Method and system for determining optimal parameters of X-ray excitation extra-high voltage composite apparatus

Abstract

The invention discloses a method and a system for determining optimal parameters of an X-ray excited extra-high voltage composite apparatus, and relates to the technical field of extra-high voltage electrical equipment detection. The method comprises the following steps: remotely operating the X-ray machine and calculating partial discharge intensity; for the current irradiation position, the tube voltage and the tube current are gradually adjusted, parameters are dynamically optimized based on the quantitative relation between the partial discharge intensity and the radiation dose rate change rate, and the optimal parameters are recorded; and changing position repetition parameter optimization, and constructing a BP neural network model after data set training optimization so as to predict an optimal irradiation position. The optimized BP neural network adopts an improved sparrow search algorithm to optimize an initial weight and a threshold value, and combines Bayesian to optimize a learning rate and a hidden node number. Accurate optimization of excitation parameters and irradiation positions is achieved, detection accuracy and safety are both considered, the optimal irradiation position positioning efficiency is improved, and the method is suitable for ultra-high voltage GIS detection of different voltage grades.

Description

Technical Field

[0001] This invention relates to the field of ultra-high voltage electrical equipment testing technology, specifically to a method and system for determining the optimal parameters of an X-ray-excited ultra-high voltage combined electrical appliance. Background Technology

[0002] Ultra-high voltage (UHV) and extra-high voltage (UHV) power transmission networks are crucial for long-distance power transmission, and UHV combined switchgear (GIS) is an important component. Many operation and maintenance (O&M) technologies have been developed to ensure its safe and stable operation. However, due to the high operating voltage of UHV GIS, even minor insulation defects can trigger discharges with serious consequences. X-ray excitation for detecting insulation defects within UHV GIS is an effective technology. Patent publication number CN120539190A discloses a method for detecting insulation defects within GIS using X-ray excitation and imaging efficacy. However, this method does not propose how to adjust X-ray parameters to guide field application; it only suggests that detection can be performed using X-ray excitation efficacy. Due to the large size and thick outer metal casing of UHV GIS, X-rays experience significant attenuation when penetrating metal. However, increasing the X-ray intensity increases radiation exposure to field workers, affecting their safety. Therefore, determining the X-ray intensity presents a conflict between excitation effectiveness and worker safety. Thus, there is an urgent need to develop a method for determining the optimal X-ray excitation parameters for field application. Summary of the Invention

[0003] To address the shortcomings of existing technologies, this invention provides a method and system for determining the optimal parameters of X-ray-excited ultra-high voltage combined electrical appliances. The purpose is to achieve optimal X-ray excitation of GIS insulation defects while minimizing X-ray radiation, ensuring the safety of operators, and guiding on-site operators in adjusting X-ray parameters.

[0004] To achieve the above objectives, the present invention provides the following technical solution: a method for determining the optimal parameters of an X-ray-excited ultra-high voltage combined electrical appliance, comprising the following steps: Step S1: Position the X-ray machine's emission port directly against the wall of the ultra-high voltage GIS cavity, and set the X-ray machine's tube voltage and tube current to 0. Measure the radiation dose rate in the environment. If the radiation dose rate meets the preset safety conditions, proceed to the next step. Step S2: The operator operates the X-ray machine from a preset safe distance and turns on the ultra-high frequency partial discharge detection device to filter out background noise in the space and calculate the partial discharge intensity; Step S3: For the current irradiation position, gradually increase the X-ray machine tube voltage and tube current starting from 0, monitor the change amplitude of partial discharge intensity and radiation dose rate in real time, and after each adjustment, further adjust the X-ray machine tube voltage and tube current according to the quantitative relationship of the change amplitude until the stopping condition is met. Record the maximum value of partial discharge intensity at the irradiation position and the corresponding optimal tube voltage, tube current and irradiation geometric position parameters. Step S4: Move the X-ray machine and repeat step S3 at each spatial location to obtain the maximum value of the partial discharge intensity at different spatial locations; compare the partial discharge intensity to determine the global optimal irradiation location and use this location as the output label; conduct experiments on different types or specifications of ultra-high voltage GIS to obtain output data under different inputs, collect this output data as historical data and construct a dataset, train the optimized BP neural network model through the dataset, and obtain the optimal irradiation location prediction model. Step S5: Input the real-time X-ray electrical parameters and irradiation geometric position parameters into the optimal irradiation position prediction model, and output the optimal irradiation position of the X-ray machine.

