GIS circuit breaker spring energy storage detection method, system, device and medium

Abstract

The invention relates to the technical field of information detection, and discloses a spring energy storage detection method, system and device for a GIS circuit breaker and a medium, and the method comprises the steps: determining a stress concentration coefficient and a material structure change index corresponding to an internal spring of the GIS circuit breaker based on grating reflection wavelengths obtained at a plurality of measurement points on the outer surface of the GIS circuit breaker, the material structure change index represents the grain size change rate; predicting the residual life of the spring based on the stress concentration coefficient and the material structure change index, and determining the predicted residual life of the spring; determining the real-time energy storage density of the spring energy storage system in the GIS circuit breaker under the condition that the predicted residual life is lower than a preset residual life threshold value; and determining a predicted energy storage demand corresponding to the real-time energy storage density, and dynamically adjusting the mechanical-hydraulic energy storage ratio of the spring energy storage system according to the predicted energy storage demand. The invention has the advantages of timely finding out the abnormity of the spring and prolonging the service life.

Description

Technical Field

[0001] The present application relates to the field of information detection technology, and in particular to a spring energy storage detection method, system, equipment and medium for a GIS circuit breaker. Background Art

[0002] The GIS circuit breaker spring energy storage system faces potential technical difficulties in long-term operation. Frequent charge and discharge cycles and continuous mechanical stress will cause the spring material to gradually fail. This process is hidden and gradual, and it is difficult to detect in time with conventional testing. This failure may cause the energy storage function to fail.

[0003] Therefore, it is urgent to develop a real-time, non-destructive detection technology that can dynamically adjust the energy storage ratio, so as to detect spring abnormalities in time and extend its service life. Summary of the invention

[0004] The present application provides a spring energy storage detection method, system, device and medium for a GIS circuit breaker, which solves the technical problem that it is difficult to timely detect the gradual failure of spring materials, and achieves the technical effect of timely detecting spring abnormalities and extending their service life.

[0005] In order to achieve the above objectives, the main technical solutions adopted in this application include:

[0006] In a first aspect, an embodiment of the present application provides a spring energy storage detection method for a GIS circuit breaker, the method comprising:

[0007] Based on the grating reflection wavelengths obtained from multiple measuring points on the outer surface of the GIS circuit breaker, the stress concentration factor and material structure change index corresponding to the internal spring of the GIS circuit breaker are determined, wherein the material structure change index represents the grain size change rate;

[0008] Predicting the remaining life of the spring based on the stress concentration factor and the material structure change index to determine the predicted remaining life of the spring;

[0009] When the predicted remaining life is lower than a preset remaining life threshold, determining a real-time energy storage density of a spring energy storage system in the GIS circuit breaker;

[0010] The predicted energy storage demand corresponding to the real-time energy storage density is determined, and the mechanical-hydraulic energy storage ratio of the spring energy storage system is dynamically adjusted according to the predicted energy storage demand.

[0011] The present embodiment provides a spring energy storage detection method for a GIS circuit breaker, which obtains the stress concentration factor and material structure change index of the internal spring of the GIS circuit breaker by monitoring the change in the grating reflection wavelength. These parameters will be used to predict the remaining life of the spring, where the stress concentration factor reflects the location where cracks may occur, and the material structure change index affects its fatigue resistance and strength. Comprehensive consideration of these factors can improve the accuracy of the remaining life prediction. When the predicted remaining life is lower than the preset threshold, it is necessary to evaluate the state of the spring energy storage system by determining the real-time energy storage density. The real-time energy storage density is defined as the ratio of the total energy storage in the spring energy storage system to the total volume of the hydraulic cylinder. Based on the prediction of the real-time energy storage density, the mechanical-hydraulic energy storage ratio is further adjusted to optimize the energy storage distribution ratio between the spring and the hydraulic cylinder to ensure that the spring energy storage system is in the best operating state.

[0012] Optionally, determining the stress concentration factor and material structure change index corresponding to the internal spring of the GIS circuit breaker based on the grating reflection wavelength obtained from multiple measuring points on the outer surface of the GIS circuit breaker includes:

[0013] According to the grating reflection wavelength, the strain value and the maximum strain value of each measuring point on the outer surface of the GIS circuit breaker are calculated using the Bragg wavelength drift principle;

[0014] Processing the strain value using a spatial interpolation algorithm to generate a continuous strain distribution field;

[0015] Performing gradient calculation on each spatial point in the strain distribution field to obtain a gradient value and a maximum gradient value of each spatial point;

[0016] Determine the spatial point corresponding to the gradient value being greater than a preset gradient threshold as a stress concentration area;

[0017] According to the maximum strain value, the maximum gradient value and the area of ​​the stress concentration region, the stress concentration coefficient and the material structure change index are determined through a preset strain distribution-stress model.

[0018] This embodiment first uses the Bragg wavelength drift principle to calculate the strain value of each measuring point. Subsequently, a spatial interpolation algorithm is used to process these strain values ​​to generate a continuous strain distribution field, thereby effectively inferring the strain distribution of the entire GIS circuit breaker outer surface and providing accurate spatial data. Next, the strain distribution field is gradient calculated to obtain the gradient value of each spatial point. These values ​​reflect the rate of strain change and help determine the stress concentration area in the structure. By setting the gradient threshold, the stress concentration areas in the strain distribution field are identified and marked. These areas may be key stress concentration areas in the structure and require special attention and analysis. Finally, a mapping relationship between the strain distribution pattern and the internal spring stress state is established using a preset strain distribution-stress model, and then the stress concentration coefficient and material structure change index are predicted, which helps to monitor and analyze the performance of the spring in the GIS circuit breaker in real time.

[0019] Optionally, predicting the remaining life of the spring based on the stress concentration factor and the material structure change index to determine the predicted remaining life of the spring includes:

[0020] Acquire an image of the spring, and extract microscopic damage features of the image, wherein the microscopic damage features characterize damage and defects of the spring;

[0021] Classifying the microscopic damage characteristics to determine different damage levels;

[0022] Determining a stress accumulation factor and a fatigue index corresponding to the spring;

[0023] The stress concentration factor, the material structure change index, the damage level, the stress accumulation factor and the fatigue index are input into a preset random forest tree algorithm model to obtain the predicted remaining spring life.

[0024] This embodiment obtains an image of the spring, and then performs image preprocessing and Otsu threshold segmentation to extract microscopic damage features, which can accurately characterize the damage and defects of the spring. The extracted microscopic damage features are then classified to determine different damage levels (low, medium, and high). The stress accumulation factor and fatigue index of the spring in the past cycle are then predicted to obtain the predicted stress accumulation factor and fatigue index. Finally, the stress concentration factor, material structure change index, damage level, stress accumulation factor, and fatigue index are used as input to predict the remaining life of the spring using the random forest algorithm model.

[0025] Optionally, when the predicted remaining life is lower than a preset remaining life threshold, determining the real-time energy storage density of the spring energy storage system in the GIS circuit breaker includes:

[0026] When the predicted remaining life is lower than a preset remaining life threshold, constructing an energy conversion model between the spring and the hydraulic cylinder in the spring energy storage system;

[0027] Calculating energy conversion data in the energy storage process according to the energy conversion model;

[0028] The real-time energy storage density is determined according to the energy conversion data and the total volume of the hydraulic cylinder.

[0029] When the predicted remaining life is lower than the preset remaining life threshold, this embodiment reduces the reliance on the spring by adjusting the energy storage ratio, thereby reducing the stress of the spring and extending its service life. By constructing an energy conversion model between the spring and the hydraulic cylinder, the real-time energy storage density can be obtained according to the real-time energy conversion data, so as to adjust the energy storage ratio in combination with the real-time energy storage density to ensure that the spring energy storage system operates within the optimal performance range and optimize the overall performance of the spring energy storage system.

[0030] Optionally, constructing an energy conversion model between a spring and a hydraulic cylinder in the spring energy storage system includes:

[0031] Obtaining the spring compression displacement, the force applied to the spring, the hydraulic cylinder pressure, and the total volume of the hydraulic cylinder;

[0032] Establishing a force-displacement relationship function of the spring according to the applied force and the compression displacement of the spring;

[0033] Establishing a pressure-volume relationship function of the hydraulic cylinder according to the hydraulic cylinder pressure and the total volume;

[0034] The energy conversion model is obtained by integrating the force-displacement relationship function and the pressure-volume relationship function.

[0035] This embodiment uses the data obtained by the sensor, combined with the mechanical characteristics, to construct a force-displacement relationship function and a pressure-volume relationship function, which can accurately calculate the mechanical energy storage and hydraulic energy storage. Through the comprehensive adjustment of the mechanical energy storage and hydraulic energy storage, a comprehensive energy conversion model is obtained, which can better predict and optimize the energy efficiency and performance of the entire spring energy storage system.

[0036] Optionally, determining the predicted energy storage demand corresponding to the real-time energy storage density, and dynamically adjusting the mechanical-hydraulic energy storage ratio of the spring energy storage system according to the predicted energy storage demand, includes:

[0037] According to the real-time energy storage density, a preset long short-term memory network is used to predict the energy storage demand in the future time period to obtain the predicted energy storage demand corresponding to each time point in the future time period;

[0038] For the predicted energy storage demand corresponding to any point in time, calculate the energy storage change from the current state to the next state;

[0039] With the goal of maximizing energy storage efficiency, dynamically plan all the energy storage changes and determine the optimal path;

[0040] Determining an optimal mechanical-hydraulic energy storage ratio at each time point according to the optimal path;

[0041] The mechanical-hydraulic energy storage ratio of the spring energy storage system is dynamically adjusted according to the optimal mechanical-hydraulic energy storage ratio.

[0042] This embodiment uses a long short-term memory network to predict the energy storage demand at each time point in the future time period, ensuring the accuracy of the prediction results. Then, the optimal path is determined through dynamic planning, thereby optimizing the energy storage change process and maximizing the energy storage efficiency. Then, based on the determined optimal path, the mechanical-hydraulic energy storage ratio at each time point is determined, so as to achieve the best energy storage configuration at different time points and ensure that the performance of the energy storage system at each time point can reach the best. By dynamically adjusting the mechanical-hydraulic energy storage ratio, it is possible to respond to changes in energy storage demand in real time, so that the system always remains efficient and stable in actual operation.

[0043] Optionally, the method further comprises:

[0044] Obtain the reflected wave signal of the spring in the GIS circuit breaker;

[0045] Performing waveform analysis on the reflected wave signal and extracting waveform feature vectors;

[0046] Calculating the Euclidean distance between the waveform feature vector and a standard waveform feature vector in a preset standard feature library;

[0047] When the Euclidean distance is greater than a preset distance threshold, it is determined that microscopic damage exists in the spring.

