An intelligent evaluation method for high-voltage circuit breaker life cycle based on deep belief network

Abstract

The present invention discloses a method for intelligently evaluating the life cycle of a high-voltage circuit breaker based on a deep belief network, comprising the following steps: S1: acquiring historical maintenance data and environmental information of multiple high-voltage circuit breakers of the same model as detection data; S2: setting a fault source, performing multiple opening and closing actions on the high-voltage circuit breakers of the same model until a fault occurs, during which periodic data detection is performed on the high-voltage circuit breaker, and the detected data is used as fault reference data; the present invention realizes intelligent evaluation of the life cycle of a high-voltage circuit breaker by collecting the maintenance data and environmental information of the high-voltage circuit breaker and combining it with a deep belief network algorithm, so that the service life of the high-voltage circuit breaker can be more accurately evaluated, and the high-voltage circuit breaker can be replaced in time, thereby reducing the fault frequency of the high-voltage circuit breaker and saving the equipment cost and manpower cost of the power grid on the high-voltage circuit breaker.

Description

Technical Field

[0001] The present invention relates to the technical field of high-voltage circuit breakers, and in particular to an intelligent evaluation method for a high-voltage circuit breaker life cycle based on a deep belief network. Background Art

[0002] High-voltage circuit breakers can not only cut off or close the no-load current and load current in the high-voltage circuit, but also cut off the overload current and short-circuit current through the action of the relay protection device when the system fails. They have a fairly complete arc extinguishing structure and sufficient current-breaking capacity. High-voltage circuit breakers are responsible for the dual tasks of control and protection in the power system, and the quality of their performance is directly related to the safe operation of the power system. Therefore, in the maintenance of the power grid, the detection and maintenance of high-voltage circuit breakers is an important part. There are a large number of high-voltage circuit breakers in the power grid, and there are many detection items. The detection and maintenance of high-voltage circuit breakers include secondary circuit detection, mechanical property detection, contact resistance detection and other items. After the detection is completed, it is necessary to analyze and judge the detection data to determine whether the detected high-voltage circuit breaker has safety hazards and evaluate the status of the high-voltage circuit breaker. Among them, the mechanical characteristic parameter is one of the important parameters for judging the performance of the circuit breaker. According to the survey results of CIGRE and China Electric Power Research Institute, mechanical failures account for nearly 37% of switchgear failures, so it is extremely necessary to detect mechanical failures of switchgear. The current "expiry maintenance" maintenance method has serious shortcomings. Such as temporary insufficient maintenance, excessive maintenance, blind maintenance or maintenance accidents caused by improper maintenance. Condition maintenance based on the operating status of the equipment is currently the most advanced equipment maintenance method. Therefore, the life cycle assessment and fault warning research of high-voltage circuit breakers has important economic and technical significance.

[0003] For example, Chinese patent CN105467309A, published on April 6, 2015, is a method for evaluating the contact status of a high-voltage circuit breaker and a maintenance strategy. Its main technical features are: connecting a dynamic resistance tester, a large-capacity battery, a speed sensor and a current sensor to the high-voltage circuit breaker contact to form a test circuit and measuring the dynamic resistance of the high-voltage circuit breaker contact during the opening process; the dynamic resistance tester calculates and analyzes the measured speed, current and voltage signals to obtain the dynamic resistance stroke curve of the contact during the opening process, and obtains the contact resistance value and length value through the dynamic resistance stroke curve of the contact; the obtained contact resistance value and length value are compared with the standard value to evaluate the contact status and formulate an effective maintenance strategy. Its technical solution uses the contact resistance value and length value to intuitively express the contact status, which provides an important reference for evaluating the circuit breaker contact status and formulating maintenance strategies. However, it cannot provide technical guidance for the detection of secondary circuits and mechanical properties, and cannot solve the problem of lack of life cycle assessment technology for high-voltage circuit breakers. Summary of the invention

[0004] The present invention mainly solves the problem that the detection of high-voltage circuit breakers in the existing technology cannot meet the life cycle assessment problem of high-voltage circuit breakers; and provides a high-voltage circuit breaker life cycle intelligent assessment method based on a deep belief network.