[0005] Furthermore, the X-ray machine is a continuous X-ray machine, and the distance between the X-ray machine and the ultra-high voltage GIS... Set the voltage level accordingly; use a radiation dose detector to detect the radiation dose rate in the environment to determine if there is a risk of leakage from the radiation machine. If the radiation dose rate meets the preset safety conditions, proceed to the next step.

[0006] Further, the specific process of step S2 is as follows: noise filtering is performed on the voltage signal collected by the UHF partial discharge detection device to remove spatial background noise; the time of partial discharge occurrence is located by the energy accumulation function, and the partial discharge data after the time of partial discharge occurrence is recorded; Fourier transform is performed on the partial discharge data to obtain the frequency domain signal; the frequency domain signal is normalized, the equivalent average frequency and the effective frequency range are calculated, and the frequency domain signal within the effective frequency range is retained; the inverse Fourier transform is performed on the retained frequency domain signal to obtain the effective partial discharge energy signal, and the partial discharge intensity is calculated after noise is filtered out.

[0007] Furthermore, step S3 includes: gradually increasing the X-ray tube voltage starting from 0 for the current irradiation position. and tube current , Indicates the adjustment level / step. This indicates the adjustment step size / iteration number; the X-ray radiation dose rate at the operator's location is detected in real time using a dose meter. Establish constraints: , , ; when hour, Indicates the effective partial discharge energy signal. Indicates ambient noise; stop increasing the tube voltage. Record the radiation dose rate at this time. If the value satisfies the constraint condition, then continue to increase the tube voltage and tube current; after each increase in tube voltage and tube current, analyze the partial discharge intensity. and radiation dose rate The changing trend of partial discharge intensity is used to calculate the partial discharge intensity. and radiation dose rate The magnitude of the change; like Then continue to increase the tube voltage and tube current; Indicates the first The amplitude of the change in partial discharge intensity under each adjustment cycle; Indicates the first Real-time value of partial discharge intensity under each adjustment cycle; like ,and If the voltage of the transistor is increased, the current of the transistor remains unchanged. Indicates the first The amplitude of radiation dose rate change under each adjustment cycle; Indicates the first Real-time radiation dose rate value under each adjustment cycle; like ,and Then the tube voltage and tube current remain unchanged; like Then the tube voltage and tube current remain unchanged; Record the maximum partial discharge intensity at the irradiation location and the corresponding optimal tube voltage, tube current, and irradiation geometric position parameters.

[0008] Further, the specific process of step S4 is as follows: Define the input variables of the optimized BP neural network model as X-ray electrical parameters and irradiation geometric position parameters, and the output variable as the optimal irradiation position, and construct a mapping relationship; the X-ray electrical parameters include tube voltage and tube current; the irradiation geometric position parameters include the irradiation angle of the ultra-high voltage GIS and the straight-line distance of the irradiation from the insulator; move the X-ray machine and repeat step S3 at each spatial position to obtain the maximum value of the partial discharge intensity under different spatial positions, determine the global optimal irradiation position by comparing the partial discharge intensity, and use this position as the output label; conduct experiments on different types or specifications of ultra-high voltage GIS to obtain output data under different inputs, collect this output data as historical data and construct a dataset; use the dataset to train the optimized BP neural network model to obtain the optimal irradiation position prediction model.

[0009] Furthermore, the optimized BP neural network model includes: optimizing the initial weights and thresholds of the BP neural network model based on the improved sparrow search algorithm; and optimizing the learning rate and the number of hidden nodes of the BP neural network model using Bayesian optimization. The improved sparrow search algorithm improves the initial population generation method and connection weight of the original sparrow search algorithm, including introducing adaptive mapping to uniformly order the initial population and improving the connection weight of the sparrow search algorithm.

[0010] Furthermore, the connection weighting of the sparrow search algorithm is improved as follows: ; ; In the formula, Indicates the first The connection ratio in each iteration; Indicates the current iteration number; Indicates the maximum number of iterations; Represents individual sparrows No. During the nth iteration, at the... The position of the dimension; Represents a random number between 0 and 1; Represents individual sparrows No. During the nth iteration, at the... The position of the dimension; and All of these represent system adjustment parameters; Indicates the defined boundary; This indicates the foraging range of an individual sparrow.

