CN110705038B - A high-voltage circuit breaker life cycle assessment and fault early warning method - Google Patents

Abstract

The invention relates to the technical field of power equipment maintenance, and specifically relates to a high-voltage circuit breaker life cycle assessment and fault early warning method, which includes the following steps: A) obtaining detection data; B setting a fault source so that the high-voltage circuit breaker continuously opens and closes until an occurrence occurs. Fault, the number of opening and closing actions is N, and detection is performed periodically; C) obtain the life cycle assessment results; D) train the fault warning neural network model; E) input the detection data of the high-voltage circuit breaker to be evaluated and warned into the fault warning Neural network model, if the fault early warning neural network model outputs a fault type, a fault early warning is issued, and the corresponding fault is the fault type output by the fault early warning neural network model. The substantial effect of the present invention is to obtain a representation of the life cycle of the high-voltage circuit breaker, make the maintenance of the high-voltage circuit breaker more targeted, and improve the stability and reliability of the operation of the high-voltage circuit breaker.

Description

A high-voltage circuit breaker life cycle assessment and fault early warning method

Technical field

The invention relates to the technical field of power equipment maintenance, and in particular to a high-voltage circuit breaker life cycle assessment and fault early warning method.

Background technique

The high-voltage circuit breaker 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. It has a fairly complete arc extinguishing structure and sufficient Current interruption capability. High-voltage circuit breakers are responsible for the dual tasks of control and protection in power systems, and their performance is directly related to the safe operation of the power system. Therefore, in the maintenance of power grid, the detection and maintenance of high-voltage circuit breakers is an important content. There are a large number of high-voltage circuit breakers in the power grid, and there are many testing items. The detection and maintenance of high-voltage circuit breakers include secondary circuit detection, mechanical characteristics detection, contact resistance detection and other items. After the detection is completed, the detection data needs to be analyzed and judged to determine whether the detected high-voltage circuit breaker is safe. Hidden dangers, evaluate the status of high-voltage circuit breakers. Among them, the mechanical characteristic parameters are 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 faults account for nearly 37% of switch equipment failures, so it is extremely necessary to detect mechanical faults in switch equipment. The current "due maintenance" maintenance method has serious shortcomings. Such as temporary insufficient maintenance, excessive maintenance, blind maintenance or maintenance accidents caused by improper maintenance. Condition-based maintenance based on the operating status of the equipment is currently the most advanced equipment maintenance method. Therefore, it is of great economic and technical significance to conduct life cycle assessment and fault warning research on high-voltage circuit breakers.

For example, Chinese patent CN105467309A, published on April 6, 2015, describes a high-voltage circuit breaker contact status evaluation method and maintenance strategy. Its main technical features are: combining a dynamic resistance tester, a large-capacity battery, a speed sensor and a current sensor with The high-voltage circuit breaker contacts are connected into a test circuit and the dynamic resistance of the high-voltage circuit breaker contacts during the opening process is measured; the dynamic resistance tester calculates and analyzes the measured speed, current and voltage signals to obtain the contact dynamics during the opening process. Resistance-stroke curve, and obtain the contact resistance value and length value through the contact dynamic resistance-stroke curve; compare the obtained contact resistance value and length value with the standard value, evaluate the contact status, and formulate an effective Maintenance strategy. Its technical solution uses contact resistance and length values to intuitively represent the contact status, providing 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 characteristics, and cannot solve the problem of the lack of life cycle assessment technology for high-voltage circuit breakers.

Contents of the invention

The technical problem to be solved by the present invention is that the current detection of high-voltage circuit breakers cannot provide the technical problem of life cycle assessment of high-voltage circuit breakers. This paper proposes a life cycle assessment and fault warning method for high-voltage circuit breakers based on big data technology.

In order to solve the above technical problems, the technical solution adopted by the present invention is: a high-voltage circuit breaker life cycle assessment and fault warning method, which includes the following steps: A) Obtain the detection data of the same type of high-voltage circuit breaker during historical maintenance; B) Obtain the same type of high-voltage circuit breaker, and artificially set the fault source under laboratory conditions, so that the high-voltage circuit breaker will continue to perform live opening and closing actions in the presence of the fault source until a fault occurs. The high-voltage circuit breaker will perform live opening and closing before the fault. The number of gate operations is N, and detection is performed periodically at the same time. The detection data is used as reference data and is associated with the number of live opening and closing operations n corresponding to the test; C) Compare the detection data of the high-voltage circuit breaker with the reference data to obtain the detection data The number of times n corresponding to the closest reference data, use n/N as the life cycle assessment result of the high-voltage circuit breaker; D) associate the reference data with the fault type that occurs to form sample data, and use the sample data to train the fault warning neural network model ; E) Input the detection data of the high-voltage circuit breaker to be evaluated and warned into the fault warning neural network model obtained in step D). If the fault warning neural network model outputs the fault type, a fault warning is issued, and the corresponding fault is the fault warning neural network. The fault type output by the model. By correlating the opening and closing times with the detection data, the equivalent opening and closing times can be obtained by inputting the detection data, thereby obtaining a representation of the life cycle of the high-voltage circuit breaker, making the maintenance of the high-voltage circuit breaker more targeted; through the fault early warning neural network The model provides fault warning information for high-voltage circuit breakers, allowing maintenance personnel to maintain targeted high-voltage circuit breakers and improve the stability and reliability of high-voltage circuit breakers.