[0005] The above technical problem of the present invention is mainly solved by the following technical solution: a high-voltage circuit breaker life cycle intelligent evaluation method based on a deep belief network, comprising the following steps:

[0006] S1: Obtain historical maintenance data and environmental information of multiple high-voltage circuit breakers of the same model as detection data;

[0007] S2: Set the fault source and perform multiple opening and closing actions on the high-voltage circuit breaker of the same model until a fault occurs. During this process, perform periodic data detection on the high-voltage circuit breaker and use the detected data as fault reference data;

[0008] S3: Associate the fault reference data with the fault type to create a fault data set, and use the fault data set and the detection data as sample data;

[0009] S4: Divide the sample data into a training set and a test set, build a deep belief network model, use the training set to train the deep belief network, and use the test set to test the deep belief network;

[0010] S5: Use the tested deep belief network to perform life cycle assessment of high-voltage circuit breakers.

[0011] By collecting historical maintenance data and environmental information of multiple high-voltage circuit breakers of the same model, and setting up fault sources under laboratory conditions to obtain high-voltage circuit breaker fault type data, combined with the deep belief network algorithm, the life cycle assessment of high-voltage circuit breakers is carried out with the support of sufficient data, which is more intelligent. At the same time, the life cycle assessment of high-voltage circuit breakers is more accurate, and timely replacement of high-voltage circuit breakers can be achieved, thereby reducing the failure frequency of high-voltage circuit breakers and saving the equipment and labor costs of the power grid on high-voltage circuit breakers.

[0012] Preferably, the historical maintenance data includes historical fault information of high-voltage circuit breakers, historical fault types of high-voltage circuit breakers, operating years of high-voltage circuit breakers after fault repair, mechanical properties of high-voltage circuit breakers and working status of high-voltage circuit breakers, and the environmental information includes temperature information, humidity information and magnetic field strength information of the high-voltage circuit breaker.

[0013] Preferably, the mechanical properties of the high-voltage circuit breaker are intelligently evaluated using a neural network, specifically comprising the following steps:

[0014] S11: Acquire high-voltage circuit breaker opening completion image data, high-voltage circuit breaker opening and closing process image data, high-voltage circuit breaker closing completion image data, and current waveform data of the high-voltage circuit breaker opening and closing process as mechanical property sample data;

[0015] S12: preprocessing the mechanical property sample data, and dividing the preprocessed mechanical property sample data into a training data set, a test data set, and a verification data set;

[0016] S13: construct a deep residual network model, train the deep residual network model using the training data set, verify the deep residual network model using the validation data set, and test the deep residual network model using the test data set;

[0017] S14: Mechanical performance evaluation of high-voltage circuit breakers using the trained deep residual network model.

[0018] By collecting images when the high-voltage circuit breaker is opened, during the opening and closing process, and when the closing is completed, it is determined whether the high-voltage circuit breaker has mechanical faults such as defects, deformation, and jamming. The mechanical performance state of the circuit breaker is determined according to the opening and closing state of the high-voltage circuit breaker. Machine intelligence learning is performed through the deep residual network algorithm to intelligently determine the mechanical performance state of the high-voltage circuit breaker. At the same time, the current state of the circuit breaker is also compared and detected during the opening and closing process of the circuit breaker, making the judgment of the deep residual network more accurate, thereby improving the accuracy and reliability of the evaluation results.

[0019] Preferably, the method for acquiring the image data of the high-voltage circuit breaker opening and closing process is:

[0020] S111: finding two adjacent planes of the high-voltage circuit breaker at a certain angle, and emitting circular light and dark stripes to the two adjacent planes during the opening and closing process of the high-voltage circuit breaker;

[0021] S112: Collect video data of at least two opening and closing processes of the high-voltage circuit breaker, with the starting point, end point and duration of each opening and closing of the high-voltage circuit breaker being consistent;

[0022] S113: Perform frame extraction processing on the collected video data at a period T to obtain a plurality of images of the opening and closing process of the high-voltage circuit breaker. The setting method of the period T is: Among them, t is the duration of the circuit breaker opening and closing process, δ is the vibration frequency of the high-voltage circuit breaker opening and closing process, and v is the angular velocity of the high-voltage circuit breaker opening and closing process.

