as the reference time. Then, for each of the remaining sensors, the time when it receives the partial discharge signal is calculated. Relative to the reference time time difference Based on the characteristic of partial discharge signals propagating at a uniform speed in GIS equipment, the location of the discharge source is determined. (Defect location) and spatial coordinates of each sensor The distance difference between them should satisfy the relationship that the propagation speed is equal to the product of the time difference, which can be established as the following hyperbolic equation system: ; Finally, the spatial coordinates of the final discharge location were obtained by solving the hyperbola equations using the least squares method. , represented as: ; In the formula, Indicates the number of sensors.

[0032] like Figure 2 As shown, a second embodiment of the present invention also provides a GIS defect detection system based on adaptive X-ray intensity control, comprising: The GIS parameter input module is used to execute: The initial X-ray intensity is set based on the input electrical parameters (rated voltage) and the measured structural parameters (shell parameters); Input the safety parameters and safety convergence threshold of the GIS equipment; The adaptive control module is used to execute: Establish X-ray intensity update principles, and execute multi-mode adaptive control procedures based on X-ray intensity update principles; Radiation safety control is implemented, and the current radiation dose is monitored in real time. If the current radiation dose is greater than or equal to the safe radiation dose threshold, the X-ray intensity is immediately reduced to the minimum allowable value, and radiation safety control is re-implemented. Otherwise, different control strategies are implemented to update the X-ray intensity based on the comparison between the average discharge amount of the real-time detected partial discharge signal and the discharge amount threshold. Check whether the changes in average discharge quantity and X-ray intensity before and after iteration are less than the convergence tolerance thresholds for average discharge quantity and X-ray intensity. If both are satisfied, it is determined that the system has converged to the optimal operating point and the adaptive control is completed; otherwise, iterative control continues. The partial discharge detection module is used to perform: The final X-ray intensity, which converges to the optimal operating point after adaptive adjustment, is output. This intensity is used to continuously excite the GIS equipment, inducing partial discharge signals. The characteristic parameters of the partial discharge signals are extracted, and the time difference method is used to automatically identify the defect location.

[0033] A third embodiment of the present invention also provides an electronic device, including a processor, a memory, and a bus, wherein the processor and the memory are connected via the bus, wherein the memory is used to store a set of program code, and the processor is used to call the program code stored in the memory to execute a GIS defect detection method based on adaptive X-ray intensity control.

[0034] A fourth embodiment of the present invention also provides a non-volatile computer storage medium storing computer-executable instructions that execute a GIS defect detection method based on adaptive X-ray intensity control.

[0035] Example 5

[0036] Based on Example 2, taking a typical 1100kV UHV GIS equipment as an example, its operation begins in the initialization phase. The system automatically sets the initial X-ray intensity to tube voltage of 180kV and tube current of 3.0mA according to the input rated voltage and shell parameters. At the same time, it determines the safe radiation dose threshold to be 5.0μSv / h and the discharge threshold to be 2.5pC, and limits the intensity adjustment range to between 1.0mA and 5.0mA to ensure operational safety.

[0037] During the adaptive control phase, after the system applies initial X-ray excitation, it simultaneously acquires the average discharge quantity and radiation dose of the partial discharge signal and enters a multi-mode adaptive control process. When the radiation dose reaches 5.0 μSv / h, the system immediately and unconditionally reduces the X-ray intensity to the minimum safe value. If the average discharge quantity of the partial discharge signal is below 2.5 pC and the radiation is within the safe range, it enters the enhanced excitation mode. The system dynamically adjusts the X-ray intensity using the fourth-order Runge-Kutta method with adaptive step size and the Adams prediction-correction method, rapidly approaching the preset upper limit while ensuring stability. When the average discharge quantity of the partial discharge signal is higher than or equal to the discharge quantity threshold, the system switches to the optimization adjustment mode to achieve optimization search. In this mode, multiple objective functions comprehensively balance the discharge signal quality, radiation safety, and equipment energy consumption.

[0038] This adaptive control process continues until the average discharge change is less than 0.2 pC and the intensity change is less than 0.1 mA over three consecutive control cycles. At this point, the system determines that it has converged to the optimal operating point. Finally, the system outputs the optimized X-ray intensity, extracts the statistical features of the discharge signal, and uses a pattern recognition algorithm to automatically identify the defect type and location. It then generates a standardized inspection report containing specific defect interpretation, confidence assessment, and health index, completing the entire intelligent inspection process.