[0011] A system for determining the optimal parameters of an X-ray-excited ultra-high voltage combined electrical appliance includes: The X-ray radiation dose rate measurement module is used to point the X-ray machine's emission port directly at the wall of the ultra-high voltage GIS cavity, and set the X-ray machine tube voltage and tube current to 0 to measure the radiation dose rate in the environment. If the radiation dose rate meets the preset safety conditions, proceed to the next step. The partial discharge intensity calculation module is used by operators to calculate the partial discharge intensity when operating the X-ray machine at a preset safe distance and turning on the ultra-high frequency partial discharge detection device to filter out background noise in the space. The parameter adjustment module is used to gradually increase the X-ray machine tube voltage and tube current from 0 for the current irradiation position, monitor the change amplitude of partial discharge intensity and radiation dose rate in real time, and further adjust the X-ray machine tube voltage and tube current according to the quantitative relationship of the change amplitude after each adjustment until the stop condition is met. It records the maximum value of partial discharge intensity at the irradiation position and the corresponding optimal tube voltage, tube current and irradiation geometric position parameters. The model training module is used to move the X-ray machine and repeat step S3 at each spatial location to obtain the maximum value of the partial discharge intensity at different spatial locations; compare the partial discharge intensity to determine the global optimal irradiation location and use this location as the output label; conduct experiments on different types or specifications of ultra-high voltage GIS to obtain output data under different inputs, collect this output data as historical data and construct a dataset, and train the optimized BP neural network model through the dataset to obtain the optimal irradiation location prediction model; The output module is used to input real-time X-ray electrical parameters and irradiation geometric position parameters into the optimal irradiation position prediction model, and output the optimal irradiation position of the X-ray machine.

[0012] 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 codes, and the processor is used to call the program codes stored in the memory to execute a method for determining the optimal parameters of an X-ray-excited ultra-high voltage combined electrical appliance.

[0013] A non-volatile computer storage medium storing computer-executable instructions that execute a method for determining the optimal parameters of an X-ray-excited ultra-high voltage combined electrical appliance.

[0014] Compared with existing technologies, the present invention has the following advantages: (1) This invention integrates innovative designs such as pre-radiation safety detection, precise processing of UHF partial discharge signals, dynamic quantitative adjustment of parameters and optimized BP neural network prediction, and achieves in-depth synergistic optimization of the excitation effect of insulation defects in ultra-high voltage combined electrical appliances, radiation safety of operators and positioning accuracy of the best irradiation position. It can systematically and accurately determine the optimal parameters of X-ray machine tube voltage, tube current and irradiation geometric position in all dimensions, providing scientific and reliable technical support for on-site detection operations.

[0015] (2) This invention, through radiation safety screening in the initial stage, setting equipment spacing according to voltage level differences, and dynamic parameter control based on the relationship between partial discharge intensity and radiation dose rate, maximizes the excitation of insulation defect signals to ensure detection sensitivity, while strictly controlling the X-ray radiation dose within the safe threshold. This effectively avoids the radiation risk to operators and avoids the excitation saturation problem caused by redundant radiation intensity, significantly reducing the hardware investment and operating losses of X-ray machines.

[0016] (3) This invention constructs a comprehensive dataset by collecting parameters from multiple illumination locations and combines a BP neural network model optimized by both the improved sparrow search algorithm and Bayesian optimization to predict the best illumination location. This not only solves the problems of inefficiency and insufficient accuracy of traditional positioning methods, but also significantly improves the positioning efficiency and accuracy of the best illumination location. Furthermore, it can comprehensively cover potential defect areas by adapting to the detection needs of devices with different voltage levels, thus significantly enhancing the method's scene adaptability and detection comprehensiveness. Attached Figure Description

[0017] Figure 1 This is a flowchart of the method of the present invention.

[0018] Figure 2 This is a flowchart of the optimized BP neural network model parameters obtained in this invention.