[0023] The opening and closing process of the circuit breaker is illuminated through two adjacent planes. If the circuit breaker is in a normal state during the opening and closing process, the distance between the light and dark stripes will not change. However, once the circuit breaker has a large fluctuation during the opening and closing process, resulting in a large vibration amplitude, the distance between the light and dark stripes will change. By collecting data at least twice and comparing the data, it can be clearly found whether the circuit breaker is in a normal state. Video collection is performed first and then frame extraction is performed to prevent the problem of the camera being unable to capture the picture in time. After periodic frame extraction processing, the image of the circuit breaker opening and closing process is more representative.

[0024] Preferably, in step S12, the data preprocessing of the current waveform data of the circuit breaker opening and closing process includes the following steps:

[0025] S221: collecting the current waveform of the circuit breaker during the opening and closing process, filtering the collected current waveform through a low-pass filter and a high-pass filter to obtain a filtered current signal;

[0026] S222: performing truncation processing on the filtered current signal at a period T to obtain a plurality of truncation current signals;

[0027] S223: performing Fourier transformation on a plurality of truncated current signals to obtain frequency and amplitude information at each moment;

[0028] S224: After normalizing the amplitude, a time-frequency image is obtained according to the frequency at each moment and the normalized amplitude;

[0029] S225: Using the obtained time-frequency image as a current characteristic image of the circuit breaker opening and closing process, and using the current characteristic image of the circuit breaker opening and closing process as preprocessed sample data.

[0030] The collected current signal is truncated with the same period T, so that the current characteristic image of the current waveform corresponds to the circuit breaker opening and closing process image. The current characteristic image can more accurately reflect the mechanical properties of the circuit breaker and improve the accuracy of the evaluation result.

[0031] Preferably, constructing a deep residual network model in step S13 includes the following steps:

[0032] S331: Construct the basic architecture of deep residual network;

[0033] S332: Using the training data set to train the deep residual network, and adjusting basic parameters of the deep residual network;

[0034] S333: Using the validation dataset to adjust the hyperparameters of the deep residual network;

[0035] S334: Use the test dataset to test the deep residual network's ability to evaluate the mechanical properties of high-voltage circuit breakers.

[0036] Preferably, in step S12, the data preprocessing of the circuit breaker opening completion image data, the circuit breaker opening and closing process image data, and the circuit breaker closing completion image data includes the following steps:

[0037] A: grayscale processing is performed on the image data to obtain a grayscale image;

[0038] B: Set a golden circle, the interior of which is filled with grayscale features between 0 and 255. The specific method for setting the grayscale value of the internal feature is as follows: Where K is the grayscale value of the feature inside the golden circle, N is the number of features of the circuit breaker in the grayscale image, and L i is the gray value of the i-th feature of the circuit breaker in the gray image;

[0039] C: Use the golden circle to detect the coverage of the grayscale image, retain some features in the grayscale image whose grayscale value is greater than the golden circle, and return the grayscale value of some features in the grayscale image whose grayscale value is less than the golden circle to 0, and obtain the circuit breaker opening completion feature image, circuit breaker opening and closing process feature image and circuit breaker closing completion feature image.

[0040] Preferably, the fault reference data includes the detection data including closing time, opening time, closing speed, opening speed, three-phase different phases, same-phase different phases, short circuit time, no-current time, maximum speed of moving contact, average speed of moving contact, moving contact action time, bounce time, number of bounces, maximum bounce amplitude, opening and closing stroke, opening and closing process current waveform curve, time-speed-stroke dynamic curve within the opening and closing stroke of moving contact, opening distance and contact resistance.

[0041] Preferably, before dividing the sample data into a training set and a test set, the lengths of the fault data set and the detection data are supplemented to make the lengths of the fault data set and the detection data consistent, and then the fault data set and the detection data are normalized.