Preferably, step D) also includes: associating the detection data M times before the failure with the corresponding fault type to form fault sample data, using the fault sample data to train the fault research and judgment neural network model; using the (N-M) times before the failure The detection data is associated with the fault type that occurs, forming sample data to train the fault early warning neural network model; step E also includes: inputting the detection data of the high-voltage circuit breaker into the fault research and judgment neural network model obtained in step D). If the fault analysis and judgment neural network model If the network model outputs the fault type, a fault alarm will be issued. The data close to the fault can be used as sample data for fault analysis and judgment, and thus used to train the fault analysis and judgment neural network model.

As a preference, in step B), the method of artificially setting the fault source includes the following steps: B11) Conduct several tests on the high-voltage circuit breaker; B12) According to the maintenance requirements of the high-voltage circuit breaker, select one maintenance requirement in order to make it fail to meet the standards. After several live opening and closing actions, conduct several tests; B13) Select two maintenance requirements in order to make them fail to meet the standards, and after several live opening and closing actions, conduct several tests; B14) Use liquid nitrogen or dry ice Quickly cool the high-voltage circuit breaker, conduct several mechanical characteristic tests, and obtain detection data of the mechanical characteristic tests. Actively generate faults to collect fault data, effectively solving the problem of insufficient fault data samples. After being cooled by liquid nitrogen or dry ice, the lubricating performance of the lubricant or lubricating oil is reduced, thereby simulating a stuck state. After the test is completed, the lubricating performance of the lubricant or lubricating oil is restored, thereby non-destructively simulating the stuck state of mechanical parts. Fault type, obtain status data under this fault type. Naturally, mechanical parts get stuck due to poor lubrication or the ingress of dust particles.

As an option, in step D, the method of establishing a high-voltage circuit breaker fault research and judgment model includes: D21) Obtaining all detection data, associating the detection data with the corresponding fault types as sample data; D22) Preprocessing and normalizing the sample data processing, train the neural network model, and use the trained neural network model as a fault analysis and judgment model.

Preferably, in step D21), the method of associating the detection data with the corresponding fault type includes: D211) obtaining the detection data in step B11) as historical detection data; D212) combining several of the detection data in step B12) and step B13) Several sets of detection data obtained from each inspection are compared with historical detection data in sequence. If the difference between the detection data and historical detection data is greater than the preset threshold, then the set of detection data is associated with unqualified maintenance requirements; D213) associate the steps in step B14) with Several sets of detection data obtained from several mechanical characteristic tests are compared with historical detection data. If the difference between the detection data and historical detection data is greater than the preset threshold, the set of detection data will be associated with the mechanical component jamming fault.

As a preferred method, the method for judging whether the difference between the detection data and the historical detection data is greater than the preset threshold includes: D31) segmenting the numerical quantities in the detection data and historical detection data, using the segmented intervals as names, and converting the numerical quantities into states. Quantity; D32) Convert the state quantities in the detection data and historical detection data into Boolean quantities, and use {0,1} to represent false and true respectively; D32) Treat the Boolean quantities of the processed historical detection data as numerical values to calculate the average , round the mean to an integer, and re-treat the obtained integer as a Boolean quantity. Sort the processed detection data and historical detection data according to the settings to form a detection vector and a historical detection vector respectively; D33) Calculate the detection vector and the historical detection vector The distance is compared with the preset distance threshold. If the distance is greater than the preset distance threshold, it is determined that the distance between the detection data corresponding to the detection vector and the historical detection data is greater than the preset threshold. Otherwise, it is determined that the distance between the detection data corresponding to the detection vector and the historical detection data is greater than the preset threshold. The distance of the data is no greater than the preset threshold. Boolean can only eliminate differences between ranges of data values.