[0019] Figure 3 A sequence of discharge pulses used in GIS defect detection. Detailed Implementation

[0020] like Figure 1 As shown, the present invention provides a technical solution: a method for determining the optimal parameters of an X-ray-excited ultra-high voltage combined electrical appliance, comprising the following steps: Step S1: Position the X-ray machine's emission port directly against the wall of the ultra-high voltage GIS cavity, and set the X-ray machine's tube voltage and tube current to 0. Measure the radiation dose rate in the environment. If the radiation dose rate meets the preset safety conditions, proceed to the next step.

[0021] In step S1, the X-ray machine is a continuous X-ray machine, and the distance between the X-ray machine and the ultra-high voltage GIS is... The distance to ultra-high voltage GIS is set according to the voltage level, which is 500kV and below. For ultra-high voltage GIS distances within 5 centimeters and greater than 500kV but less than 750kV, the distance is... The distance between ultra-high voltage GIS (Gas-Inductor) and extra-high voltage (UHV) systems is 5 to 10 centimeters, and is 750 kV and above. The radiation dose rate in the environment is measured at a depth of 10 to 20 centimeters using a radiation dose detector to determine if there is a risk of leakage from the radiation machine. If the radiation dose rate is <5 mGy / s, proceed to the next step.

[0022] Step S2: The operator operates the X-ray machine from a preset safe distance and turns on the ultra-high frequency partial discharge detection device to filter out background noise in the space and calculate the partial discharge intensity.

[0023] The specific process of step S2 is as follows: Step S21: Noise filtering is performed on the voltage signal acquired by the UHF partial discharge detection device to remove spatial background noise, and the moment of partial discharge occurrence is located by the energy accumulation function. Record the time of partial discharge. Subsequent partial discharge data; The energy accumulation function is expressed as: ; In the formula, Indicates the cumulative amount of energy. Indicates the current moment; This indicates the voltage amplitude acquired by the ultra-high frequency partial discharge detection device; Indicates spatial background noise; Represents the time element of the differential of the integral; when The moment when a significant increase occurs is the time when partial discharge occurs, denoted as . Record the subsequent partial discharge data Furthermore, the coupled electromagnetic wave signals are eliminated.

[0024] Step S22: Process partial discharge data The frequency domain signal is obtained by performing a Fourier transform. For frequency domain signals After normalization, the equivalent average frequency and effective frequency range are calculated. The frequency domain signal within the effective frequency range is retained and represented as: ; ; ; In the formula, Represents the normalized frequency domain signal ; Indicates angular frequency; To represent the differential; Indicates the equivalent average frequency; Characteristic parameters representing the effective frequency range; Filtering frequency domain signals The signal within the range, the preserved frequency domain signal is .

[0025] Step S23: For the retained frequency domain signal Perform inverse Fourier transform to obtain effective partial discharge energy signal The partial discharge intensity was calculated after filtering out noise. , is represented as: ; In the formula, This indicates the gain of the ultra-high frequency sensor, which is the core component of the ultra-high frequency partial discharge detection device. This represents the system calibration coefficient.

[0026] Step S3: For the current irradiation position, gradually increase the X-ray machine tube voltage and tube current starting from 0, monitor the changes in partial discharge intensity and radiation dose rate in real time, and after each adjustment, further adjust the X-ray machine tube voltage and tube current according to the quantitative relationship of the change in amplitude until the stopping condition is met. Record the maximum value of partial discharge intensity at the irradiation position and the corresponding optimal tube voltage, tube current and irradiation geometric position parameters.

[0027] Step S3 includes: Gradually increasing the X-ray tube voltage, starting from 0, for the current irradiation position. and tube current , Indicates the adjustment level / step. Indicates the adjustment step size / number of iterations; tube voltage. Each increase in amplitude is 5kV, tube current Each increment is 0.5 mA; the X-ray radiation dose rate at the worker's location is monitored in real time using a dose detector. Establish variables ( , , Constraints: , , .