[0048] This embodiment achieves an accurate description of the internal structure and state of the spring by acquiring the reflected wave signal, performing waveform analysis and feature vector extraction. This method helps to identify the working state and potential problems of the spring. Furthermore, by comparing the waveform feature vector with the data in the standard feature library and calculating the Euclidean distance, it is possible to accurately determine whether the spring has microscopic damage, thereby quickly locating the anomaly and effectively evaluating the health status of the spring.

[0049] Optionally, the waveform feature vector includes amplitude, frequency and phase; the method further includes:

[0050] determining an amplitude decay rate according to a ratio of the amplitude to an initial amplitude;

[0051] Determining a frequency offset according to a difference between the frequency and an initial frequency;

[0052] Determining a phase difference according to a difference between the phase and an initial phase;

[0053] A total health status score corresponding to the spring is determined based on the amplitude decay rate, the frequency offset and the phase difference.

[0054] This embodiment can accurately evaluate the health status of the spring by measuring and calculating the amplitude attenuation rate, frequency offset and phase difference of the reflected wave signal. The health status of the spring is divided into different levels (severe abnormality, moderate abnormality, slight abnormality or normal) based on the scores of various parameters. It can provide a clear health status and help decision makers take corresponding measures according to the actual health status.

[0055] Optionally, the method further comprises:

[0056] Obtain the original strain signal of the spring in the GIS circuit breaker;

[0057] Performing time-frequency analysis on the original strain signal to extract time domain features and frequency domain features;

[0058] Inputting the time domain features and the frequency domain features into a preset strain signal feature-elastic modulus model to obtain an estimated value of the elastic modulus;

[0059] Determining a degree of reduction in the elastic modulus corresponding to the elastic modulus estimate;

[0060] When the reduction degree of the elastic modulus exceeds a preset reduction degree threshold, it is determined that the spring is abnormal.

[0061] This embodiment can effectively characterize the changes and characteristics of the signal by extracting time domain features and frequency domain features from the original strain signal obtained. These features provide important information for the subsequent elastic modulus estimation, which helps to accurately identify the state of the spring. By mapping the time domain and frequency domain features to the elastic modulus estimate of the spring through the strain signal feature-elastic modulus model, the elastic modulus estimate can be accurately obtained. Calculating the degree of reduction in the elastic modulus can quantify the performance changes of the spring and provide a clear standard to evaluate whether the spring is abnormal. By setting a threshold to determine whether the spring is abnormal, automatic monitoring and alarm can be performed. If the degree of reduction in the elastic modulus exceeds a preset threshold, an alarm will be triggered. This mechanism can promptly detect the performance degradation of the spring, thereby preventing potential failures or equipment damage and ensuring the reliability and safety of the equipment.

[0062] In a second aspect, an embodiment of the present application provides a spring energy storage detection system for a GIS circuit breaker, the system comprising:

[0063] A concentration factor and change index determination module, used to determine the stress concentration factor and material structure change index corresponding to the internal spring of the GIS circuit breaker based on the grating reflection wavelength obtained from multiple measurement points on the outer surface of the GIS circuit breaker, wherein the material structure change index represents the grain size change rate;

[0064] A remaining life prediction module, used to predict the remaining life of the spring based on the stress concentration factor and the material structure change index, and determine the predicted remaining life of the spring;

[0065] An energy storage density determination module is used to determine the real-time energy storage density of the spring energy storage system in the GIS circuit breaker when the predicted remaining life is lower than a preset remaining life threshold;

[0066] A dynamic adjustment module is used to determine the predicted energy storage demand corresponding to the real-time energy storage density, and dynamically adjust the mechanical-hydraulic energy storage ratio of the spring energy storage system according to the predicted energy storage demand.

[0067] The present embodiment provides a spring energy storage detection system for a GIS circuit breaker, which obtains the stress concentration factor and material structure change index of the internal spring of the GIS circuit breaker by monitoring the change in the grating reflection wavelength. These parameters will be used to predict the remaining life of the spring, where the stress concentration factor reflects the location where cracks may occur, and the material structure change index affects its fatigue resistance and strength. Comprehensive consideration of these factors can improve the accuracy of the remaining life prediction. When the predicted remaining life is lower than the preset threshold, it is necessary to evaluate the state of the spring energy storage system by determining the real-time energy storage density. The real-time energy storage density is defined as the ratio of the total energy storage in the spring energy storage system to the total volume of the hydraulic cylinder. Based on the prediction of the real-time energy storage density, the mechanical-hydraulic energy storage ratio is further adjusted to optimize the energy storage distribution ratio between the spring and the hydraulic cylinder to ensure that the spring energy storage system is in the best operating state.

[0068] Optionally, the concentration coefficient and change index determination module includes:

[0069] According to the grating reflection wavelength, the strain value and the maximum strain value of each measuring point on the outer surface of the GIS circuit breaker are calculated using the Bragg wavelength drift principle;

[0070] Processing the strain value using a spatial interpolation algorithm to generate a continuous strain distribution field;

[0071] Performing gradient calculation on each spatial point in the strain distribution field to obtain a gradient value and a maximum gradient value of each spatial point;

[0072] Determine the spatial point corresponding to the gradient value being greater than a preset gradient threshold as a stress concentration area;

[0073] According to the maximum strain value, the maximum gradient value and the area of ​​the stress concentration region, the stress concentration coefficient and the material structure change index are determined through a preset strain distribution-stress model.

[0074] This embodiment first uses the Bragg wavelength drift principle to calculate the strain value of each measuring point. Subsequently, a spatial interpolation algorithm is used to process these strain values ​​to generate a continuous strain distribution field, thereby effectively inferring the strain distribution of the entire GIS circuit breaker outer surface and providing accurate spatial data. Next, the strain distribution field is gradient calculated to obtain the gradient value of each spatial point. These values ​​reflect the rate of strain change and help determine the stress concentration area in the structure. By setting the gradient threshold, the stress concentration areas in the strain distribution field are identified and marked. These areas may be key stress concentration areas in the structure and require special attention and analysis. Finally, a mapping relationship between the strain distribution pattern and the internal spring stress state is established using a preset strain distribution-stress model, and then the stress concentration coefficient and material structure change index are predicted, which helps to monitor and analyze the performance of the spring in the GIS circuit breaker in real time.

[0075] Optionally, the remaining life prediction module includes:

[0076] Acquire an image of the spring, and extract microscopic damage features of the image, wherein the microscopic damage features characterize damage and defects of the spring;

[0077] Classifying the microscopic damage characteristics to determine different damage levels;

[0078] Determining a stress accumulation factor and a fatigue index corresponding to the spring;

[0079] The stress concentration factor, the material structure change index, the damage level, the stress accumulation factor and the fatigue index are input into a preset random forest tree algorithm model to obtain the predicted remaining spring life.

[0080] This embodiment obtains an image of the spring, and then performs image preprocessing and Otsu threshold segmentation to extract microscopic damage features, which can accurately characterize the damage and defects of the spring. The extracted microscopic damage features are then classified to determine different damage levels (low, medium, and high). The stress accumulation factor and fatigue index of the spring in the past cycle are then predicted to obtain the predicted stress accumulation factor and fatigue index. Finally, the stress concentration factor, material structure change index, damage level, stress accumulation factor, and fatigue index are used as input to predict the remaining life of the spring using the random forest algorithm model.

[0081] Optionally, the module for determining energy storage density includes:

[0082] When the predicted remaining life is lower than a preset remaining life threshold, constructing an energy conversion model between the spring and the hydraulic cylinder in the spring energy storage system;

[0083] Calculating energy conversion data in the energy storage process according to the energy conversion model;

[0084] The real-time energy storage density is determined according to the energy conversion data and the total volume of the hydraulic cylinder.

[0085] When the predicted remaining life is lower than the preset remaining life threshold, this embodiment reduces the reliance on the spring by adjusting the energy storage ratio, thereby reducing the stress of the spring and extending its service life. By constructing an energy conversion model between the spring and the hydraulic cylinder, the real-time energy storage density can be obtained according to the real-time energy conversion data, so as to adjust the energy storage ratio in combination with the real-time energy storage density to ensure that the spring energy storage system operates within the optimal performance range and optimize the overall performance of the spring energy storage system.

[0086] Optionally, constructing an energy conversion model between a spring and a hydraulic cylinder in the spring energy storage system includes:

[0087] Obtaining the spring compression displacement, the force applied to the spring, the hydraulic cylinder pressure, and the total volume of the hydraulic cylinder;

[0088] Establishing a force-displacement relationship function of the spring according to the applied force and the compression displacement of the spring;

[0089] Establishing a pressure-volume relationship function of the hydraulic cylinder according to the hydraulic cylinder pressure and the total volume;

[0090] The energy conversion model is obtained by integrating the force-displacement relationship function and the pressure-volume relationship function.

[0091] This embodiment uses the data obtained by the sensor, combined with the mechanical characteristics, to construct a force-displacement relationship function and a pressure-volume relationship function, which can accurately calculate the mechanical energy storage and hydraulic energy storage. Through the comprehensive adjustment of the mechanical energy storage and hydraulic energy storage, a comprehensive energy conversion model is obtained, which can better predict and optimize the energy efficiency and performance of the entire spring energy storage system.

[0092] Optionally, the dynamic adjustment module includes:

[0093] According to the real-time energy storage density, a preset long short-term memory network is used to predict the energy storage demand in the future time period to obtain the predicted energy storage demand corresponding to each time point in the future time period;

[0094] For the predicted energy storage demand corresponding to any point in time, calculate the energy storage change from the current state to the next state;

[0095] With the goal of maximizing energy storage efficiency, dynamically plan all the energy storage changes and determine the optimal path;

[0096] Determining an optimal mechanical-hydraulic energy storage ratio at each time point according to the optimal path;

[0097] The mechanical-hydraulic energy storage ratio of the spring energy storage system is dynamically adjusted according to the optimal mechanical-hydraulic energy storage ratio.

[0098] This embodiment uses a long short-term memory network to predict the energy storage demand at each time point in the future time period, ensuring the accuracy of the prediction results. Then, the optimal path is determined through dynamic planning, thereby optimizing the energy storage change process and maximizing the energy storage efficiency. Then, based on the determined optimal path, the mechanical-hydraulic energy storage ratio at each time point is determined, so as to achieve the best energy storage configuration at different time points and ensure that the performance of the energy storage system at each time point can reach the best. By dynamically adjusting the mechanical-hydraulic energy storage ratio, it is possible to respond to changes in energy storage demand in real time, so that the system always remains efficient and stable in actual operation.

[0099] Optionally, the system further comprises:

[0100] Obtain the reflected wave signal of the spring in the GIS circuit breaker;

[0101] Performing waveform analysis on the reflected wave signal and extracting waveform feature vectors;

[0102] Calculating the Euclidean distance between the waveform feature vector and a standard waveform feature vector in a preset standard feature library;

[0103] When the Euclidean distance is greater than a preset distance threshold, it is determined that microscopic damage exists in the spring.