As an alternative, the method of judging whether the difference between the detection data and the historical detection data is greater than the preset threshold includes: D41) Convert the status quantities in the detection data and historical detection data into Boolean quantities, and use {0,1} to represent false and true respectively ; D42) Normalize the numerical quantities in the detection data and historical detection data, obtain the minimum and maximum values of each item of the normalized historical detection data, and convert the processed Boolean quantities and numerical quantities according to the settings Arrange in a certain order, the detection data is sorted to form a detection vector, the minimum values of the historical detection data are sorted to form the historical detection left vector, and the maximum values of the historical detection data are sorted to form the historical detection right vector; D43) Calculate the detection vectors respectively The distance from the historical detection left vector and the historical detection right vector is compared with the preset distance threshold. If the distance between the detection vector and the historical detection left vector and the historical detection right vector are both greater than the preset distance threshold, then the detection corresponding to the detection vector is determined. The distance between the data and the historical detection data is greater than the preset threshold. On the contrary, it is determined that the distance between the detection data corresponding to the detection vector and the historical detection data is not greater than the preset threshold.

As an alternative, the method of judging whether the difference between the detection data and the historical detection data is greater than the preset threshold includes: D51) segmenting the numerical quantities in the detection data and historical detection data, using the segmented interval as the name, and converting the numerical quantities into states D52) Convert the state quantities in the detection data and historical detection data into Boolean quantities, and use {0,1} to represent false and true respectively; D52) Convert the Boolean quantities in the processed historical detection data, there are different values Delete the Boolean quantities of the value, sort the remaining Boolean quantities of the historical detection data according to the settings, and form the historical detection vectors respectively. Select the Boolean quantities corresponding to the historical detection vectors in the detection data, and sort them according to the settings to form the detection vector; D53 ) Calculate the distance between the detection vector and the historical detection vector, and compare it with the preset distance threshold. If the distance is greater than the preset distance threshold, it is determined that the distance between the detection data corresponding to the detection vector and the historical detection data is greater than the preset threshold. Otherwise, it is determined The distance between the detection data corresponding to the detection vector and the historical detection data is not greater than the preset threshold. By constructing a vector of Boolean type, the deviation caused by numerical data can be eliminated. Deleting Boolean quantities with different values and retaining only Boolean quantities with consistent values can eliminate the influence of irrelevant factors and make the judgment results more valuable for reference.

Preferably, in step D51), the method of segmenting the numerical quantities in the detection data and historical detection data includes: D511) selecting a numerical quantity, obtaining all the values of the numerical quantity in the historical detection data, pressing The numerical values are arranged in order and recorded as a set Ki. The minimum value in the set Ki is k _min and the maximum value is k _max ; D512) Assign the initial value of the partition starting point k _s as k _min and the partition end point k _e as k _max , the inspection value k _m = _ks +n×Δk, Δk is the artificially set step size, n is a positive integer, and the initial value of n is 1; D513) n continues to increase by 1, if the inspection value k _m meets the following conditions:

Among them, the function N(x, y) represents the set Ki and the number of data whose data values are in the numerical interval (x, y). Then (2k m -k s) is used as the interval dividing point and added to the dividing point set Km, and (2k _m -k _s ) is added to the dividing point set Km. 2k _m -k _s ) is assigned to k _s , and n continues to increase by 1 until k _m > k _max ; D514) Add k _min and k _max to the set Km, and use the values within Km as the dividing point, Divide the numerical quantity data into numerical intervals; D515) Select the next numerical quantity and repeat steps D511) to D514) until all numerical quantities are divided into interval segments; D516) The detection data adopts the corresponding numerical quantity in the historical detection data. Interval division. Segmentation based on the aggregation characteristics of the value itself can make the segmentation closer to the different states of the value.

Preferably, in step D51), using the segmented interval as the name, the method of converting the numerical quantity into the state quantity includes the following steps: D511) divide the numerical quantity data into several intervals, [n _m(1) , n _{m( 2)} ], [n _m(2) , n _m(3) ]...[n _m(k-1) , n _m(k) ], where n _m(1) and n _m(k) are respectively The starting point and end point of the numerical interval, nm(2)~nm(k-1) are the middle dividing points of the numerical interval, and Respectively as the state name of the corresponding numerical interval; D512) If the data of the historical detection numerical quantity falls into the interval [n _m(d) , n _m(d+1) ], d∈[1, k-1], then status name As the value of this numerical quantity, the conversion of numerical quantity data into state quantity data is completed. It can quickly convert numerical quantities into status quantities.

Preferably, in step D52), the method of converting the state quantity in the detection data and historical detection data into a Boolean quantity includes the following steps: D521) Obtaining all the state values of the state quantity data; D522) Using the state value as the field name The status quantity field is split into multiple fields; D523) Set the field whose field name is the same as the status quantity data value to 1, and set the other split fields to 0 to complete the splitting of the status quantity data into Boolean data. Splitting the state quantity into Boolean quantities can speed up the training efficiency of the neural network.

Preferably, in step D512), the setting method of the step size Δk includes: calculating the pairwise differences of the numerical quantity data in the set Ki, eliminating the differences that are zero, performing an absolute value operation on the remaining differences, and calculating the The minimum value is used as the step size Δk and participates in the calculation.