[0028] Due to partial discharge intensity and radiation dose rate The optimal value is contradictory, namely, the desired partial discharge intensity The larger the better, radiation dose rate The smaller the better, but the partial discharge intensity Increased radiation dose rate It will also increase, so a balance needs to be struck between the two; when Stop increasing the tube voltage Record the radiation dose rate at this time. If the value satisfies the above constraints, then continue to increase the tube voltage and tube current; after each increase in tube voltage and tube current, analyze the partial discharge intensity. and radiation dose rate The changing trend of partial discharge intensity is used to calculate the partial discharge intensity. and radiation dose rate Amplitude of change: ; ; In the formula, Indicates the first The amplitude of the change in partial discharge intensity under each adjustment cycle; , They represent the first , Real-time value of partial discharge intensity under each adjustment cycle; Indicates the first The amplitude of radiation dose rate change under each adjustment cycle; , They represent the first , Real-time radiation dose rate value under each adjustment cycle; like Then continue to increase the tube voltage and tube current; like ,and If the voltage of the transistor is increased, the current of the transistor remains unchanged. like ,and Then the tube voltage and tube current remain unchanged; like Then the tube voltage and tube current remain unchanged; Record the maximum partial discharge intensity at the irradiation location and the corresponding optimal tube voltage, tube current, and irradiation geometric position parameters.

[0029] Step S4: Move the X-ray machine and repeat step S3 at each spatial location to obtain the maximum value of the partial discharge intensity at different spatial locations; compare the partial discharge intensity to determine the global optimal irradiation location and use this location as the output label; conduct experiments on different types or specifications of ultra-high voltage GIS to obtain output data under different inputs, collect this output data as historical data and construct a dataset, train the optimized BP neural network model through the dataset, and obtain the optimal irradiation location prediction model.

[0030] The specific process of step S4 is as follows: Define the model input variables as X-ray electrical parameters (tube voltage). tube current ) and irradiation geometric position parameters (irradiation angle of ultra-high voltage GIS) The straight-line distance of the irradiation distance from the insulator The output variable is the optimal irradiation position, and a mapping relationship is constructed. The X-ray machine is moved, and step S3 is repeated at each spatial position to obtain the maximum value of the local discharge intensity at different spatial positions. The global optimal irradiation position is determined by comparing the local discharge intensity, and this position is used as the output label. Experiments are carried out on different types or specifications of ultra-high voltage GIS to obtain output data under different inputs. This output data is collected as historical data and a dataset is constructed. The optimized BP neural network model is trained using the dataset to obtain the optimal irradiation position prediction model.

[0031] Among them, conducting experiments on different types or specifications of ultra-high voltage GIS and obtaining output data under different inputs is actually the same as applying the previous steps to different types or specifications of ultra-high voltage GIS to obtain the corresponding X-ray electrical parameters, irradiation geometric position parameters, and the corresponding global optimal irradiation position.

[0032] Step S5: Input the real-time X-ray electrical parameters and irradiation geometric position parameters into the optimal irradiation position prediction model, and output the optimal irradiation position of the X-ray machine.

[0033] Among them, the tube voltage tube current Irradiation angle of ultra-high voltage GIS The straight-line distance of the irradiation distance from the insulator As input variables, the mapping output is the optimal illumination position. Construct the optimal illumination position mapping function , Represents the function mapping symbol.

[0034] like Figure 2 As shown, the optimized BP neural network model includes: 1. Optimize the initial weights and thresholds of the BP neural network model based on the improved sparrow search algorithm: The improved sparrow search algorithm improves the initial population generation method and connection weight of the original sparrow search algorithm: Introducing an adaptive mapping to uniformly order the initial population: ; In the formula, Indicates the first The first initial solution is the next initial solution generated after adaptive mapping, which is used to construct a more uniform and ordered initial population and serve as the input sparrow individuals for the sparrow search algorithm. Indicates the first An initial solution; Represents a uniformly distributed random number between 0 and 1; This represents the total number of initial solutions, i.e., the size of the initial population; Improve the connection weighting of the sparrow search algorithm: ; ; In the formula, Indicates the first The connection weight of each iteration is used to control the magnitude of the sparrow's individual position update, and it changes dynamically with the number of iterations; Indicates the current iteration number; Indicates the maximum number of iterations; Represents individual sparrows No. During the nth iteration, at the... The position of the dimension; Represents a random number between 0 and 1; Represents individual sparrows No. During the nth iteration, at the... The position of the dimension; and These all represent system adjustment parameters used to adjust the step size and direction of position updates; they are hyperparameters of the algorithm. Indicates the defined boundary; This represents the foraging range of an individual sparrow, reflecting the size of the foraging area of ​​the current individual sparrow, and is a state parameter of the algorithm; The initial weights and thresholds of the BP neural network model are encoded as individual sparrows in the improved sparrow search algorithm for optimization. When the improved sparrow search algorithm reaches its own loop termination condition (such as reaching the maximum number of iterations or accuracy requirements), the optimal sparrow individual obtained through optimization is decoded, and the weights and thresholds are updated. If the termination condition is not met, the optimization loop continues.