[0104] This embodiment achieves an accurate description of the internal structure and state of the spring by acquiring the reflected wave signal, performing waveform analysis and feature vector extraction. This method helps to identify the working state and potential problems of the spring. Furthermore, by comparing the waveform feature vector with the data in the standard feature library and calculating the Euclidean distance, it is possible to accurately determine whether the spring has microscopic damage, thereby quickly locating the anomaly and effectively evaluating the health status of the spring.

[0105] Optionally, the waveform feature vector includes amplitude, frequency and phase; and the system further includes:

[0106] determining an amplitude decay rate according to a ratio of the amplitude to an initial amplitude;

[0107] Determining a frequency offset according to a difference between the frequency and an initial frequency;

[0108] Determining a phase difference according to a difference between the phase and an initial phase;

[0109] A total health status score corresponding to the spring is determined based on the amplitude decay rate, the frequency offset and the phase difference.

[0110] This embodiment can accurately evaluate the health status of the spring by measuring and calculating the amplitude attenuation rate, frequency offset and phase difference of the reflected wave signal. The health status of the spring is divided into different levels (severe abnormality, moderate abnormality, slight abnormality or normal) based on the scores of various parameters. It can provide a clear health status and help decision makers take corresponding measures according to the actual health status.

[0111] Optionally, the system further comprises:

[0112] Obtain the original strain signal of the spring in the GIS circuit breaker;

[0113] Performing time-frequency analysis on the original strain signal to extract time domain features and frequency domain features;

[0114] Inputting the time domain features and the frequency domain features into a preset strain signal feature-elastic modulus model to obtain an estimated value of the elastic modulus;

[0115] Determining a degree of reduction in the elastic modulus corresponding to the elastic modulus estimate;

[0116] When the reduction degree of the elastic modulus exceeds a preset reduction degree threshold, it is determined that the spring is abnormal.

[0117] This embodiment can effectively characterize the changes and characteristics of the signal by extracting time domain features and frequency domain features from the original strain signal obtained. These features provide important information for the subsequent elastic modulus estimation, which helps to accurately identify the state of the spring. By mapping the time domain and frequency domain features to the elastic modulus estimate of the spring through the strain signal feature-elastic modulus model, the elastic modulus estimate can be accurately obtained. Calculating the degree of reduction in the elastic modulus can quantify the performance changes of the spring and provide a clear standard to evaluate whether the spring is abnormal. By setting a threshold to determine whether the spring is abnormal, automatic monitoring and alarm can be performed. If the degree of reduction in the elastic modulus exceeds a preset threshold, an alarm will be triggered. This mechanism can promptly detect the performance degradation of the spring, thereby preventing potential failures or equipment damage and ensuring the reliability and safety of the equipment.

[0118] In a third aspect, an embodiment of the present application provides a computer device, including:

[0119] A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the above-mentioned spring energy storage detection method for the GIS circuit breaker by executing the computer instructions.

[0120] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, on which computer instructions are stored, and the computer instructions are used to enable a computer to execute the spring energy storage detection method for the GIS circuit breaker described above. BRIEF DESCRIPTION OF THE DRAWINGS

[0121] In order to more clearly illustrate the specific implementation methods of the present application or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0122] Figure 1 A flow chart of a spring energy storage detection method for a GIS circuit breaker provided in an embodiment of the present application;

[0123] Figure 2 A flowchart of step S1 provided in an embodiment of the present application;

[0124] Figure 3 A flowchart of step S3 provided in an embodiment of the present application;

[0125] Figure 4 A flowchart of step S5 provided in an embodiment of the present application;

[0126] Figure 5 A flowchart of step S51 provided in an embodiment of the present application;

[0127] Figure 6 A flowchart of step S7 provided in an embodiment of the present application;

[0128] Figure 7 A flow chart for determining whether a spring has microscopic damage provided in an embodiment of the present application;

[0129] Figure 8 A flow chart for determining the total health status score corresponding to a spring provided in an embodiment of the present application;

[0130] Fig. 9 A flow chart for determining whether a spring is abnormal provided in an embodiment of the present application;

[0131] Fig.10 A flowchart for determining the total cumulative damage amount provided in an embodiment of the present application;

[0132] Fig.11 A flowchart of step S407 provided in an embodiment of the present application;

[0133] Fig.12 A block diagram of a spring energy storage detection system for a GIS circuit breaker provided in an embodiment of the present application;

[0134] Fig.13 A schematic diagram of the structure of a computer device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0135] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present application.

[0136] According to an embodiment of the present application, an embodiment of a spring energy storage detection method for a GIS circuit breaker is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0137] In this embodiment, a spring energy storage detection method for a GIS circuit breaker is provided. Figure 1 A flow chart of a spring energy storage detection method for a GIS circuit breaker provided in an embodiment of the present application is as follows: Figure 1 As shown, the process includes the following steps:

[0138] Step S1, based on the grating reflection wavelength obtained from multiple measuring points on the outer surface of the GIS circuit breaker, determine the stress concentration factor and material structure change index corresponding to the internal spring of the GIS circuit breaker, wherein the material structure change index represents the grain size change rate.

[0139] The stress concentration factor reflects the degree of stress concentration caused by defects (such as holes, scratches, etc.) or uneven shapes in the material. In contrast, the material structure change index describes the rate of change of grain size, which is usually caused by deformation of the material. For non-intrusive monitoring of GIS circuit breakers, the grating reflection wavelength can be used for analysis to obtain information from these two data.

[0140] Specifically, a fiber grating sensor can be made of single-mode optical fiber, with a grating length of 10 mm, a reflectivity of 80%, and a central wavelength of 1550 nm. One fiber grating sensor is arranged every 30 degrees around the GIS circuit breaker casing. One is arranged every 50 mm along the axial direction to monitor the deformation at different positions, forming a complete 360-degree grid structure, with a total of 48 measuring points. The grating reflection wavelength is measured at multiple measuring points, the sampling frequency is set to 100 Hz, and the wavelength resolution is 1pm, so as to capture the dynamic changes caused by the operation of the GIS circuit breaker. Then, the collected grating reflection wavelength is analyzed to determine the stress concentration factor and material structure change index corresponding to the internal spring of the GIS circuit breaker. The intensity of the stress concentration area and the trend and degree of the grain size change inside the material can be inferred by the offset of the grating reflection wavelength or the change of the spectrum, so as to evaluate the force condition and stress distribution of the spring in the GIS circuit breaker. Machine learning algorithms (such as regression analysis, support vector machines, deep learning models, etc.) can also be used to establish a prediction model, and the corresponding stress concentration factor and material structure change index of the spring can be predicted based on the spectral characteristic information of the grating reflection wavelength. During model training, the existing labeled data set can be used for supervised learning to optimize the model parameters to improve the prediction accuracy.

[0141] Step S3, predicting the remaining life of the spring based on the stress concentration factor and the material structure change index, and determining the predicted remaining life of the spring.

[0142] Specifically, since the spring may experience stress concentration during use, that is, some parts are subjected to much higher stress than other parts. These stress concentration points are often the first places where fatigue cracks appear in the spring. A higher stress concentration factor means that these points are more prone to cracks and damage, thereby shortening the service life of the spring. With the change of time and usage conditions, the material structure of the spring may change. These changes will cause the material structure change index of the spring to change, thereby affecting its fatigue resistance and strength. Therefore, it is necessary to comprehensively consider the stress concentration factor and the material structure change index, and their mutual influence jointly determines the remaining life of the spring under given working conditions.

[0143] Based on this, a remaining life prediction model can be established using machine learning or statistical methods, such as regression analysis, survival analysis (such as the COX model), deep learning model, random forest tree algorithm model, etc., according to the stress concentration factor and material structure change index. Use historical data sets to train and optimize the model to ensure that the model can accurately predict the remaining life. Adjust the parameters of the model to improve the accuracy and generalization of the prediction. Verify the accuracy and reliability of the model by comparing the remaining life predicted by the model with the actual time of spring failure. The predicted remaining life here is expressed as a percentage of life, ranging from 0-100%, where the new state of the spring is defined as 100% life and the failure state is defined as 0% life.

[0144] Step S5, when the predicted remaining life is lower than a preset remaining life threshold, determining the real-time energy storage density of the spring energy storage system in the GIS circuit breaker.

[0145] Since the predicted remaining life is lower than a warning value of a preset remaining life threshold, it means that the spring may face the risk of performance degradation or even breakage due to aging. In order to extend the service life of the spring and ensure the safety of the GIS circuit breaker, it is necessary to adjust the force of the spring in the spring energy storage system in real time. Before adjusting, it is crucial to determine the real-time energy storage density of the spring energy storage system. Preferably, the preset remaining life threshold is 80%.

[0146] Specifically, the spring energy storage system consists of a spring and a hydraulic cylinder. First, the force-displacement and pressure-volume energy conversion models can be established by analyzing the mechanical properties of the spring and the relevant parameters of the hydraulic cylinder. The force-displacement relationship of the spring is based on Hooke's law, which describes the functional relationship between force and displacement; while the pressure-volume relationship of the hydraulic cylinder describes the change in the volume of the hydraulic cylinder under different pressures.

[0147] During the energy storage process, the real-time energy storage density is defined as the ratio of the total energy storage in the spring energy storage system to the total volume of the hydraulic cylinder. In order to calculate the energy data during the energy storage process, the energy conversion model can be numerically integrated using an integration method, such as the Simpson integration method. By integrating the energy conversion model of the spring and the hydraulic cylinder, and based on the total volume of the hydraulic cylinder, the accurate value of the real-time energy storage density can be obtained.

[0148] Step S7, determining the predicted energy storage demand corresponding to the real-time energy storage density, and dynamically adjusting the mechanical-hydraulic energy storage ratio of the spring energy storage system according to the predicted energy storage demand.

[0149] Specifically, the real-time energy storage density refers to the ratio of the real-time energy storage density to the total volume of the hydraulic cylinder in the spring energy storage system. First, by analyzing the real-time energy storage density and combining it with the current operating status, a calculation method for predicting energy storage demand can be derived. For example, the historical energy storage density and operating status information of the GIS circuit breaker, such as the mechanical spring compression, hydraulic cylinder pressure, and ambient temperature, can be collected as input data for the prediction model. Subsequently, the wavelet denoising algorithm is used to process the historical energy storage density and operating status information to eliminate noise in the data and retain key features. The denoised data is further Z-score standardized to ensure that all features have the same numerical range, thereby avoiding the influence of model bias caused by data scale differences. Finally, the long short-term memory network (LSTM) is applied to perform time series analysis and prediction on the standardized data. By training the LSTM model to learn the patterns of historical data and current status, future energy storage demand can be predicted. The prediction results can be adjusted and optimized according to the real-time updated current status information to reflect the changing conditions in the system. Finally, the real-time energy storage density is input into the trained long short-term memory network to obtain the predicted energy storage demand.