[0042] The beneficial effects of the present invention are as follows: by collecting images when the high-voltage circuit breaker is opened, during the opening and closing process, and when the closing is completed, it is determined whether the high-voltage circuit breaker has mechanical faults such as defects, deformation, and jamming; the mechanical performance state of the circuit breaker is determined according to the opening and closing state of the high-voltage circuit breaker; machine intelligent learning is performed through a deep residual network algorithm to intelligently determine the mechanical performance state of the high-voltage circuit breaker, thereby improving the accuracy and reliability of the evaluation results, so that the power grid can accurately grasp the performance state of the high-voltage circuit breaker; and then by collecting other maintenance data and environmental information of the high-voltage circuit breaker, combined with a deep confidence network algorithm, an intelligent evaluation of the life cycle of the high-voltage circuit breaker is achieved, so that the service life of the high-voltage circuit breaker can be more accurately evaluated, and the high-voltage circuit breaker can be replaced in time, thereby reducing the failure frequency of the high-voltage circuit breaker and saving the equipment cost and manpower cost of the power grid on the high-voltage circuit breaker. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 It is a flowchart of a high-voltage circuit breaker life cycle intelligent assessment method according to an embodiment of the present invention.

[0044] Figure 2 It is a flowchart of a method for preprocessing current waveform data according to an embodiment of the present invention. DETAILED DESCRIPTION

[0045] The technical solution of the present invention is further specifically described below through embodiments and in conjunction with the accompanying drawings.

[0046] Embodiment: A method for intelligent evaluation of high-voltage circuit breaker life cycle based on deep belief network, such as Figure 1 As shown, the following steps are included:

[0047] S1: Obtain historical maintenance data and environmental information of multiple high-voltage circuit breakers of the same model as detection data; the historical maintenance data includes historical fault information of the high-voltage circuit breaker, historical fault types of the high-voltage circuit breaker, operating years after fault repair of the high-voltage circuit breaker, mechanical properties of the high-voltage circuit breaker and working status of the high-voltage circuit breaker; the environmental information includes temperature information, humidity information and magnetic field strength information of the high-voltage circuit breaker.

[0048] The mechanical performance of high-voltage circuit breakers is intelligently evaluated using neural networks, which specifically includes the following steps:

[0049] S11: Acquire the image data of the high-voltage circuit breaker opening completion, the image data of the high-voltage circuit breaker opening and closing process, the image data of the high-voltage circuit breaker closing completion, and the current waveform data of the high-voltage circuit breaker opening and closing process as the mechanical performance sample data; the method for acquiring the image data of the high-voltage circuit breaker opening and closing process is:

[0050] S111: finding two adjacent planes of the high-voltage circuit breaker at a certain angle, and emitting circular light and dark stripes to the two adjacent planes during the opening and closing process of the high-voltage circuit breaker;

[0051] S112: Collect video data of at least two opening and closing processes of the high-voltage circuit breaker, with the starting point, end point and duration of each opening and closing of the high-voltage circuit breaker being consistent;

[0052] S113: Perform frame extraction processing on the collected video data at a period T to obtain a plurality of images of the opening and closing process of the high-voltage circuit breaker. The setting method of the period T is: Among them, t is the duration of the circuit breaker opening and closing process, δ is the vibration frequency of the high-voltage circuit breaker opening and closing process, and v is the angular velocity of the high-voltage circuit breaker opening and closing process.

[0053] S12: preprocessing the mechanical property sample data, and dividing the preprocessed mechanical property sample data into a training data set, a test data set, and a verification data set; Figure 2 As shown, the data preprocessing of the current waveform data of the circuit breaker opening and closing process includes the following steps:

[0054] S221: collecting the current waveform of the circuit breaker during the opening and closing process, filtering the collected current waveform through a low-pass filter and a high-pass filter to obtain a filtered current signal;

[0055] S222: performing truncation processing on the filtered current signal at a period T to obtain a plurality of truncation current signals;

[0056] S223: performing Fourier transformation on a plurality of truncated current signals to obtain frequency and amplitude information at each moment;

[0057] S224: After normalizing the amplitude, a time-frequency image is obtained according to the frequency at each moment and the normalized amplitude;

[0058] S225: Using the obtained time-frequency image as a current characteristic image of the circuit breaker opening and closing process, and using the current characteristic image of the circuit breaker opening and closing process as preprocessed sample data.