[0035] 2. Bayesian optimization is used to optimize the learning rate and the number of hidden nodes in the BP neural network model: Step 1: Define the target function and the range of values ​​for the variables (learning rate, number of hidden nodes); Step 2: Select a finite number of variables, calculate the corresponding function values, and define them as observed values; Step 3: Based on the observed values, estimate the function using a probabilistic surrogate model (a mathematical model that simulates the objective function) to obtain the objective value; Step 4: Select the next observation point to be calculated according to the uniform sampling rule of variables; Step 5: Repeat steps 2 to 4 until the maximum number of observations is reached, and output the optimal objective function value (i.e., the optimized learning rate and number of hidden nodes).

[0036] The optimal learning rate and number of hidden nodes recommended by Bayesian optimization are updated into the BP neural network model. The current model performance is evaluated to see if it has reached the global optimum or meets the termination condition. If not, the loop continues (the current learning rate, number of hidden nodes, and optimized weights and thresholds are used as the starting point for the next large loop).

[0037] The relevant parameters in this invention are set as follows:

[0038] A system for determining the optimal parameters of an X-ray-excited ultra-high voltage combined electrical appliance includes: The X-ray radiation dose rate measurement module is used to point the X-ray machine's emission port directly at the wall of the ultra-high voltage GIS cavity, and set the X-ray machine tube voltage and tube current to 0 to measure the radiation dose rate in the environment. If the radiation dose rate meets the preset safety conditions, proceed to the next step. The partial discharge intensity calculation module is used by operators to calculate the partial discharge intensity when operating the X-ray machine at a preset safe distance and turning on the ultra-high frequency partial discharge detection device to filter out background noise in the space. The parameter adjustment module is used to gradually increase the X-ray machine tube voltage and tube current from 0 for the current irradiation position, monitor the change amplitude of partial discharge intensity and radiation dose rate in real time, and further adjust the X-ray machine tube voltage and tube current according to the quantitative relationship of the change amplitude after each adjustment until the stop condition is met. It records the maximum value of partial discharge intensity at the irradiation position and the corresponding optimal tube voltage, tube current and irradiation geometric position parameters. The model training module is used to move the X-ray machine and repeat step S3 at each spatial location to obtain the maximum value of the partial discharge intensity at different spatial locations; compare the partial discharge intensity to determine the global optimal irradiation location and use this location as the output label; conduct experiments on different types or specifications of ultra-high voltage GIS to obtain output data under different inputs, collect this output data as historical data and construct a dataset, and train the optimized BP neural network model through the dataset to obtain the optimal irradiation location prediction model; The output module is used to input real-time X-ray electrical parameters and irradiation geometric position parameters into the optimal irradiation position prediction model, and output the optimal irradiation position of the X-ray machine.

[0039] 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 codes, and the processor is used to call the program codes stored in the memory to execute a method for determining the optimal parameters of an X-ray-excited ultra-high voltage combined electrical appliance.

[0040] A non-volatile computer storage medium storing computer-executable instructions that execute a method for determining the optimal parameters of an X-ray-excited ultra-high voltage combined electrical appliance.

[0041] Figure 3 To determine the discharge pulse sequence during GIS defect detection, this invention is used to determine X-ray parameters (distance between the X-ray machine and the ultra-high voltage GIS). After irradiation with X-ray electrical parameters and geometric position parameters, the discharge signal is significantly enhanced, which can improve the detection sensitivity.