[0039] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A GIS defect detection method based on adaptive X-ray intensity control, characterized in that, Includes the following steps: Step S1: Set the initial X-ray intensity; input the safety parameters and safety convergence threshold of the GIS equipment; the safety parameters include the safety radiation dose threshold and the discharge threshold; the safety convergence threshold includes the average discharge convergence tolerance threshold and the X-ray intensity convergence tolerance threshold. Step S2: Set the X-ray intensity update principle, and execute the multi-mode adaptive control process of steps S3-S5 based on the X-ray intensity update principle; Step S3: Perform radiation safety control and monitor the current radiation dose in real time. If the current radiation dose is greater than or equal to the safe radiation dose threshold, immediately reduce the X-ray intensity to the minimum allowable value and repeat step S3; otherwise, proceed to step S4. Step S4: Based on the comparison between the average discharge quantity of the real-time detected partial discharge signal and the discharge quantity threshold, different control strategies are executed to update the X-ray intensity; Step S5: Check whether the changes in average discharge quantity and X-ray intensity before and after iteration are less than the average discharge quantity convergence tolerance threshold and the X-ray intensity convergence tolerance threshold. If both are satisfied, it is determined that the process has converged to the optimal operating point, the adaptive control is completed, and the process proceeds to step S6. Otherwise, return to step S2 and continue iterative adjustment; Step S6: Output the final X-ray intensity that converges to the optimal working point after adaptive adjustment. Use this intensity to continuously excite the GIS equipment to induce partial discharge signals. Extract the characteristic parameters of the partial discharge signals and use the time difference method to automatically identify the defect location.

2. The GIS defect detection method based on adaptive X-ray intensity control according to claim 1, characterized in that: Step S4 includes: Step S4.1: If the average discharge quantity of the partial discharge signal is lower than the discharge quantity threshold and the current radiation dose is lower than the safe radiation dose threshold, then enter the enhanced excitation mode; the enhanced excitation mode calculates the preliminary intensity correction amount through the proportional-integral controller based on the error between the current discharge signal and the target value, and integrates the fourth-order Runge-Kutta method and the Adams prediction-correction method to improve the control accuracy and dynamically enhance the X-ray intensity. Step S4.2: If the average discharge amount of the partial discharge signal is higher than or equal to the discharge amount threshold, then enter the optimization and adjustment mode. The optimization and adjustment mode constructs a multi-objective optimization function that comprehensively balances signal quality, radiation safety and equipment energy consumption, and uses an intelligent optimization algorithm based on gradient descent to iteratively optimize and gradually adjust the X-ray intensity to the lowest effective level for identifying defects.

3. The GIS defect detection method based on adaptive X-ray intensity control according to claim 2, characterized in that: The enhanced incentive model specifically includes: Step S4.11: Calculate the discharge quantity error at the current moment based on the average discharge quantity and discharge quantity threshold of the partial discharge signal, and calculate the preliminary X-ray intensity correction based on the proportional-integral controller; Step S4.12: Treat the initial X-ray intensity correction output by the proportional-integral controller as a function of X-ray intensity with respect to time; Step S4.13: Apply the fourth-order Runge-Kutta method to calculate four intermediate slope values ​​in sequence based on the current X-ray intensity, time step, and X-ray intensity as a function of time; Step S4.14: Calculate the weighted average of the slopes to obtain the predicted value of the X-ray intensity; Step S4.15: Based on the predicted X-ray intensity, the Adams prediction-correction method is then used to calculate the corrected control output; Step S4.16: Based on the corrected control output, estimate the error of the slope in the last step at the current step size; Step S4.17: Adaptively adjust the time step based on the relationship between the error of the slope in the last step under the current step size and the preset tolerance threshold; Step S4.18: Update the X-ray intensity using the adaptively adjusted time step and the corrected control output to complete this iteration.

4. The GIS defect detection method based on adaptive X-ray intensity control according to claim 3, characterized in that: The optimization and adjustment mode specifically includes: Step S4.21: Establish a multi-objective function that balances partial discharge signal quality, radiation safety, and equipment energy consumption; Step S4.22: Calculate the gradient of the multi-objective function and apply the fourth-order Runge-Kutta method to calculate four intermediate gradient step values ​​based on the current X-ray intensity, learning rate and time step. Step S4.23: Take a weighted average of the four intermediate gradient step values ​​to obtain the predicted value of X-ray intensity; Step S4.24: Determine whether the multi-objective function value is smaller than before the iteration. If yes, use the predicted value of X-ray intensity from step S4.23; otherwise, keep the original X-ray intensity.