Preferably, in step D521), the method for obtaining all state values of the state quantity data is: if the state quantity data is the state of the circuit breaker itself, then all the state values include all possible values of the state; if the state quantity data If the data is state quantity data converted from numerical quantity data, all state values only include values that have appeared in historical states.

Preferably, the detection data includes closing time, opening time, just-closing speed, just-opening speed, three-phase different periods, same-phase different periods, golden short time, no flow time, maximum speed of moving contact, moving contact Contact average speed, moving contact action time, bounce time, bounce times, maximum bounce amplitude, opening and closing stroke, current waveform curve during opening and closing process, time speed stroke dynamic curve within the opening and closing stroke of the moving contact, opening and closing stroke distance and contact resistance. By obtaining various data of the high-voltage circuit breaker, the status data of the high-voltage circuit breaker is made more comprehensive, which helps to improve the accuracy of fault analysis and judgment, and at the same time provides conditions for discovering unobvious abnormal data.

As a preferred method, in step D, before using the fault sample data to train the fault analysis and judgment neural network model, the fault sample data is normalized, including: D11) enumerating the numerical data in the fault sample data to obtain the theoretical boundaries of the numerical data The value is used as the boundary value. If there is no theoretical boundary value, the historical boundary value of the data is obtained as the boundary value. The left boundary value of the boundary value is regarded as 0, and the right boundary value of the boundary value is regarded as 1. The numerical data is subtracted from the left boundary value. The difference between the boundary values divided by the difference between the right boundary value and the left boundary value is used as the normalized value of the numerical data; D12) Split the state quantity data into several Boolean data; D13) Convert the Boolean data is a numerical value and normalized.

As a preference, in step D12, the method of splitting the state amount data into several Boolean data includes: D121) Obtaining all state values of the state amount data; D122) Using the state value as the field name to split the state amount field into Multiple fields; D123) Set the field whose field name is the same as the value of the state quantity data, and set the remaining split fields to zero to complete the splitting of the state quantity data into Boolean data.

Preferably, in step B), before testing the high-voltage circuit breaker, a non-contact displacement sensor is installed on each mechanical moving part of the high-voltage circuit breaker, and the displacement data measured by the non-contact displacement sensor is added to the high-voltage circuit breaker. in the detection data of the circuit breaker.

Preferably, in step E), a non-contact displacement sensor is installed on each mechanical moving part of the high-voltage circuit breaker to be evaluated and warned, and the displacement data measured by the non-contact displacement sensor is obtained. The high-voltage circuit breaker status data and the displacement data measured by the non-contact displacement sensor are used as the input of the fault research and judgment neural network model to train the fault research and judgment neural network model and conduct fault research and judgment.

Preferably, in step B), a non-contact displacement sensor is installed on each mechanical moving part of the normal high-voltage circuit breaker. Under power-off conditions, the opening and closing test of the high-voltage circuit breaker is repeated until the high-voltage circuit breaker is disconnected. If the mechanical parts of the circuit breaker are damaged, record the number of opening and closing times K during the test, and the displacement data of each mechanical moving part during the opening and closing process as historical displacement data; in step E), if the high-voltage circuit breaker to be determined is If the fault analysis result is no fault, then a non-contact displacement sensor is installed on each mechanical moving part of the high-voltage circuit breaker to be analyzed, and the high-voltage circuit breaker to be analyzed is opened and closed once to obtain the measured value of the non-contact displacement sensor. The obtained displacement data is compared with the historical displacement data to obtain the number of opening and closing tests k corresponding to the closest historical displacement data, and (K-k) is used as the remaining service life of the high-voltage circuit breaker to be determined.

The non-contact displacement sensor includes a laser emitter, current limiting resistor, photoresistor, power supply module, reflective sticker, voltage sensor and communication module. The laser emitter is fixedly installed in the shell of the high-voltage circuit breaker and is aligned with the mechanical moving parts in the normal direction. An alignment point on the outer surface is adjusted so that the emitted light of the laser emitter has an angle with the normal direction of the outer surface of the mechanical moving part. Within the stroke of the mechanical moving part, the alignment point of the laser emitter moves along the outer surface of the mechanical moving part, forming Moving range, reflective stickers are attached to the mechanical moving parts and cover the moving range of the alignment point. The reflective stickers have several highly reflective areas arranged at equal intervals along the stroke of the mechanical moving parts. Between adjacent high reflective areas are Low-reflection zone, the width of the high-reflection zone is equal to the width of the low-reflection zone, the spot diameter of the laser emitter is equal to an integral multiple of the interval width, the photoresistor is installed on the other side of the laser emitter that is symmetrical with respect to the normal direction of the outer surface of the mechanical moving part, One end of the photoresistor is grounded, and the other end is connected to the power supply module through a current limiting resistor. The voltage sensor collects the voltage at the connection point between the photoresistor and the current limiting resistor, and the voltage sensor is connected to the communication module.