[0042] 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 method for determining optimum parameters of an X-ray excited ultrahigh voltage combined electric appliance, characterized in that, Comprising the following steps: Step S1: The emission port of the X-ray machine is directed to the cavity wall of the ultra-high voltage GIS, and the tube voltage and tube current of the X-ray machine are set to 0. The radiation dose rate in the environment is measured. If the radiation dose rate meets the preset safety conditions, the next step is entered; Step S2: The operator operates the X-ray machine outside the preset safety distance, and turns on the ultra-high frequency partial discharge detection device to filter out the background noise in the space. The partial discharge intensity is calculated; Step S3: For the current irradiation position, the tube voltage and tube current of the X-ray machine are gradually increased from 0. The change amplitude of the partial discharge intensity and the radiation dose rate is monitored in real time. After each adjustment, the tube voltage and tube current of the X-ray machine are further adjusted according to the quantitative relationship of the change amplitude until the stopping condition is met. The maximum partial discharge intensity at the irradiation position and the corresponding optimal tube voltage, tube current and irradiation geometric position parameters are recorded; Step S4: Move the X-ray machine. Repeat step S3 at each spatial position to obtain the maximum partial discharge intensity at different spatial positions. Compare the partial discharge intensities to determine the global optimal irradiation position. The position is taken as the output label. Experiments are carried out on different types or specifications of ultra-high voltage GIS to obtain output data under different inputs. The output data is collected as historical data and a data set is constructed. The optimized BP neural network model is trained through the data set to obtain a best irradiation position prediction model; Step S5: The real-time X-ray electrical parameters and irradiation geometric position parameters are input into the best irradiation position prediction model to output the best irradiation position of the X-ray machine.

2. A method for determining optimum parameters of an X-ray excited ultrahigh voltage combined electric appliance according to claim 1, characterized in that: The X-ray machine is a continuous X-ray machine, and the distance between the X-ray machine and the ultra-high voltage GIS According to the voltage level, the ray radiation dose detector is used to detect the ray radiation dose rate in the environment, to determine whether the X-ray machine has a leakage risk, and if the ray radiation dose rate meets the preset safety condition, the next step is entered.

3. The method for determining the optimum parameters of an X-ray excited ultrahigh voltage combined electric appliance according to claim 2, characterized in that: The specific process of step S2 is: the voltage signal collected by the ultra-high frequency partial discharge detection device is filtered to remove the spatial background noise. The occurrence time of partial discharge is located by the energy accumulation function, and the partial discharge data after the occurrence time of partial discharge is recorded. The partial discharge data is subjected to Fourier transform to obtain frequency domain signals. The frequency domain signals are normalized, and the equivalent average frequency and effective frequency range are calculated. The frequency domain signals in the effective frequency range are retained; The retained frequency domain signals are subjected to inverse Fourier transform to obtain effective partial discharge energy signals. The partial discharge intensity is calculated after filtering out the noise.

4. The method for determining the optimal parameters of an X-ray excited ultrahigh voltage combined electric appliance according to claim 3, characterized in that: Step S3 includes: gradually increasing the X-ray machine tube voltage from 0 for the current irradiation position and tube current , indicates the adjustment gear step size, indicates the adjustment step size / iteration number; the dose detector is used in real time to detect the X-ray radiation dose rate at the operator , and a constraint condition is established: , , ; when hour, Indicates the effective partial discharge energy signal. Indicates ambient noise; stop increasing the tube voltage. Record the radiation dose rate at this time. If the value satisfies the constraint condition, then continue to increase the tube voltage and tube current; after each increase in tube voltage and tube current, analyze the partial discharge intensity. and radiation dose rate The changing trend of partial discharge intensity is used to calculate the partial discharge intensity. and radiation dose rate The magnitude of the change; If , then continue to increase the tube voltage and tube current; represents the first adjustment cycle, the change amplitude of the partial discharge intensity; represents the first adjustment cycle, the real-time value of the partial discharge intensity; If , and , only increase the tube voltage, the tube current size is not changed; indicates the first adjustment cycle, the change amplitude of the radiation dose rate; indicates the first adjustment cycle, the real-time value of the radiation dose rate; If , and , then the tube voltage, tube current remain unchanged; If then the tube voltage and tube current remain unchanged; The maximum partial discharge intensity at the irradiation position and the corresponding optimal tube voltage, tube current and irradiation geometric position parameters are recorded.