[0150] Next, the mechanical-hydraulic energy storage ratio is the ratio of the energy storage distribution of the mechanical spring and the hydraulic cylinder in the spring energy storage system. According to the analysis of the real-time energy storage density and the current operating status of the system, the compression of the mechanical spring and the working pressure of the hydraulic cylinder can be dynamically adjusted to optimize the energy storage performance of the system. Specifically, the dynamic programming algorithm can be used to calculate the optimal mechanical-hydraulic energy storage ratio and obtain the optimal energy storage distribution plan. Once the optimal energy storage distribution plan is determined, the spring energy storage system can be adaptively controlled, that is, adjusted according to the real-time energy storage density and operating status to maintain the best working state.

[0151] The present embodiment provides a spring energy storage detection method for a GIS circuit breaker, which obtains the stress concentration factor and material structure change index of the internal spring of the GIS circuit breaker by monitoring the change in the grating reflection wavelength. These parameters will be used to predict the remaining life of the spring, where the stress concentration factor reflects the location where cracks may occur, and the material structure change index affects its fatigue resistance and strength. Comprehensive consideration of these factors can improve the accuracy of the remaining life prediction. When the predicted remaining life is lower than the preset threshold, it is necessary to evaluate the state of the spring energy storage system by determining the real-time energy storage density. The real-time energy storage density is defined as the ratio of the total energy storage in the spring energy storage system to the total volume of the hydraulic cylinder. Based on the prediction of the real-time energy storage density, the mechanical-hydraulic energy storage ratio is further adjusted to optimize the energy storage distribution ratio between the spring and the hydraulic cylinder to ensure that the spring energy storage system is in the best operating state.

[0152] Figure 2 The flowchart of step S1 provided in the embodiment of the present application may include the following steps:

[0153] Step S11, according to the grating reflection wavelength, using the Bragg wavelength drift principle to calculate the strain value and the maximum strain value of each measuring point on the outer surface of the GIS circuit breaker.

[0154] Specifically, the measured grating reflection wavelength is compared with the original Bragg wavelength of the grating, the wavelength change is calculated, and the strain value of each measuring point of the shell is calculated using the Bragg wavelength drift principle. After all the strain values ​​are obtained, the maximum strain value is found.

[0155]

[0156] Among them, the strain sensitivity coefficient is 1.2pm / με and the linear range is ±4000με.

[0157] For example, first, measure and record the original Bragg wavelength of the fiber Bragg grating in the strain-free state. During the operation of the GIS circuit breaker, use the fiber Bragg grating sensor to record the grating reflection wavelength of each measuring point. For each measuring point, calculate the difference between the grating reflection wavelength and the original Bragg wavelength, that is, the wavelength change. Then use the strain sensitivity coefficient 1.2pm / με to convert the wavelength change into a strain value. Note that it is also necessary to ensure that the strain value is within the linear range of the fiber Bragg grating sensor ±4000με. Repeat the above calculations for all measuring points to obtain the strain value of each measuring point on the outer surface of the GIS circuit breaker. Find the maximum value from the strain values ​​of all measuring points, which represents the maximum strain value recorded during the measurement process.

[0158] Step S13: Process the strain value using a spatial interpolation algorithm to generate a continuous strain distribution field.

[0159] Specifically, the spatial interpolation algorithm can use Kriging interpolation. The strain values ​​are processed using the spatial interpolation algorithm. Before processing, the range is first set to 200 mm, representing the spatial scale of the correlation between strain values. The nugget effect is set to 0.1 to affect the smoothness of the interpolation result. According to the strain value, the semivariance function is calculated; the spatial covariance matrix is ​​constructed using the semivariance function; the interpolation algorithm is executed using the covariance matrix and the strain value to generate the interpolation result for each spatial position; the interpolation results are integrated to generate the continuous strain distribution field of the outer surface of the GIS circuit breaker.

[0160] Step S15, performing gradient calculation on each spatial point in the strain distribution field to obtain the gradient value and the maximum gradient value of each spatial point.

[0161] Specifically, the gradient is a vector field whose direction points to the direction where the strain value increases fastest, and its magnitude indicates the rate at which the strain value changes. Each component of the gradient vector is the partial derivative of the strain value with respect to each coordinate axis: Gradient value:

[0162] Traverse each spatial point in the strain distribution field, calculate the gradient value, and find the maximum gradient value from all calculated gradient values.

[0163] Step S17, determining the spatial point corresponding to the gradient value being greater than a preset gradient threshold as a stress concentration area.

[0164] Specifically, all spatial points can be traversed to check whether the gradient value is greater than a preset gradient threshold. If it is greater than the preset gradient threshold, the spatial point is marked as part of the strain concentration area. The preset gradient threshold here is preferably 0.01 με / mm.

[0165] Step S19, determining the stress concentration coefficient and the material structure change index according to the maximum strain value, the maximum gradient value and the area of ​​the stress concentration region through a preset strain distribution-stress model.

[0166] Specifically, a preset strain distribution-stress model is used to determine the stress concentration factor and the material structure change index. The support vector regression (SVR) algorithm can be selected to establish the mapping relationship between the strain distribution pattern and the internal spring stress state. The radial basis function (RBF) is selected as the kernel function in the SVR algorithm, and the five-fold cross validation method is used to optimize the parameters C and γ of the SVR model. Parameter C controls the penalty of the error term, while γ determines the width of the radial basis kernel. The SVR model is trained using historical data, with the maximum strain, the maximum gradient, and the area of ​​the stress concentration region as input, and the stress concentration factor and the material structure change index as output. Five-fold cross validation is used to evaluate the performance of the model, and the model parameters are adjusted until satisfactory accuracy is achieved. The stress concentration factor and the material structure change index output by the model are analyzed and compared with the actual measured values ​​to verify the prediction ability of the model, and finally a trained strain distribution-stress model is obtained. The strain distribution-stress model can be used to predict the stress concentration factor and the material structure change index.

[0167] The normal value of the stress concentration coefficient here is <2.5, and the normal value of the material structure change index is <0.2. These two data can also be used to determine whether there is an abnormality. If there is an abnormality, an alarm can be sounded to remind the staff to perform maintenance and other operations.

[0168] and Figure 1Compared with the embodiment shown in the figure, this embodiment first uses the Bragg wavelength drift principle to calculate the strain value of each measuring point. Subsequently, a spatial interpolation algorithm is used to process these strain values ​​to generate a continuous strain distribution field, thereby effectively inferring the strain distribution of the entire GIS circuit breaker outer surface and providing accurate spatial data. Next, the strain distribution field is gradient calculated to obtain the gradient value of each spatial point. These values ​​reflect the rate of strain change and help determine the stress concentration area in the structure. By setting the gradient threshold, the stress concentration areas in the strain distribution field are identified and marked. These areas may be key stress concentration areas in the structure and require special attention and analysis. Finally, a preset strain distribution-stress model is used to establish a mapping relationship between the strain distribution pattern and the internal spring stress state, thereby predicting the stress concentration coefficient and material structure change index, which helps to monitor and analyze the performance of the spring in the GIS circuit breaker in real time.

[0169] Figure 3 The flowchart of step S3 provided in the embodiment of the present application may include the following steps:

[0170] Step S31, acquiring an image of the spring and extracting microscopic damage features of the image, wherein the microscopic damage features characterize the damage and defects of the spring.

[0171] Specifically, a scanning electron microscope (SEM) or a transmission electron microscope (TEM) can be used to take a comprehensive picture of the spring with a magnification of 5000 times. Ensure that the image clarity is sufficient to reveal the microscopic damage characteristics. The collected images are preprocessed, including adjusting the brightness, contrast, denoising, etc. to improve the image quality. The image is processed using the Otsu threshold segmentation algorithm to separate the damage features in the image from the background. Microscopic damage features are extracted from the processed image, and the microscopic damage features can include crack density (the number of cracks per square millimeter), defect size distribution (statistical distribution of defect diameters), and dislocation density (dislocation line length per square micrometer).

[0172] Step S33, classifying the microscopic damage features to determine different damage levels.

[0173] Specifically, support vector machine is selected as the classification model to classify microscopic damage features, and radial basis function (RBF) kernel function is used. The penalty parameter C (10 -2 -10 2 ) and the kernel parameter γ(10 -4-1) and use the grid search method to traverse all possible combinations of C and γ to find the best parameter pair. A 5-fold cross validation is used to evaluate the model performance under different parameter combinations to ensure the generalization ability of the model. The SVM model is trained using historical data, with microscopic damage features as input and damage level as output. The trained SVM model is used to classify the new microscopic damage feature data and determine its damage level, which is divided into three levels: low, medium, and high.

[0174] Step S35, determining the stress accumulation factor and fatigue index corresponding to the spring.

[0175] Specifically, first, the model is composed of 3 convolutional layers and 2 fully connected layers. Each convolutional layer has 64 3×3 convolution kernels followed by a ReLU activation function. Each fully connected layer has 128 neurons and also uses the ReLU activation function. The model uses the Adam optimizer, and the initial learning rate is set to 0.001. The input data is the time series of the stress accumulation factor and fatigue index of the spring in the past 100 cycles, where the stress accumulation factor is normalized to the range of 0-1 and the fatigue index is normalized to the range of 0-100. The output of the model is the predicted stress accumulation factor and fatigue index for the next 10 cycles. The training process is 1000 rounds, and the entire data set is traversed once in each round. After the training is completed, the model is applied to the actual spring operation process, and the stress accumulation factor and fatigue index data of the past 100 cycles are collected in real time through sensors and data acquisition systems, and predictions are made to predict the stress accumulation factor and fatigue index of the spring in the next 10 cycles.

[0176] Step S37, inputting the stress concentration factor, the material structure change index, the damage level, the stress accumulation factor and the fatigue index into a preset random forest tree algorithm model to obtain the predicted remaining spring life.

[0177] Specifically, the random forest algorithm is used in the prediction process of the remaining life of the spring. First, the Gini index is used as a criterion for feature selection to help identify the features that contribute most to the prediction. The importance of each feature is evaluated through the Out-of-Bag (OOB) error, and it is found that the damage level is the most important feature, with a contribution of 40%. It is followed by the stress accumulation factor, which accounts for 30%, and the fatigue index, which contributes 20%. In addition, the stress concentration factor and the material structure change index also have an impact on the prediction, each accounting for 5%.

[0178] After determining the importance of the features, the key parameters of the random forest algorithm model are set: 100 trees are selected and the maximum depth of each tree is set to 10. Such parameter settings are intended to balance the complexity and generalization ability of the model. In order to evaluate the prediction accuracy of the model, the validation set and cross-validation method are used to ensure the reliability of its prediction results, and the prediction error is within ±500 working cycles. Finally, the trained and validated random forest algorithm model is used to predict the remaining life of the spring.