[0059] The data preprocessing of the circuit breaker opening completion image data, the circuit breaker opening and closing process image data, and the circuit breaker closing completion image data includes the following steps:

[0060] A: grayscale processing is performed on the image data to obtain a grayscale image;

[0061] B: Set a golden circle, the interior of which is filled with grayscale features between 0 and 255. The specific method for setting the grayscale value of the internal feature is as follows: Where K is the grayscale value of the feature inside the golden circle, N is the number of features of the circuit breaker in the grayscale image, and L i is the gray value of the i-th feature of the circuit breaker in the gray image;

[0062] C: Use the golden circle to detect the coverage of the grayscale image, retain some features in the grayscale image whose grayscale value is greater than the golden circle, and return the grayscale value of some features in the grayscale image whose grayscale value is less than the golden circle to 0, and obtain the circuit breaker opening completion feature image, circuit breaker opening and closing process feature image and circuit breaker closing completion feature image.

[0063] S13: constructing a deep residual network model, training the deep residual network model using a training data set, verifying the deep residual network model using a verification data set, and testing the deep residual network model using a test data set; constructing a deep residual network model, including the following steps:

[0064] S331: Build the basic architecture of the deep residual network, set the number of convolutional layers, residual modules, pooling layers, and fully connected layers;

[0065] S332: Using the training data set to train the deep residual network, and using the gradient descent method and the binary cross entropy loss function to optimize and adjust the basic parameters;

[0066] S333: Using the validation data set, adjust the hyperparameters of the deep residual network; the hyperparameters include the number of sample data, the size of a single sample data, the number of neurons, the learning rate, and the number of iterations;

[0067] S334: Use the test dataset to test the deep residual network's ability to evaluate the mechanical properties of high-voltage circuit breakers.

[0068] S14: Use the trained deep residual network model to evaluate the mechanical properties of high-voltage circuit breakers; use the mechanical performance indicators of high-voltage circuit breakers obtained by the evaluation to use the deep belief network to perform life cycle assessment of high-voltage circuit breakers.

[0069] S2: Setting a fault source, performing multiple opening and closing actions on a high-voltage circuit breaker of the same model until a fault occurs, during which periodic data detection is performed on the high-voltage circuit breaker, and the detected data is used as fault reference data; the fault reference data includes the detection data including closing time, opening time, closing speed, opening speed, three-phase different phases, same-phase different phases, short-circuit time, no-current time, maximum speed of the moving contact, average speed of the moving contact, moving contact action time, bounce time, number of bounces, maximum bounce amplitude, opening and closing stroke, opening and closing process current waveform curve, time-speed-stroke dynamic curve within the opening and closing stroke of the moving contact, open distance and contact resistance.

[0070] S3: Associate the fault reference data with the fault type to create a fault data set, and use the fault data set and the detection data as sample data;

[0071] S4: Supplement the length of the fault data set and the detection data to make them consistent in length, then normalize the fault data set and the detection data, divide the normalized sample data into a training set and a test set, build a deep belief network model, use the training set to train the deep belief network, and use the test set to test the deep belief network.

[0072] S5: Use the tested deep belief network to perform life cycle assessment of high-voltage circuit breakers.