The substantial effect of the present invention is: by correlating the opening and closing times with detection data, the equivalent opening and closing times can be obtained by inputting the detection data, thereby obtaining a representation of the life cycle of the high-voltage circuit breaker and making the maintenance of the high-voltage circuit breaker more efficient. Targeted; the fault warning neural network model provides fault warning information for high-voltage circuit breakers, allowing maintenance personnel to maintain targeted high-voltage circuit breakers and improve the stability and reliability of high-voltage circuit breakers; by proactively setting fault sources, it can solve The technical problem is that the number of detection data samples under fault is small, and the detection data and fault are more relevant, which helps to improve the accuracy of fault analysis. By normalizing the sample data, the fault assessment model can be accelerated. The convergence speed speeds up the establishment efficiency of the fault assessment model and improves the accuracy of the fault assessment model.

Description of the drawings

Figure 1 is a flow chart of Embodiment 1.

FIG. 2 is a flow chart of a method for manually setting a fault source according to the embodiment.

Figure 3 is a flow chart of a method for associating detection data with corresponding fault types in Embodiment 1.

Figure 4 is a flow chart of a method for determining whether the difference is greater than a preset threshold in Embodiment 1.

Figure 5 is a schematic structural diagram of a non-contact displacement sensor according to the first embodiment.

Figures 6 and 7 are measurement schematic diagrams of the non-contact displacement sensor according to the first embodiment.

Figure 8 is a flow chart of a method for determining whether the difference is greater than a preset threshold in Embodiment 2.

Figure 9 is a flow chart of a method for determining whether the difference is greater than a preset threshold in Embodiment 3.

Among them: 1. Linear reflection sticker, 2. Laser transmitter, 3. Cylindrical reflection sticker, 4. Cam, 5. Cylindrical end reflection sticker, 6. Moving parts, 7. Alignment point trajectory, 8. Arc reflection sticker, 100. Voltage sensor, 200. Communication module.

Detailed ways

The specific implementation manner of the present invention will be further described in detail below through specific examples and in conjunction with the accompanying drawings.

Example 1:

A method for life cycle assessment and fault warning of high-voltage circuit breakers, as shown in Figure 1, includes the following steps: A) Obtain historical maintenance detection data of high-voltage circuit breakers of the same model. The detection data includes closing time, opening time, just closing speed, just opening speed, three-phase different phases, same phase and different phases, golden short time, no flow time, maximum speed of moving contacts, average speed of moving contacts, Moving contact action time, bounce time, number of bounces, maximum bounce amplitude, opening and closing stroke, current waveform curve during opening and closing process, time speed stroke dynamic curve within the opening and closing stroke of the moving contact, opening distance and contact resistance. By obtaining various data of the high-voltage circuit breaker, the status data of the high-voltage circuit breaker is made more comprehensive, which helps to improve the accuracy of fault analysis and judgment, and at the same time provides conditions for discovering unobvious abnormal data.

B) Obtain a high-voltage circuit breaker of the same model, install a non-contact displacement sensor on each mechanical moving part of the high-voltage circuit breaker, and add the displacement data measured by the non-contact displacement sensor to the detection data of the high-voltage circuit breaker , the fault source is artificially set up under laboratory conditions, so that the high-voltage circuit breaker continuously performs live opening and closing actions in the presence of the fault source until a fault occurs. The number of live opening and closing actions of the high-voltage circuit breaker before the fault is N, and at the same time Detection is performed periodically, and the detection data is used as reference data and is associated with the number of live opening and closing operations n corresponding to the test. A non-contact displacement sensor is installed on each mechanical moving part of a normal high-voltage circuit breaker. Under power outage conditions, the opening and closing test of the high-voltage circuit breaker is repeated until the mechanical parts of the high-voltage circuit breaker are damaged, and the records are recorded The number of opening and closing times K during the test and the displacement data of each mechanical moving part during the opening and closing process are used as historical displacement data.

As shown in Figure 2, the method of artificially setting the fault source includes the following steps: B11) Test the high-voltage circuit breaker several times; B12) According to the maintenance requirements of the high-voltage circuit breaker, select one maintenance requirement in order to make it fail to meet the standards, and conduct several After several live opening and closing actions, conduct several tests; B13) Select two maintenance requirements in order to make them fail to meet the standards, and after several live opening and closing actions, conduct several tests; B14) Use liquid nitrogen or dry ice for rapid cooling For high-voltage circuit breakers, conduct several mechanical characteristic tests to obtain detection data of mechanical characteristic tests. Actively generate faults to collect fault data, effectively solving the problem of insufficient fault data samples. After being cooled by liquid nitrogen or dry ice, the lubricating performance of the lubricant or lubricating oil is reduced, thereby simulating a stuck state. After the test is completed, the lubricating performance of the lubricant or lubricating oil is restored, thereby non-destructively simulating the stuck state of mechanical parts. Fault type, obtain status data under this fault type. Naturally, mechanical parts get stuck due to poor lubrication or the ingress of dust particles.