[0179] and Figure 1 Compared with the embodiment shown in the figure, this embodiment obtains an image of the spring, and then performs image preprocessing and Otsu threshold segmentation to extract microscopic damage features, which can accurately characterize the damage and defects of the spring. The extracted microscopic damage features are then classified to determine different damage levels (low, medium, and high). The stress accumulation factor and fatigue index of the spring in the past cycle are then predicted to obtain the predicted stress accumulation factor and fatigue index. Finally, the stress concentration factor, material structure change index, damage level, stress accumulation factor, and fatigue index are used as inputs to predict the remaining life of the spring using the random forest algorithm model.

[0180] Figure 4 The flowchart of step S5 provided in the embodiment of the present application may include the following steps:

[0181] Step S51, when the predicted remaining life is lower than a preset remaining life threshold, constructing an energy conversion model between the spring and the hydraulic cylinder in the spring energy storage system.

[0182] Specifically, the predicted remaining life is lower than the preset remaining life threshold, which means that the spring may face the risk of performance degradation or even breakage over time. In order to extend the service life of the spring and ensure the safety of the GIS circuit breaker, it is necessary to adjust the force of the spring in the spring energy storage system in real time. Therefore, it is necessary to build an energy conversion model to determine the energy conversion data.

[0183] First, a parallel structure of mechanical spring and hydraulic cylinder is adopted to ensure that the two can work at the same time and jointly provide the required energy. The maximum compression of the mechanical spring is 50mm, and the maximum stroke of the hydraulic cylinder is 100mm. The working pressure of the hydraulic cylinder is accurately adjusted by the electro-hydraulic proportional valve 0-20MPa to change the output of the hydraulic cylinder, thereby adjusting the energy storage ratio of the mechanical spring and the hydraulic cylinder. In addition, high-precision displacement sensors and pressure sensors are installed on the mechanical spring and hydraulic cylinder of the spring energy storage system, and a data acquisition module (DAQ) is used to collect sensor data in real time. The sampling frequency is set to 1kHz (i.e., 1000 samples per second) and the resolution of the displacement sensor is 0.01mm, which ensures the accurate measurement of the spring compression displacement. The accuracy of the pressure sensor is 0.1% FS (Full Scale), which ensures the accuracy of the hydraulic cylinder pressure measurement. The 32-point sliding window average filter algorithm is used to reduce the noise and fluctuations in the data. Setting the window overlap rate to 50% means that each window overlaps the previous window by 16 data points (half of 32 points), which helps maintain the continuity and stability of the filtered data.

[0184] Then, according to the characteristics of the mechanical spring and the mechanical properties of the hydraulic cylinder, the force-displacement relationship function of the spring and the pressure-volume relationship function of the hydraulic cylinder are established respectively. After considering the leakage loss and friction loss, the two relationship functions are combined to obtain the total energy conversion model, which can be obtained by adding or weighted summing the force-displacement relationship function of the spring and the pressure-volume relationship function of the hydraulic cylinder. This energy conversion model is helpful for the subsequent calculation of energy conversion data and helps to accurately control the working state of the spring and the hydraulic cylinder to ensure that the spring energy storage system operates within the optimal performance range.

[0185] Step S53, calculating energy conversion data during the energy storage process according to the energy conversion model.

[0186] Specifically, the spring compression displacement and the pressure of the hydraulic cylinder are obtained in real time through the displacement sensor and the pressure sensor, and then the energy conversion model is used to calculate the real-time energy conversion data. The energy conversion model can be numerically integrated by an integration method, such as the Simpson integration method, to obtain the energy conversion data. The integration step in the Simpson integration method is 0.1 mm (displacement) and 0.1 MPa (pressure).

[0187] Step S55, determining the real-time energy storage density according to the energy conversion data and the total volume of the hydraulic cylinder.

[0188] Specifically, the real-time energy storage density can be calculated according to the following formula:

[0189]

[0190] Where ρ is the real-time energy storage density, E total,adjusted is the energy conversion data, and V is the total volume of the hydraulic cylinder.

[0191] In addition, the Kalman filter algorithm is used to smooth the real-time energy storage density data, where the process noise covariance is set to 0.01 and the measurement noise covariance is set to 0.1. Through Kalman filtering, random fluctuations in the data can be reduced and smoother real-time energy storage density data can be obtained.

[0192] and Figure 1 Compared with the embodiment shown in the figure, this embodiment reduces the reliance on the spring by adjusting the energy storage ratio when the predicted remaining life is lower than the preset remaining life threshold, thereby reducing the stress of the spring and extending its service life. By constructing an energy conversion model between the spring and the hydraulic cylinder, the real-time energy storage density can be obtained according to the real-time energy conversion data, so as to adjust the energy storage ratio in combination with the real-time energy storage density to ensure that the spring energy storage system operates within the optimal performance range and optimize the overall performance of the spring energy storage system.

[0193] Figure 5 The flowchart of step S51 provided in the embodiment of the present application may include the following steps:

[0194] Step S511, obtaining the spring compression displacement, the force acting on the spring, the hydraulic cylinder pressure and the total volume of the hydraulic cylinder.

[0195] Specifically, the spring compression displacement, the force acting on the spring, and the hydraulic cylinder pressure are obtained through sensors, and the total volume of the hydraulic cylinder is obtained through the product manual provided by the supplier for subsequent calculation of the energy storage density.

[0196] Step S513: establishing a force-displacement relationship function of the spring according to the applied force and the spring compression displacement.

[0197] Specifically, mechanical energy storage can be expressed as the change in spring potential energy, introducing the mechanical friction loss rate α (5-10%), that is, the force-displacement relationship function:

[0198] F(x)=k·x

[0199]

[0200] E m,adjusted =E m ×(1-α)

[0201] Where F(x) is the applied force; x is the spring compression displacement; k is the spring stiffness, k = 100 N / mm; E m,adjusted Mechanical energy storage taking into account friction losses.

[0202] Step S515: establishing a pressure-volume relationship function of the hydraulic cylinder according to the hydraulic cylinder pressure and the total volume.

[0203] Specifically, hydraulic energy storage can be expressed as the change in pressure work of the hydraulic system, introducing the leakage loss rate β (3-8%), that is, the pressure-volume relationship function:

[0204] P=F' / A

[0205]

[0206] E h,adjusted =E h ×(1-β)

[0207] Where P is the hydraulic cylinder pressure; F' is the force acting on the piston; A is the area of ​​the piston, A = 50cm 2 ; V is the total volume; E h,adjusted Hydraulic energy storage taking leakage loss into account.

[0208] Step S517, combining the force-displacement relationship function and the pressure-volume relationship function to obtain the energy conversion model.

[0209] Specifically, the energy conversion model: E total,adjusted =E m,adjusted +E h,adjusted , where E total,adjusted Energy conversion data.

[0210] and Figure 4 Compared with the embodiment shown in the figure, this embodiment uses the data obtained by the sensor, combined with the mechanical characteristics, to construct the force-displacement relationship function and the pressure-volume relationship function, so as to accurately calculate the mechanical energy storage and the hydraulic energy storage. Through the comprehensive adjustment of the mechanical energy storage and the hydraulic energy storage, a comprehensive energy conversion model is obtained, which can better predict and optimize the energy efficiency and performance of the entire spring energy storage system.

[0211] Figure 6 The flowchart of step S7 provided in the embodiment of the present application may include the following steps:

[0212] Step S71, based on the real-time energy storage density, use a preset long short-term memory network to predict the energy storage demand in the future time period, and obtain the predicted energy storage demand corresponding to each time point in the future time period.

[0213] Specifically, a three-layer LSTM network is designed, each containing 64 neurons, to process time series data. The input features include energy storage density, spring compression displacement, hydraulic cylinder pressure, and ambient temperature, and the output is the energy storage demand forecast for the next 24 hours. First, the historical energy storage density and current state information, such as spring compression displacement, hydraulic cylinder pressure, and ambient temperature, are collected. The db4 wavelet 5-layer decomposition is applied to the data for denoising to reduce the influence of environmental noise and measurement error. Then, Z-score standardization is used to make the data have zero mean and unit standard deviation to improve the model training effect. The historical data is divided into 24-hour windows with a sliding step of 1 hour for time series analysis. The LSTM network is used for in-depth analysis to capture the dynamic features and long-term dependencies in the data. The Adam optimizer (learning rate 0.001) is used for 1000 rounds of training, and the validation set loss drops below 0.05. After training, the LSTM network can be used to predict the energy storage demand in the future time period.

[0214] Step S73, for the predicted energy storage demand corresponding to any time point, calculate the energy storage change from the current state to the next state.

[0215] Step S75, with the goal of maximizing energy storage efficiency, dynamically plan all the energy storage changes and determine the optimal path.

[0216] Step S77, determining the optimal mechanical-hydraulic energy storage ratio at each time point according to the optimal path.

[0217] Step S79, dynamically adjusting the mechanical-hydraulic energy storage ratio of the spring energy storage system according to the optimal mechanical-hydraulic energy storage ratio.

[0218] In step S73 to step S79, the dynamic programming algorithm is used to calculate and optimize the energy storage change from the current state to the next state for the predicted energy storage demand at each time point. With the goal of maximizing energy storage efficiency, the dynamic programming algorithm is used to plan all energy storage changes and determine the optimal energy distribution path. In this process, it is necessary to consider constraints such as total energy storage, remaining spring life, and hydraulic system pressure limit to ensure that the system operates within the safety and design range. Define the objective function η = (E m +E h ) / (E m,max +E h,max ), used to maximize energy storage efficiency, where E m and E h are mechanical energy storage and hydraulic energy storage, E m,max and E h,max Set constraints, including total energy storage (50-100 kJ), remaining spring life (greater than 10 ^6cycles), hydraulic system pressure limit (less than 25MPa).

[0219] Specifically, a 24-hour period is divided into 1-hour intervals, and the state of dynamic programming is established for each time point. For each time point, a state transition equation for dynamic programming is established, considering the change in energy storage from the current state to the next state. The initial distribution of mechanical energy storage and hydraulic energy storage is defined for the initial state of the dynamic programming algorithm (usually the beginning of the time series). For each time point, all possible paths for transfer from the current state to the next state are calculated, and the efficiency of each path is evaluated. Constraints are applied to the state transfer to exclude those transfers that violate the total energy storage, remaining spring life, or hydraulic pressure limits. Using the iterative method of dynamic programming, the optimal path from the initial state to the final state is gradually determined, which maximizes the objective function. Based on the optimal path, the optimal mechanical energy storage and hydraulic energy storage distribution ratio is calculated for each time point.

[0220] In addition, the model predictive control (MPC) method is used to achieve rolling optimization of the energy storage system by predicting for 6 hours and controlling in real time for 1 hour. The initial PID controller parameters are determined by the Ziegler-Nichols method and adjusted online in combination with the fuzzy adaptive algorithm with an adjustment range of ±20% to meet the needs of dynamic changes in the system. Wavelet denoising and Z-score standardization are used to improve data quality and provide accurate input for the LSTM network and MPC algorithm. The establishment of the state space model further describes the energy storage dynamics and makes the control strategy more precise. Through real-time monitoring and adaptive adjustment, the entire system can respond to external environmental changes and internal state fluctuations in a timely manner, maximize energy storage efficiency, extend spring life, and ensure that the hydraulic system pressure is within a safe range.