[0073] The present invention determines whether the high-voltage circuit breaker has mechanical faults such as defects, deformation, and jamming by collecting images when the high-voltage circuit breaker is opened, during the opening and closing process, and when the closing is completed. The mechanical performance state of the circuit breaker is determined according to the opening and closing states of the high-voltage circuit breaker. Machine intelligent learning is performed through a deep residual network algorithm to intelligently judge the mechanical performance state of the high-voltage circuit breaker, thereby improving the accuracy and reliability of the evaluation results, so that the power grid can accurately grasp the performance state of the high-voltage circuit breaker. Subsequently, by collecting other maintenance data and environmental information of the high-voltage circuit breaker and combining it with a deep confidence network algorithm, an intelligent evaluation of the life cycle of the high-voltage circuit breaker is achieved, so that the service life of the high-voltage circuit breaker can be more accurately evaluated, and the high-voltage circuit breaker can be replaced in time, thereby reducing the failure frequency of the high-voltage circuit breaker and saving the equipment cost and manpower cost of the power grid on the high-voltage circuit breaker.

[0074] The above-described embodiment is only a preferred solution of the present invention and does not limit the present invention in any form. There are other variations and modifications without exceeding the technical solution described in the claims.

Claims

1. A method for intelligent evaluation of high-voltage circuit breaker life cycle based on deep belief network, characterized in that: The following steps are involved: S1: Acquire historical maintenance data and environmental information of multiple high-voltage circuit breakers of the same model as detection data; acquire image data of high-voltage circuit breaker opening completion, image data of high-voltage circuit breaker opening and closing process, image data of high-voltage circuit breaker closing completion, and current waveform data of high-voltage circuit breaker opening and closing process as mechanical performance sample data; The mechanical property sample data are preprocessed, and the preprocessed mechanical property sample data are divided into a training data set, a test data set, and a validation data set, which includes setting a golden circle, and the golden circle is filled with features of grayscale values ​​between 0 and 255; the grayscale image is covered by the golden circle, and some features of the grayscale image with grayscale values ​​greater than the golden circle are retained, and some features of the grayscale image with grayscale values ​​less than the golden circle are processed to return the grayscale values ​​to 0, so as to obtain a circuit breaker opening completion feature image, a circuit breaker opening and closing process feature image, and a circuit breaker closing completion feature image; S2: Set the fault source and perform multiple opening and closing actions on the high-voltage circuit breaker of the same model until a fault occurs. During this process, perform periodic data detection on the high-voltage circuit breaker and use the detected data as fault reference data; S3: Associate the fault reference data with the fault type to create a fault data set, and use the fault data set and the detection data as sample data; S4: Divide the sample data into a training set and a test set, build a deep belief network model, use the training set to train the deep belief network, and use the test set to test the deep belief network; S5: Use the tested deep belief network to perform life cycle assessment of high-voltage circuit breakers.

2. According to claim 1, a high-voltage circuit breaker life cycle intelligent assessment method based on deep belief network is characterized in that: The historical maintenance data includes historical fault information of high-voltage circuit breakers, historical fault types of high-voltage circuit breakers, operating years after high-voltage circuit breaker fault repairs, mechanical properties of high-voltage circuit breakers, and working status of high-voltage circuit breakers. The environmental information includes temperature information, humidity information, and magnetic field strength information of the high-voltage circuit breaker.

3. According to claim 2, a high-voltage circuit breaker life cycle intelligent assessment method based on deep belief network is characterized in that: The mechanical performance of the high-voltage circuit breaker is intelligently evaluated by using a neural network, which specifically includes the following steps: S11: Acquire high-voltage circuit breaker opening completion image data, high-voltage circuit breaker opening and closing process image data, high-voltage circuit breaker closing completion image data, and current waveform data of the high-voltage circuit breaker opening and closing process as mechanical property sample data; S12: preprocessing the mechanical property sample data, and dividing the preprocessed mechanical property sample data into a training data set, a test data set, and a verification data set; S13: construct a deep residual network model, train the deep residual network model using the training data set, verify the deep residual network model using the validation data set, and test the deep residual network model using the test data set; S14: Mechanical performance evaluation of high-voltage circuit breakers using the trained deep residual network model.