As shown in Figure 5, the non-contact displacement sensor includes a laser emitter 2, a current limiting resistor, a photoresistor, a power supply module, a reflective sticker, a voltage sensor 100 and a communication module 200. The laser emitter 2 is fixedly installed on the shell of the high-voltage circuit breaker. inside, align an alignment point along the normal direction to the outer surface of the mechanical moving part 6, and adjust it so that the light emitted by the laser emitter 2 has an angle with the normal direction of the outer surface of the mechanical moving part 6. Within the stroke of the mechanical moving part 6, the laser emits The alignment point of the device 2 moves along the outer surface of the mechanical moving part 6 to form a moving range. The reflective sticker is attached to the mechanical moving part 6 and covers the moving range of the alignment point. The reflective sticker has several equidistantly spaced edges along the stroke of the mechanical moving part 6. Arranged high-reflective areas, low-reflective areas between adjacent high-reflective areas, the width of the high-reflective area is equal to the width of the low-reflective area, the spot diameter of the laser transmitter 2 is equal to an integral multiple of the interval width, the photoresistor installation and laser emission On the other side of the device 2 that is symmetrical to the normal direction of the outer surface of the mechanical moving part 6, one end of the photoresistor is grounded, and the other end is connected to the power supply module through a current limiting resistor. The voltage sensor 100 collects the voltage at the connection point between the photoresistor and the current limiting resistor. The voltage sensor 100 is connected to the communication module 200. Figure 5 shows a linear reflection sticker 1, and the detected mechanical moving parts 6 move along a straight line, such as moving contacts, unlocking locks, etc. As shown in Figure 6, when performing non-contact displacement detection on rotating parts such as the shaft and cam 4, a cylindrical reflective sticker 3 can be attached to the outer surface of the shaft or the equal-radius arc part of the cam 4. In order to avoid The picture is blurry, and the distance between high-reflection areas and low-reflection areas in the picture is distorted. When the equal-radius arc part of the cam 4 is also the working surface, a cylindrical end surface reflective sticker 5 can be attached to the end surface of the cam 4 . As shown in Figure 7, when the moving part 6 to be detected has complex planar motion, that is, it includes both translational motion and rotational motion, select an appropriate alignment point on the moving part 6 to be detected so that the alignment point is within the stroke of the moving part 6. Always on the moving part 6, the alignment point trajectory 7 will be an arc, and an appropriate arc-shaped reflective sticker 8 is attached. The arc-shaped reflective sticker 8 arranges high-reflective areas and low-reflective areas at intervals along the arc, so that the high-reflective area And the edges of the low-reflection area are perpendicular to the arc at the corresponding position. This embodiment provides a non-contact displacement sensor implementation. In the prior art, non-contact displacement sensors are known to be used to detect vibration and displacement. Those skilled in the art can design other forms of non-contact displacement by themselves. Sensors are used to detect displacement.

C) Compare the detection data of the high-voltage circuit breaker with the reference data to obtain the number of times n corresponding to the reference data that is closest to the detection data, and use n/N as the life cycle assessment result of the high-voltage circuit breaker.

D) Associate the reference data with the type of fault that occurred to form sample data. Use the sample data to train the fault warning neural network model. Associate the detection data M times before the failure with the corresponding fault type to form the fault sample data. Use The fault sample data trains the fault analysis and judgment neural network model; the detection data (N-M) times before the fault occurs is used, and is associated with the fault type that occurs to form the sample data to train the fault warning neural network model.

The method of establishing a high-voltage circuit breaker fault research and judgment model includes: D21) Obtaining all detection data, associating the detection data with the corresponding fault types as sample data; D22) Preprocessing and normalizing the sample data, and training the neural network model , using the trained neural network model as a fault analysis and judgment model. As shown in Figure 3, in step D21), the method of associating the detection data with the corresponding fault type includes: D211) obtaining the detection data in step B11) as historical detection data; D212) combining step B12) and step B13) , several sets of detection data obtained from several inspections are compared with historical detection data in sequence. If the difference between the detection data and historical detection data is greater than the preset threshold, then the set of detection data is associated with substandard maintenance requirements; D213) Step B14 ) Compare several sets of detection data obtained from several mechanical characteristic tests with historical detection data. If the difference between the detection data and historical detection data is greater than the preset threshold, then the set of detection data will be associated with the mechanical component jamming fault.