[0221] and Figure 1 Compared with the embodiment shown, this embodiment uses a long short-term memory network to predict the energy storage demand at each time point in the future time period, ensuring the accuracy of the prediction results. Then, the optimal path is determined through dynamic planning, thereby optimizing the energy storage change process and maximizing the energy storage efficiency. Then, based on the determined optimal path, the mechanical-hydraulic energy storage ratio at each time point is determined, so as to achieve the best energy storage configuration at different time points and ensure that the performance of the energy storage system at each time point can reach the best. By dynamically adjusting the mechanical-hydraulic energy storage ratio, it is possible to respond to changes in energy storage demand in real time, so that the system always remains efficient and stable in actual operation.

[0222] Figure 7 A flow chart for determining whether a spring has microscopic damage is provided in an embodiment of the present application, and the flow chart may include the following steps:

[0223] Step S101, obtaining a reflected wave signal of a spring in a GIS circuit breaker.

[0224] Specifically, a high-frequency ultrasonic detection device is used to scan the springs in the GIS circuit breaker at multiple angles to collect reflected wave signals at different angles. A high-frequency ultrasonic detection device with a frequency of 50MHz is selected to perform a 360-degree full-range scan of the springs in the GIS circuit breaker, and a set of reflected wave signals is collected every 15 degrees to ensure that all key angles are covered. A total of 24 sets of reflected wave signals are obtained, each corresponding to a measurement result at a specific angle. The 24 sets of reflected wave signals are superimposed in phase to enhance the useful signal and reduce random noise. Through signal superposition, the signal-to-noise ratio is improved by about 6dB.

[0225] Step S103: performing waveform analysis on the reflected wave signal and extracting a waveform feature vector.

[0226] Specifically, the collected reflected wave signals are analyzed for waveforms, and key waveform features such as amplitude (0.1-1V), frequency (center frequency is 50MHz) and phase (0-360 degrees) are extracted to form waveform feature vectors. Based on the known propagation speed of ultrasound in spring materials (about 5900m / s), the arrival time of the reflected wave is analyzed to infer the internal structure information.

[0227] Step S105, calculating the Euclidean distance between the waveform feature vector and a standard waveform feature vector in a preset standard feature library.

[0228] Specifically, a standard feature library of healthy springs is created, which contains reflection wave features under 10 different working conditions, and each state has 100 sets of sample data. The Euclidean distance between the waveform feature vector to be tested and each standard vector in the standard feature library is calculated.

[0229] Step S107: When the Euclidean distance is greater than a preset distance threshold, it is determined that the spring has microscopic damage.

[0230] Specifically, if the Euclidean distance between the waveform feature vector to be measured and any standard vector in the standard feature library is greater than a preset distance threshold, it is determined that there may be microscopic damage inside the spring. Microscopic damage is microcrack extension or microscopic plastic deformation.

[0231] and Figure 1Compared with the embodiment shown in the figure, this embodiment achieves an accurate description of the internal structure and state of the spring by acquiring the reflected wave signal, performing waveform analysis and feature vector extraction on it. This method helps to identify the working state and potential problems of the spring. Furthermore, by comparing the waveform feature vector with the data in the standard feature library and calculating the Euclidean distance, it is possible to accurately determine whether the spring has microscopic damage, thereby quickly locating the anomaly and effectively evaluating the health status of the spring.

[0232] Figure 8 The flowchart for determining the total health status score corresponding to the spring provided in the embodiment of the present application, wherein the waveform characteristic vector includes amplitude, frequency and phase. The flow may include the following steps:

[0233] Step S201, determining an amplitude attenuation rate according to a ratio of the amplitude to an initial amplitude.

[0234] Specifically, the amplitude attenuation rate ADR usually refers to the ratio of the amplitude of the reflected wave signal to its initial amplitude:

[0235]

[0236] Among them, V initial is the initial amplitude, V current is the current measured amplitude.

[0237] Step S203: determining a frequency offset according to a difference between the frequency and an initial frequency.

[0238] Specifically, the frequency offset is the difference between the frequency of the reflected wave signal and its initial frequency: Δf = |f current -f initial |, where f current is the current measured frequency, f initial is the initial frequency.

[0239] Step S205: determining a phase difference according to a difference between the phase and an initial phase.

[0240] Specifically, the phase difference refers to the difference between the phase of the reflected wave signal and the initial phase: ΔΦ=|Φ current -Φ initial |, where Φ current is the current measured phase, Φ initial is the initial phase.

[0241] Step S207: determining a total health status score corresponding to the spring based on the amplitude attenuation rate, the frequency offset and the phase difference.

[0242] Specifically, the threshold range of the characteristic parameters of the reflected wave signal is set, including the amplitude attenuation rate, the frequency offset and the phase difference. If the amplitude attenuation rate is greater than 20%, the frequency offset exceeds 2MHz, and the phase difference is greater than 30 degrees, a maximum score is set for each parameter according to the degree to which these parameters exceed the threshold, for example, the maximum score for each parameter is 100 points. If the parameter does not exceed the threshold, the maximum score is obtained; if it exceeds the threshold, the corresponding score is deducted according to the degree of excess. The scores of the three parameters are combined to obtain the total health status score of the spring energy storage system. For example, if the scores of the amplitude attenuation rate, the frequency offset and the phase difference are 85, 90 and 95 respectively, the comprehensive score may be (85+90+95) / 3=90 points.

[0243] According to the comprehensive score, the health status of the spring energy storage system is divided into three levels:

[0244] 0-60 points: Severe abnormality, immediate action may be required.

[0245] 61-80 points: Moderately abnormal, requiring further monitoring and evaluation.

[0246] 81-100 points: Slightly abnormal or normal, the system is running well.

[0247] and Figure 7 Compared with the embodiment shown in the figure, this embodiment can accurately evaluate the health status of the spring by measuring and calculating the amplitude attenuation rate, frequency offset and phase difference of the reflected wave signal. The health status of the spring is divided into different levels (severe abnormality, moderate abnormality, slight abnormality or normal) based on the scores of various parameters. It can provide a clear health status and help decision makers take corresponding measures according to the actual health status.

[0248] In some embodiments, the reflected wave signal can also be analyzed using continuous wavelet transform, and Morlet wavelet is selected as the mother wavelet, and the center frequency is set to 50MHz, which matches the frequency of the ultrasonic wave in the spring material. The wavelet coefficients of the reflected wave signal are calculated, the energy distribution of the wavelet coefficients is analyzed, and the microstructural changes inside the spring material, such as microcrack extension or microplastic deformation, are identified. If the energy concentration in a certain area exceeds 10% of the total energy, it is determined that there is damage in the area. The specific location of the abnormality inside the spring is located using a method with an accuracy of 0.1mm through the center position of the energy distribution. Specifically, the wavelet coefficient energy density map at each scale can be drawn, and these images can intuitively show the characteristics of the energy distribution in the signal. Observe the areas with significant increase or decrease in energy in the energy density map, which may indicate defects or abnormalities inside the material. Look for peak points in the energy density map, which usually correspond to damaged or defective areas inside the material. Defects of different sizes may show the greatest energy changes at different scales. By comparing the energy distribution maps of different scales, the size of the defect can be estimated. Combined with the location of the peak and the characteristics of the energy distribution, the specific location of the defect in the material is determined. The wavelet coefficient calculation and energy distribution analysis of the reflected wave signal are realized, providing a scientific basis for the non-destructive testing and health assessment of the spring energy storage system of the GIS circuit breaker.

[0249] Fig. 9 A flowchart for determining whether a spring is abnormal is provided in an embodiment of the present application, and the flowchart may include the following steps:

[0250] Step S301, obtaining the original strain signal of the spring in the GIS circuit breaker.

[0251] Specifically, a micro piezoelectric strain sensor can be made of polyvinylidene fluoride (PVDF) piezoelectric film material. A micro piezoelectric strain sensor array with a thickness of 28μm is designed to ensure that it can accurately measure tiny strain changes. The stress distribution of the GIS circuit breaker housing is simulated using finite element analysis (FEA) software to determine the stress concentration area. The micro piezoelectric strain sensor array is pasted at the five locations with the largest stress determined by the finite element analysis to achieve high-precision measurement of key areas. The sensor has a sensitivity of 15mV / με and can convert tiny strain changes into electrical signals. Using the data acquisition module, a sampling frequency of 1000Hz and an acquisition time window of 10 seconds are set to obtain the original strain signal for the spring in the GIS circuit breaker. The tiny deformation signal of the GIS circuit breaker housing is collected by the data acquisition module to obtain the original strain signal of 50,000 data points.

[0252] Step S303: performing time-frequency analysis on the original strain signal to extract time domain features and frequency domain features.

[0253] Specifically, before performing time-frequency analysis on the original strain signal, it is necessary to use a 6th-order Butterworth low-pass filter to reduce the noise of the original strain signal, and set the cutoff frequency to 200Hz to filter out high-frequency interference and noise. Then, the original strain signal after noise reduction is subjected to time-frequency analysis to extract the time domain characteristics of the original strain signal: the root mean square value (RMS) range is 0.5-2με; the peak-to-peak value range is 2-8με; the peak factor range is 3-5. The signal is analyzed in the frequency domain by a 2048-point fast Fourier transform (FFT) algorithm, and the Hanning window function is used to reduce spectrum leakage. The frequency domain characteristics of the signal are extracted from the frequency domain analysis, including the first five main frequency components and the corresponding amplitudes. These frequency components are usually in the range of 10-100Hz.

[0254] Step S305: input the time domain feature and the frequency domain feature into a preset strain signal feature-elastic modulus model to obtain an elastic modulus estimation value.

[0255] Specifically, the support vector regression algorithm is used to establish the strain signal feature-elastic modulus model, which maps the input time domain and frequency domain feature parameters to the estimated elastic modulus of the spring. The radial basis function (RBF) kernel function is selected as the kernel function in the SVR algorithm, and the parameter C (2 -5 To 2 15 ) and γ(2 -15 To 2 3 ). The parameter C controls the regularization strength of the model, while γ defines the influence range of a single training sample. The extracted feature parameters are used to train the SVR model. The input is 13 feature parameters (including RMS value, peak-to-peak value, peak factor, 5 main frequency components and corresponding amplitudes), and the output is the estimated value of the elastic modulus of the spring.

[0256] Step S307: determining the elastic modulus reduction degree corresponding to the elastic modulus estimation value.

[0257] Specifically, an initial value of the spring elastic modulus is set, for example, 210 GPa. Elastic modulus reduction degree = elastic modulus estimated value / initial value of spring elastic modulus.

[0258] Step S309: When the reduction degree of the elastic modulus exceeds a preset reduction degree threshold, it is determined that the spring is abnormal.