4. According to claim 3, a high-voltage circuit breaker life cycle intelligent assessment method based on deep belief network is characterized in that: The method for acquiring the image data of the high-voltage circuit breaker opening and closing process is as follows: S111: finding two adjacent planes of the high-voltage circuit breaker at a certain angle, and emitting circular light and dark stripes to the two adjacent planes during the opening and closing process of the high-voltage circuit breaker; S112: Collect video data of at least two opening and closing processes of the high-voltage circuit breaker, with the starting point, end point and duration of each opening and closing of the high-voltage circuit breaker being consistent; S113: Frame extraction is performed on the collected video data at a period T to obtain a plurality of images of the opening and closing process of the high-voltage circuit breaker. The setting method of the period T is: Among them, t is the duration of the circuit breaker opening and closing process, δ is the vibration frequency of the high-voltage circuit breaker opening and closing process, and v is the angular velocity of the high-voltage circuit breaker opening and closing process.

5. A high-voltage circuit breaker life cycle intelligent assessment method based on deep belief network according to claim 3 or 4, characterized in that: In step S12, the data preprocessing of the current waveform data of the circuit breaker opening and closing process includes the following steps: S221: collecting the current waveform of the circuit breaker during the opening and closing process, filtering the collected current waveform through a low-pass filter and a high-pass filter to obtain a filtered current signal; S222: performing truncation processing on the filtered current signal at a period T to obtain a plurality of truncation current signals; S223: performing Fourier transformation on a plurality of truncated current signals to obtain frequency and amplitude information at each moment; S224: After normalizing the amplitude, a time-frequency image is obtained according to the frequency at each moment and the normalized amplitude; S225: Using the obtained time-frequency image as a current characteristic image of the circuit breaker opening and closing process, and using the current characteristic image of the circuit breaker opening and closing process as preprocessed sample data.

6. A high-voltage circuit breaker life cycle intelligent assessment method based on deep belief network according to claim 3 or 4, characterized in that: In step S13, a deep residual network model is constructed, including the following steps: S331: Construct the basic architecture of deep residual network; S332: Using the training data set to train the deep residual network, and adjusting basic parameters of the deep residual network; S333: Using the validation dataset to adjust the hyperparameters of the deep residual network; S334: Use the test dataset to test the deep residual network's ability to evaluate the mechanical properties of high-voltage circuit breakers.

7. A high-voltage circuit breaker life cycle intelligent assessment method based on deep belief network according to claim 3 or 4, characterized in that: In step S12, the data preprocessing of the circuit breaker opening completion image data, the circuit breaker opening and closing process image data, and the circuit breaker closing completion image data includes the following steps: A: grayscale processing is performed on the image data to obtain a grayscale image; B: Set a golden circle, the interior of which is filled with grayscale features between 0 and 255. The specific method for setting the grayscale value of the internal feature is as follows: Where K is the grayscale value of the feature inside the golden circle, N is the number of features of the circuit breaker in the grayscale image, and L i is the gray value of the i-th feature of the circuit breaker in the gray image; C: Use the golden circle to detect the coverage of the grayscale image, retain some features in the grayscale image whose grayscale value is greater than the golden circle, and return the grayscale value of some features in the grayscale image whose grayscale value is less than the golden circle to 0, and obtain the circuit breaker opening completion feature image, circuit breaker opening and closing process feature image and circuit breaker closing completion feature image.

8. The method for intelligent evaluation of high-voltage circuit breaker life cycle based on deep belief network according to claim 1 is characterized in that: The fault reference data includes the detection data including closing time, opening time, closing speed, opening speed, three-phase different phases, same-phase different phases, short circuit time, no-current time, maximum speed of moving contact, average speed of moving contact, moving contact action time, bounce time, number of bounces, maximum bounce amplitude, opening and closing stroke, opening and closing process current waveform curve, time-speed-stroke dynamic curve within the opening and closing stroke of moving contact, opening distance and contact resistance.

9. The method for intelligent evaluation of high-voltage circuit breaker life cycle based on deep belief network according to claim 1 is characterized in that: Before dividing the sample data into a training set and a test set, the length of the fault data set and the detection data is supplemented to make the length of the fault data set and the detection data consistent, and then the fault data set and the detection data are normalized.

Images