In step D, before using the fault sample data to train the fault analysis and judgment neural network model, normalize the fault sample data, including: D11) List the numerical data in the fault sample data and obtain the theoretical boundary value of the numerical data as the boundary. value. If there is no theoretical boundary value, the historical boundary value of the data is obtained as the boundary value. The left boundary value of the boundary value is regarded as 0, and the right boundary value of the boundary value is regarded as 1. The numerical data minus the left boundary value is The difference is divided by the difference between the right boundary value and the left boundary value as the normalized value of the numerical data; D12) Split the state quantity data into several Boolean data; D13) Convert the Boolean data into numerical values, and normalized.

In step D12, the method of splitting the status data into several Boolean data includes: D121) Obtaining all status values of the status data; D122) Splitting the status field into multiple fields using the status value as the field name ;D123) Set the fields whose field names are the same as the value of the state quantity data to 1, and set the remaining split fields to zero to complete the splitting of the state quantity data into Boolean data.

As shown in Figure 4, the method for judging whether the difference between the detection data and the historical detection data is greater than the preset threshold includes: D31) Segment the numerical quantities in the detection data and historical detection data, and use the segmentation interval as the name, and divide the numerical quantities into Convert to state quantities; D32) Convert the state quantities in the detection data and historical detection data into Boolean quantities, and use {0,1} to represent false and true respectively; D32) Treat the Boolean quantities of the processed historical detection data as Calculate the average of the values, round the average to an integer, and re-treat the obtained integer as a Boolean quantity. Sort the processed detection data and historical detection data according to the settings to form a detection vector and a historical detection vector respectively; D33) Calculate the detection vector and history The distance of the detection vector is compared with the preset distance threshold. If the distance is greater than the preset distance threshold, it is determined that the distance between the detection data corresponding to the detection vector and the historical detection data is greater than the preset threshold. Otherwise, the detection data corresponding to the detection vector is determined. The distance from historical detection data is not greater than the preset threshold. Boolean can only eliminate differences between ranges of data values.

E) Install a non-contact displacement sensor on each mechanical moving part of the high-voltage circuit breaker to be evaluated and warned, obtain the displacement data measured by the non-contact displacement sensor, and convert the status data of the high-voltage circuit breaker to be evaluated and warned Together with the displacement data measured by the non-contact displacement sensor, it is used as the input of the fault analysis and judgment neural network model to train the fault analysis and judgment neural network model and conduct fault analysis and judgment.

Input the detection data of the high-voltage circuit breaker to be evaluated and warned into the fault early warning neural network model obtained in step D). If the fault early warning neural network model outputs the fault type, a fault early warning will be issued, and the corresponding fault is the output of the fault early warning neural network model. Fault type. By correlating the opening and closing times with the detection data, the equivalent opening and closing times can be obtained by inputting the detection data, thereby obtaining a representation of the life cycle of the high-voltage circuit breaker, making the maintenance of the high-voltage circuit breaker more targeted; through the fault early warning neural network The model provides fault warning information for high-voltage circuit breakers, allowing maintenance personnel to maintain targeted high-voltage circuit breakers and improve the stability and reliability of high-voltage circuit breakers. Input the detection data of the high-voltage circuit breaker into the fault analysis and judgment neural network model obtained in step D). If the fault analysis and judgment neural network model outputs a fault type, a fault alarm is issued. The data close to the fault can be used as sample data for fault analysis and judgment, and thus used to train the fault analysis and judgment neural network model.

Example 2:

This embodiment makes further improvements based on Embodiment 1. As shown in Figure 8, the method for judging whether the difference between the detection data and the historical detection data is greater than the preset threshold includes: D41) Adding the detection data and the historical detection data to The state quantity is converted into a Boolean quantity, and {0,1} is used to represent false and true respectively; D42) Normalize the numerical quantities in the detection data and historical detection data to obtain the normalized historical detection data. The minimum and maximum values of each item are arranged in a set order. The detection data is sorted to form a detection vector. The minimum values of each item of historical detection data are sorted to form a historical detection left vector. Historical detection The maximum values of the data are sorted to form the historical detection right vector; D43) respectively calculate the distance between the detection vector and the historical detection left vector and the historical detection right vector, and compare it with the preset distance threshold. If the detection vector and the historical detection left vector and If the distances of the historical detection right vectors are all greater than the preset distance threshold, then it is determined that the distance between the detection data corresponding to the detection vector and the historical detection data is greater than the preset threshold. On the contrary, it is determined that the distance between the detection data corresponding to the detection vector and the historical detection data is not greater than Preset threshold.