[0259] Specifically, if the elastic modulus decreases to less than 90% (189 GPa) of the initial value, it is determined that the spring is abnormal and a spring abnormality alarm is triggered, indicating that the elastic modulus of the spring is significantly reduced.

[0260] and Figure 1Compared with the embodiment shown, this embodiment can effectively characterize the changes and characteristics of the signal by extracting time domain features and frequency domain features from the acquired original strain signal. These features provide important information for the subsequent elastic modulus estimation, which helps to accurately identify the state of the spring. By mapping the time domain and frequency domain features to the elastic modulus estimate of the spring through the strain signal feature-elastic modulus model, the elastic modulus estimate can be accurately obtained. Calculating the degree of reduction in the elastic modulus can quantify the performance changes of the spring and provide a clear standard to evaluate whether the spring is abnormal. By setting a threshold to determine whether the spring is abnormal, automatic monitoring and alarm can be performed. If the degree of reduction in the elastic modulus exceeds the preset threshold, an alarm will be triggered. This mechanism can detect the performance degradation of the spring in a timely manner, thereby preventing potential failures or equipment damage and ensuring the reliability and safety of the equipment.

[0261] Fig.10 A flowchart for determining the total cumulative damage amount provided in an embodiment of the present application may include the following steps:

[0262] Step S401, obtaining magnetic field data generated by the spring at different positions during the charging and discharging process; wherein the magnetic field data includes magnetic field intensity and magnetic field direction.

[0263] Specifically, Hall elements and fluxgate sensors are used in combination, and a group of sensors are arranged every 10 cm along the circumference of the GIS circuit breaker housing, and three layers are arranged axially to form a 3D sensor array with 72 detection points. The measurement range of each detection unit is ±2000 Gauss, and the sensitivity is 0.1 Gauss. The sampling frequency is 1kHz, and 72,000 magnetic field data are obtained per second.

[0264] Step S403, interpolating the magnetic field data using cubic spline interpolation to generate an electromagnetic field distribution map.

[0265] Specifically, the cubic spline interpolation algorithm is used to process discrete data to generate a continuous electromagnetic field distribution map with a resolution of 1 mm. The number of interpolation nodes is 100×100×30, ensuring sufficient resolution and data accuracy.

[0266] Step S405, inputting the electromagnetic field distribution diagram as a boundary condition into a preset corresponding relationship model, and using the conjugate gradient method to solve the deformation data of the spring under the electromagnetic field; wherein the corresponding relationship model is established by finite element software.

[0267] Specifically, a corresponding relationship model between spring deformation and electromagnetic field distribution is established using finite element software, wherein the corresponding relationship model uses tetrahedral meshes to divide the spring structure; the unit size is 0.5 mm, and the total number of meshes is about 2 million. The conjugate gradient method is used for reverse solution, with a maximum number of iterations of 1000 and a convergence error of 0.1%.

[0268] Step S407, determining the expected fatigue life corresponding to the deformation data, and generating a total accumulated damage amount corresponding to the expected fatigue life.

[0269] Specifically, the linear cumulative damage theory based on Miner's criterion is used to perform fatigue analysis on the spring. The damage amount of a single cycle is obtained by dividing the number of stress cycles by the fatigue life at the corresponding stress level. The damage amounts of multiple cycles are accumulated to calculate the degree of fatigue stress accumulation.

[0270] Step S409: When the total accumulated damage exceeds a preset damage threshold, it is determined that the spring has a failure risk.

[0271] Specifically, if the cumulative damage value exceeds 0.8, it is determined that the spring has a potential failure risk.

[0272] and Figure 1 Compared with the embodiment shown in the figure, this embodiment collects the magnetic field data of the spring during the charging and discharging process with high precision, and then uses cubic spline interpolation to generate an electromagnetic field distribution map to ensure the continuity and accuracy of the data, thereby accurately reflecting the magnetic field changes of the spring. The electromagnetic field distribution map is input into the finite element model as a boundary condition, and the tetrahedral mesh and conjugate gradient method are used for high-precision deformation solution. The fatigue damage of the spring is calculated using the Miner criterion. Taking into account the stress accumulation of multiple cycles, an accurate prediction of the expected fatigue life of the spring is achieved. By calculating the total cumulative damage, if it exceeds the preset damage threshold, the potential failure risk of the spring can be effectively judged, providing a scientific basis for maintenance and replacement.

[0273] Fig.11 The flowchart of step S407 provided in the embodiment of the present application may include the following steps:

[0274] Step S4071, determining the strain corresponding to the deformation data, and generating the corresponding stress according to the strain.

[0275] Specifically, strain is the relative change in displacement to the original size and can be calculated using the following formula:

[0276]

[0277] Among them, du is the deformation data, that is, the displacement change, and x is the original size;

[0278] The relationship between stress (σ) and strain (ε) can be expressed as: σ = E·ε, where E is the elastic modulus.

[0279] Step S4073, mapping the stress to a preset SN curve of the spring material, and determining the expected fatigue life of each stress; wherein the preset SN curve is a curve that describes the number of cycles that a material can withstand under different stresses.

[0280] Specifically, the SN curve of the spring material is set, and the fatigue limit is 400MPa, corresponding to 107 cycles. The fatigue analysis of the spring is carried out based on the linear cumulative damage theory of Miner criterion.

[0281] Step S4075: For each of the stresses, the damage amount of a single cycle is calculated using the Miner criterion.

[0282] Specifically,

[0283]

[0284] Where D is the amount of damage, N is the number of cycles the spring actually experiences under the current stress, and N f is the expected fatigue life under the current stress;

[0285] Step S4077, accumulating the damage amounts of all the stresses to obtain the total accumulated damage amount.

[0286] Specifically,

[0287] D total =∑D i

[0288] Among them, D total is the total cumulative damage, D i is the damage amount under the i-th stress.

[0289] and Fig.10 Compared with the embodiment shown, this embodiment obtains the strain of the spring by calculating the ratio of the displacement change to the original size. Further, the stress is calculated by multiplying the strain and the elastic modulus. The calculated stress value is mapped to the preset SN curve, and the corresponding fatigue life can be determined for each stress value. This mapping process provides the fatigue life expectation at each stress level, ensuring the accuracy of fatigue analysis. By comparing the fatigue limit and actual stress of the spring material, its durability at different stress levels can be accurately predicted. The Miner criterion is used to calculate the damage amount of a single cycle, that is, the ratio of the actual number of cycles of the spring under the current stress to the expected fatigue life. The total cumulative damage amount is obtained by accumulating the damage amount under all stresses. The total cumulative damage amount can effectively predict the overall fatigue state of the spring and help predict possible failure risks.

[0290] Accordingly, please refer to Fig.12, a block diagram of a spring energy storage detection system for a GIS circuit breaker provided in an embodiment of the present application, the system comprising:

[0291] A concentration factor and variation index determination module 111 is used to determine the stress concentration factor and material structure variation index corresponding to the internal spring of the GIS circuit breaker based on the grating reflection wavelength obtained from multiple measurement points on the outer surface of the GIS circuit breaker, wherein the material structure variation index represents the grain size variation rate;

[0292] A remaining life prediction module 222 is used to predict the remaining life of the spring based on the stress concentration factor and the material structure change index, and determine the predicted remaining life of the spring;

[0293] An energy storage density determination module 333 is used to determine the real-time energy storage density of the spring energy storage system in the GIS circuit breaker when the predicted remaining life is lower than a preset remaining life threshold;

[0294] The dynamic adjustment module 444 is used to determine the predicted energy storage demand corresponding to the real-time energy storage density, and dynamically adjust the mechanical-hydraulic energy storage ratio of the spring energy storage system according to the predicted energy storage demand.

[0295] Optionally, the concentration coefficient and change index determination module 111 includes:

[0296] According to the grating reflection wavelength, the strain value and the maximum strain value of each measuring point on the outer surface of the GIS circuit breaker are calculated using the Bragg wavelength drift principle;

[0297] Processing the strain value using a spatial interpolation algorithm to generate a continuous strain distribution field;

[0298] Performing gradient calculation on each spatial point in the strain distribution field to obtain a gradient value and a maximum gradient value of each spatial point;

[0299] Determine the spatial point corresponding to the gradient value being greater than a preset gradient threshold as a stress concentration area;

[0300] According to the maximum strain value, the maximum gradient value and the area of ​​the stress concentration region, the stress concentration coefficient and the material structure change index are determined through a preset strain distribution-stress model.

[0301] Optionally, the remaining life prediction module 222 includes:

[0302] Acquire an image of the spring, and extract microscopic damage features of the image, wherein the microscopic damage features characterize damage and defects of the spring;

[0303] Classifying the microscopic damage characteristics to determine different damage levels;

[0304] Determining a stress accumulation factor and a fatigue index corresponding to the spring;

[0305] The stress concentration factor, the material structure change index, the damage level, the stress accumulation factor and the fatigue index are input into a preset random forest tree algorithm model to obtain the predicted remaining spring life.

[0306] Optionally, the energy storage density determination module 333 includes:

[0307] When the predicted remaining life is lower than a preset remaining life threshold, constructing an energy conversion model between the spring and the hydraulic cylinder in the spring energy storage system;

[0308] Calculating energy conversion data in the energy storage process according to the energy conversion model;

[0309] The real-time energy storage density is determined according to the energy conversion data and the total volume of the hydraulic cylinder.

[0310] Optionally, constructing an energy conversion model between a spring and a hydraulic cylinder in the spring energy storage system includes:

[0311] Obtaining the spring compression displacement, the force applied to the spring, the hydraulic cylinder pressure, and the total volume of the hydraulic cylinder;

[0312] Establishing a force-displacement relationship function of the spring according to the applied force and the compression displacement of the spring;

[0313] Establishing a pressure-volume relationship function of the hydraulic cylinder according to the hydraulic cylinder pressure and the total volume;

[0314] The energy conversion model is obtained by integrating the force-displacement relationship function and the pressure-volume relationship function.

[0315] Optionally, the dynamic adjustment module 444 includes:

[0316] According to the real-time energy storage density, a preset long short-term memory network is used to predict the energy storage demand in the future time period to obtain the predicted energy storage demand corresponding to each time point in the future time period;

[0317] For the predicted energy storage demand corresponding to any point in time, calculate the energy storage change from the current state to the next state;

[0318] With the goal of maximizing energy storage efficiency, dynamically plan all the energy storage changes and determine the optimal path;

[0319] Determining an optimal mechanical-hydraulic energy storage ratio at each time point according to the optimal path;

[0320] The mechanical-hydraulic energy storage ratio of the spring energy storage system is dynamically adjusted according to the optimal mechanical-hydraulic energy storage ratio.