5. The method for determining the optimum parameters of an X-ray excited ultrahigh voltage combined electric appliance according to claim 4, characterized in that: The specific process of step S4 is: defining the input variables of the optimized BP neural network model as X-ray electrical parameters and irradiation geometric position parameters, and the output variable as the optimal irradiation position, and constructing a mapping relationship; the X-ray electrical parameters include tube voltage and tube current; the irradiation geometric position parameters include the irradiation angle of the ultra-high voltage GIS and the linear distance of the irradiation distance from the insulator; moving the X-ray machine, repeating step S3 at each spatial position to obtain the maximum partial discharge intensity at different spatial positions, determining the global optimal irradiation position by comparing the partial discharge intensities, and taking the position as the output label; experiments are carried out on different types or specifications of ultra-high voltage GIS to obtain output data under different inputs, the output data is collected as historical data and a data set is constructed; the optimized BP neural network model is trained using the data set to obtain a best irradiation position prediction model.

6. The method for determining the optimum parameters of an X-ray excited ultrahigh voltage combined electric appliance according to claim 5, characterized in that: The optimized BP neural network model includes: optimizing the initial weights and thresholds of the BP neural network model based on the improved sparrow search algorithm; and optimizing the learning rate and hidden node number of the BP neural network model using Bayesian optimization. The improved sparrow search algorithm improves the initial population generation method and connection proportion of the original sparrow search algorithm, including introducing adaptive mapping to uniformly order the initial population and improving the connection proportion of the sparrow search algorithm.

7. The method for determining the optimum parameters of an X-ray excited ultrahigh voltage combined electric appliance according to claim 6, characterized in that: The improvement of the connection proportion of the sparrow search algorithm is represented as: ； ； In the formula, Indicates the first The connection ratio in each iteration; Indicates the current iteration number; Indicates the maximum number of iterations; Represents individual sparrows No. During the nth iteration, at the... The position of the dimension; Represents a random number between 0 and 1; Represents individual sparrows No. During the nth iteration, at the... The position of the dimension; and All of these represent system adjustment parameters; Indicates the defined boundary; This indicates the foraging range of an individual sparrow.

8. A system for determining optimum parameters of an X-ray excited ultrahigh voltage combined electric appliance, characterized in that, including: The X-ray radiation dose rate measurement module is used to direct the emission port of the X-ray machine towards the cavity wall of the ultra-high voltage GIS, and set the tube voltage and tube current of the X-ray machine to 0, measure the X-ray radiation dose rate in the environment, and if the X-ray radiation dose rate meets the preset safety condition, proceed to the next step; The partial discharge intensity calculation module is used for the operator to operate the X-ray machine outside the preset safe distance, and turn on the ultra-high frequency partial discharge detection device to filter out the background noise in the space, and calculate the partial discharge intensity; The parameter adjustment module is used for gradually increasing the tube voltage and tube current of the X-ray machine from 0 for the current irradiation position, and real-time monitoring the change amplitude of the partial discharge intensity and the X-ray radiation dose rate, and further adjusting the tube voltage and tube current of the X-ray machine according to the quantization relationship of the change amplitude after each adjustment, until the stop condition is met, and recording the maximum partial discharge intensity, the optimal tube voltage, the tube current and the irradiation geometric position parameters at the irradiation position; The model training module is used to move the X-ray machine, repeat step S3 at each spatial position to obtain the maximum partial discharge intensity at different spatial positions; compare the partial discharge intensities to determine the global optimal irradiation position, and take the position as the output label; experiments are carried out on different types or specifications of ultra-high voltage GIS to obtain output data under different inputs, the output data is collected as historical data and a data set is constructed, and the optimized BP neural network model is trained through the data set to obtain a best irradiation position prediction model; The output module is used to input the real-time X-ray electrical parameters and irradiation geometric position parameters into the best irradiation position prediction model, and output the best irradiation position of the X-ray machine.

9. An electronic device, comprising: The application relates to a computer readable storage medium comprising a processor, a memory and a bus, wherein the processor and the memory are connected through the bus, the memory is used for storing a group of program codes, and the processor is used for calling the program codes stored in the memory to execute the optimal parameter determination method of the X-ray excited ultra-high voltage combined electric appliance.

10. A non-transitory computer storage medium storing computer-executable instructions, the computer-executable instructions comprising instructions for: The computer executable instructions execute the optimal parameter determination method of the X-ray excited ultra-high voltage combined electric appliance according to any one of claims 1-7.

Images