5. The GIS defect detection method based on adaptive X-ray intensity control according to claim 4, characterized in that: Applying the fourth-order Runge-Kutta method, based on the current X-ray intensity, time step, and the function of X-ray intensity with respect to time, four intermediate slope values ​​are calculated sequentially, expressed as: ； ； ； ； In the formula, This represents the first intermediate slope value; This represents the second intermediate slope value; This represents the third intermediate slope value; This represents the fourth intermediate slope value; Indicates the time step of numerical integration; Indicates the first X-ray intensity after the next iteration Regarding time The function; The gradient of the multi-objective function is calculated, and the fourth-order Runge-Kutta method is applied to calculate four intermediate gradient step values ​​based on the current X-ray intensity, learning rate, and time step, expressed as: ； ； ； ； In the formula, This represents the first intermediate gradient step value; This represents the second intermediate gradient step value; This represents the third intermediate gradient step value; This represents the 4th intermediate gradient step value; Indicates the learning rate; This represents the gradient of a multi-objective function.

6. The GIS defect detection method based on adaptive X-ray intensity control according to claim 5, characterized in that: The specific process of step S6 is as follows: Step S6.1: Under the continuous excitation of the final X-ray intensity, collect the partial discharge signal induced by the GIS equipment and extract the feature parameters for positioning, including the spatial coordinates of multiple sensors deployed around the GIS equipment, the time when each sensor receives the partial discharge signal, and the propagation speed of the partial discharge signal in the GIS equipment. Step S6.2: Defect spatial localization is performed using the time difference method on the extracted feature parameters. First, the first sensor is selected as the reference sensor, and the time when the reference sensor receives the partial discharge signal is obtained as the reference time. Then, for each of the remaining sensors, the time difference between the time when it receives the partial discharge signal and the reference time is calculated. Based on the characteristic of the uniform propagation of the partial discharge signal in the GIS equipment, the distance difference between the discharge source location and the spatial coordinates of each sensor satisfies the relationship that the distance difference is equal to the product of the propagation speed and the time difference, and a hyperbolic equation system is established. Finally, the hyperbolic equation system is solved by the least squares method to obtain the final spatial coordinates of the discharge location, thereby realizing the automatic identification of the defect location.

7. The GIS defect detection method based on adaptive X-ray intensity control according to claim 6, characterized in that: The principle for updating X-ray intensity is: the next X-ray intensity equals the current X-ray intensity plus a correction amount.

8. A GIS defect detection system based on adaptive X-ray intensity control, used to execute the GIS defect detection method based on adaptive X-ray intensity control according to any one of claims 1-7, characterized in that, include: The GIS parameter input module is used to execute: The initial X-ray intensity is set based on the input electrical parameters and the measured structural parameters; Input the safety parameters and safety convergence threshold of the GIS equipment; The adaptive control module is used to execute: Establish X-ray intensity update principles, and execute multi-mode adaptive control procedures based on X-ray intensity update principles; Radiation safety control is implemented, and the current radiation dose is monitored in real time. If the current radiation dose is greater than or equal to the safe radiation dose threshold, the X-ray intensity is immediately reduced to the minimum allowable value, and radiation safety control is re-implemented. Otherwise, different control strategies are implemented to update the X-ray intensity based on the comparison between the average discharge amount of the real-time detected partial discharge signal and the discharge amount threshold. Check whether the changes in average discharge quantity and X-ray intensity before and after the iteration are less than the average discharge quantity convergence tolerance threshold and the X-ray intensity convergence tolerance threshold. If both are satisfied, it is determined that the system has converged to the optimal operating point and the adaptive control is completed. Otherwise, continue iterative adjustments; The partial discharge detection module is used to perform: The final X-ray intensity, which converges to the optimal operating point after adaptive adjustment, is output. This intensity is used to continuously excite the GIS equipment, inducing partial discharge signals. The characteristic parameters of the partial discharge signals are extracted, and the time difference method is used to automatically identify the defect location.

9. An electronic device, characterized in that, The device includes a processor, a memory, and a bus, wherein the processor and the memory are connected via the bus, the memory is used to store a set of program code, and the processor is used to call the program code stored in the memory to execute a GIS defect detection method based on adaptive X-ray intensity control as described in any one of claims 1-7.

10. A non-volatile computer storage medium storing computer-executable instructions, characterized in that, The computer can execute instructions to perform a GIS defect detection method based on adaptive X-ray intensity control as described in any one of claims 1-7.

Images