[0321] Optionally, the system further comprises:

[0322] Obtain the reflected wave signal of the spring in the GIS circuit breaker;

[0323] Performing waveform analysis on the reflected wave signal and extracting waveform feature vectors;

[0324] Calculating the Euclidean distance between the waveform feature vector and a standard waveform feature vector in a preset standard feature library;

[0325] When the Euclidean distance is greater than a preset distance threshold, it is determined that microscopic damage exists in the spring.

[0326] Optionally, the waveform feature vector includes amplitude, frequency and phase; and the system further includes:

[0327] determining an amplitude decay rate according to a ratio of the amplitude to an initial amplitude;

[0328] Determining a frequency offset according to a difference between the frequency and an initial frequency;

[0329] Determining a phase difference according to a difference between the phase and an initial phase;

[0330] A total health status score corresponding to the spring is determined based on the amplitude decay rate, the frequency offset and the phase difference.

[0331] Optionally, the system further comprises:

[0332] Obtain the original strain signal of the spring in the GIS circuit breaker;

[0333] Performing time-frequency analysis on the original strain signal to extract time domain features and frequency domain features;

[0334] Inputting the time domain features and the frequency domain features into a preset strain signal feature-elastic modulus model to obtain an estimated value of the elastic modulus;

[0335] Determining a degree of reduction in the elastic modulus corresponding to the elastic modulus estimate;

[0336] When the reduction degree of the elastic modulus exceeds a preset reduction degree threshold, it is determined that the spring is abnormal.

[0337] The further functional description of each of the above modules and units is the same as that of the above corresponding embodiments and will not be repeated here.

[0338] In this embodiment, a spring energy storage detection system for a GIS circuit breaker is presented in the form of a functional unit, where the unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that executes one or more software or fixed programs, and / or other devices that can provide the above functions.

[0339] See also Fig.13 , Fig.13 A schematic diagram of the structure of a computer device provided in an embodiment of the present application is shown in FIG. Fig.13 As shown, the computer device includes: one or more processors 10, a memory 20, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. Various components are connected to each other using different buses for communication, and can be installed on a common mainboard or installed in other ways as needed. The processor can process the instructions executed in the computer device, including instructions stored in or on the memory to display the graphical information of the GUI on an external input / output device (such as, a display device coupled to the interface). In some optional embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Similarly, multiple computer devices can be connected, and each device provides some necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system). Fig.13 A processor 10 is taken as an example.

[0340] The processor 10 may be a central processing unit, a network processor or a combination thereof. The processor 10 may further include a hardware chip. The hardware chip may be a dedicated integrated circuit, a programmable logic device or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general purpose array logic or any combination thereof.

[0341] The memory 20 stores instructions executable by at least one processor 10, so that the at least one processor 10 executes the method shown in the above embodiment.

[0342] The memory 20 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created according to the use of the computer device, etc. In addition, the memory 20 may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some optional embodiments, the memory 20 may optionally include a memory remotely arranged relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0343] The memory 20 may include a volatile memory, such as a random access memory; the memory may also include a non-volatile memory, such as a flash memory, a hard disk or a solid state drive; the memory 20 may also include a combination of the above types of memory.

[0344] The computer device further comprises a communication interface 30 for the computer device to communicate with other devices or a communication network.

[0345] The embodiment of the present application also provides a computer-readable storage medium. The above method according to the embodiment of the present application can be implemented in hardware, firmware, or can be implemented as a computer code that can be recorded in a storage medium, or can be implemented as a computer code that is originally stored in a remote storage medium or a non-temporary machine-readable storage medium and will be stored in a local storage medium through a network download, so that the method described herein can be stored in such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only storage memory, a random access memory, a flash memory, a hard disk or a solid-state hard disk, etc.; further, the storage medium can also include a combination of the above types of memories. It can be understood that a computer, a processor, a microprocessor controller, or programmable hardware includes a storage component that can store or receive software or computer code. When the software or computer code is accessed and executed by a computer, a processor, or hardware, the method shown in the above embodiment is implemented.

[0346] The systems, modules or units described in the above embodiments may be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, the computer may be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

[0347] For the convenience of description, the above device is described in various units according to their functions. Of course, when implementing the present application, the functions of each unit can be implemented in the same or multiple software and / or hardware.

[0348] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods and systems. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0349] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems) according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0350] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0351] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0352] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0353] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0354] The above is only an embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.

[0355] Although the embodiments of the present application have been described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present application, and such modifications and variations are all within the scope defined by the appended claims.

Claims

1. A spring energy storage detection method for a GIS circuit breaker, characterized in that: The method comprises: Based on the grating reflection wavelengths obtained from multiple measuring points on the outer surface of the GIS circuit breaker, the stress concentration factor and material structure change index corresponding to the internal spring of the GIS circuit breaker are determined, wherein the material structure change index represents the grain size change rate; Predicting the remaining life of the spring based on the stress concentration factor and the material structure change index to determine the predicted remaining life of the spring; When the predicted remaining life is lower than a preset remaining life threshold, determining a real-time energy storage density of a spring energy storage system in the GIS circuit breaker; The predicted energy storage demand corresponding to the real-time energy storage density is determined, and the mechanical-hydraulic energy storage ratio of the spring energy storage system is dynamically adjusted according to the predicted energy storage demand.

2. The method according to claim 1, characterized in that Determining the stress concentration factor and material structure change index corresponding to the internal spring of the GIS circuit breaker based on the grating reflection wavelength obtained from multiple measuring points on the outer surface of the GIS circuit breaker includes: According to the grating reflection wavelength, the strain value and the maximum strain value of each measuring point on the outer surface of the GIS circuit breaker are calculated using the Bragg wavelength drift principle; Processing the strain value using a spatial interpolation algorithm to generate a continuous strain distribution field; Performing gradient calculation on each spatial point in the strain distribution field to obtain a gradient value and a maximum gradient value of each spatial point; Determine the spatial point corresponding to the gradient value being greater than a preset gradient threshold as a stress concentration area; According to the maximum strain value, the maximum gradient value and the area of ​​the stress concentration region, the stress concentration coefficient and the material structure change index are determined through a preset strain distribution-stress model.

3. The method according to claim 1, characterized in that: The predicting the remaining life of the spring based on the stress concentration factor and the material structure change index to determine the predicted remaining life of the spring includes: Acquire an image of the spring, and extract microscopic damage features of the image, wherein the microscopic damage features characterize damage and defects of the spring; Classifying the microscopic damage characteristics to determine different damage levels; Determining a stress accumulation factor and a fatigue index corresponding to the spring; The stress concentration factor, the material structure change index, the damage level, the stress accumulation factor and the fatigue index are input into a preset random forest tree algorithm model to obtain the predicted remaining spring life.

4. The method according to claim 1, characterized in that: When the predicted remaining life is lower than a preset remaining life threshold, determining the real-time energy storage density of the spring energy storage system in the GIS circuit breaker includes: When the predicted remaining life is lower than a preset remaining life threshold, constructing an energy conversion model between the spring and the hydraulic cylinder in the spring energy storage system; Calculating energy conversion data in the energy storage process according to the energy conversion model; The real-time energy storage density is determined according to the energy conversion data and the total volume of the hydraulic cylinder.

5. The method according to claim 4, characterized in that The energy conversion model between the spring and the hydraulic cylinder in the spring energy storage system is constructed, including: Obtaining the spring compression displacement, the force applied to the spring, the hydraulic cylinder pressure, and the total volume of the hydraulic cylinder; Establishing a force-displacement relationship function of the spring according to the applied force and the compression displacement of the spring; Establishing a pressure-volume relationship function of the hydraulic cylinder according to the hydraulic cylinder pressure and the total volume; The energy conversion model is obtained by integrating the force-displacement relationship function and the pressure-volume relationship function.

6. The method according to claim 1, characterized in that The determining of the predicted energy storage demand corresponding to the real-time energy storage density and dynamically adjusting the mechanical-hydraulic energy storage ratio of the spring energy storage system according to the predicted energy storage demand includes: According to the real-time energy storage density, a preset long short-term memory network is used to predict the energy storage demand in the future time period to obtain the predicted energy storage demand corresponding to each time point in the future time period; For the predicted energy storage demand corresponding to any point in time, calculate the energy storage change from the current state to the next state; With the goal of maximizing energy storage efficiency, dynamically plan all the energy storage changes and determine the optimal path; Determining an optimal mechanical-hydraulic energy storage ratio at each time point according to the optimal path; The mechanical-hydraulic energy storage ratio of the spring energy storage system is dynamically adjusted according to the optimal mechanical-hydraulic energy storage ratio.

7. The method according to claim 1, characterized in that The method further comprises: Obtain the reflected wave signal of the spring in the GIS circuit breaker; Performing waveform analysis on the reflected wave signal and extracting waveform feature vectors; Calculating the Euclidean distance between the waveform feature vector and a standard waveform feature vector in a preset standard feature library; When the Euclidean distance is greater than a preset distance threshold, it is determined that microscopic damage exists in the spring.

8. The method according to claim 7, characterized in that The waveform feature vector includes amplitude, frequency and phase; the method further includes: determining an amplitude decay rate according to a ratio of the amplitude to an initial amplitude; Determining a frequency offset according to a difference between the frequency and an initial frequency; Determining a phase difference according to a difference between the phase and an initial phase; A total health status score corresponding to the spring is determined based on the amplitude decay rate, the frequency offset and the phase difference.

9. The method according to claim 1, characterized in that: The method further comprises: Obtain the original strain signal of the spring in the GIS circuit breaker; Performing time-frequency analysis on the original strain signal to extract time domain features and frequency domain features; Inputting the time domain features and the frequency domain features into a preset strain signal feature-elastic modulus model to obtain an estimated value of the elastic modulus; Determining a degree of reduction in the elastic modulus corresponding to the elastic modulus estimate; When the reduction degree of the elastic modulus exceeds a preset reduction degree threshold, it is determined that the spring is abnormal.

10. A spring energy storage detection system for a GIS circuit breaker, characterized in that: The system comprises: A concentration factor and change index determination module, used to determine the stress concentration factor and material structure change index corresponding to the internal spring of the GIS circuit breaker based on the grating reflection wavelength obtained from multiple measurement points on the outer surface of the GIS circuit breaker, wherein the material structure change index represents the grain size change rate; A remaining life prediction module, used to predict the remaining life of the spring based on the stress concentration factor and the material structure change index, and determine the predicted remaining life of the spring; An energy storage density determination module is used to determine the real-time energy storage density of the spring energy storage system in the GIS circuit breaker when the predicted remaining life is lower than a preset remaining life threshold; A dynamic adjustment module is used to determine the predicted energy storage demand corresponding to the real-time energy storage density, and dynamically adjust the mechanical-hydraulic energy storage ratio of the spring energy storage system according to the predicted energy storage demand.

11. A computer device, characterized in that: include: A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the spring energy storage detection method for a GIS circuit breaker according to any one of claims 1 to 9 by executing the computer instructions.

12. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute the spring energy storage detection method for a GIS circuit breaker according to any one of claims 1 to 9.

Images