Embodiment three:

This embodiment makes further improvements on the basis of Embodiment 1. As shown in Figure 9, the method for judging whether the difference between the detection data and the historical detection data is greater than the preset threshold includes: D51) Adding the detection data and the historical detection data to The numerical quantity is processed in segments, with the segmented interval as the name, and the numerical quantity is converted into a state quantity; D52) The state quantity in the detection data and historical detection data is converted into a Boolean quantity, and {0,1} is used to represent false respectively. Hezhen; D52) Delete the Boolean quantities with different values in the processed historical detection data, sort the remaining Boolean quantities in the historical detection data according to the settings, respectively constitute historical detection vectors, and select the Boolean quantities in the detection data The Boolean quantity corresponding to the historical detection vector is sorted according to the setting to form the detection vector; D53) Calculate the distance between the detection vector and the historical detection vector, and compare it with the preset distance threshold. If the distance is greater than the preset distance threshold, the detection is determined The distance between the detection data corresponding to the vector and the historical detection data is greater than the preset threshold. Otherwise, it is determined that the distance between the detection data corresponding to the detection vector and the historical detection data is not greater than the preset threshold. By constructing a vector of Boolean type, the deviation caused by numerical data can be eliminated. Deleting Boolean quantities with different values and retaining only Boolean quantities with consistent values can eliminate the influence of irrelevant factors and make the judgment results more valuable for reference.

In step D51), the method of segmenting the numerical quantities in the detection data and historical detection data includes: D511) selecting a numerical quantity, obtaining all the values of the numerical quantity in the historical detection data, in order of numerical value. The arrangement is recorded as the set Ki, the minimum value in the set Ki is k _min and the maximum value is k _max ; D512) Assign the initial value of the partition starting point k _s as k _min , and assign the initial value of the partition end point k _e as k _max , and examine the values k _m =k _s +n×Δk, Δk is the artificially set step size, n is a positive integer, and the initial value of n is 1. The setting method of step size Δk includes: calculating the pairwise difference of numerical data in the set Ki value, eliminate the difference that is zero, perform absolute value calculation on the remaining difference, and use the minimum value as the step size Δk to participate in the calculation.

;D513) n keeps adding 1 to itself, if the inspection value k _m satisfies the following conditions:

Among them, the function N(x, y) represents the set Ki and the number of data whose data values are in the numerical interval (x, y). Then (2k m -k s) is used as the interval dividing point and added to the dividing point set Km, and (2k _m -k _s ) is added to the dividing point set Km. 2k _m -k _s ) is assigned to k _s , and n continues to increase by 1 until k _m > k _max ; D514) Add k _min and k _max to the set Km, and use the values within Km as the dividing point, Divide the numerical quantity data into numerical intervals; D515) Select the next numerical quantity and repeat steps D511) to D514) until all numerical quantities are divided into interval segments; D516) The detection data adopts the corresponding numerical quantity in the historical detection data. Interval division. Segmentation based on the aggregation characteristics of the value itself can make the segmentation closer to the different states of the value.

In step D51), with the segmented interval as the name, the method of converting the numerical quantity into the state quantity includes the following steps: D511) Divide the numerical quantity data into several intervals, [n _m(1) , n _m(2) ] , [n _m(2) , n _m(3) ]...[n _m(k-1) , n _m(k) ], where n _m(1) and n _{m(k) are the starting point and n m(k)} of the numerical interval respectively. The end point, n _m(2) ~n _m(k-1) is the middle dividing point of the numerical interval, and Respectively as the state name of the corresponding numerical interval; D512) If the data of the historical detection numerical quantity falls into the interval [n _m(d) , n _m(d+1) ], d∈[1, k-1], then Status name/> As the value of this numerical quantity, the conversion of numerical quantity data into state quantity data is completed. It can quickly convert numerical quantities into status quantities.

The method of converting the status quantities in the detection data and historical detection data into Boolean quantities includes the following steps: D521) Obtain all status values of the status quantity data. If the status quantity data is the status of the circuit breaker itself, then all status values Includes all possible values of the state; if the state quantity data is state quantity data converted from numerical quantity data, all state values only include values that have appeared in historical states. ; D522) Use the status value as the field name to split the status field into multiple fields; D523) Set the fields with the same field name as the status data value to 1, and set the remaining split fields to 0 to complete the split of the status data. Divided into Boolean data. Splitting the state quantity into Boolean quantities can speed up the training efficiency of the neural network.

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 (15)

1. A life cycle assessment and fault early warning method for a high-voltage circuit breaker is characterized in that,

the method comprises the following steps:

a) Acquiring historical maintenance detection data of the high-voltage circuit breakers of the same type;

b) Acquiring a high-voltage circuit breaker of the same type, manually setting a fault source under laboratory conditions, enabling the high-voltage circuit breaker to continuously perform electrified switching-on and switching-off actions under the condition that the fault source exists until faults occur, detecting periodically while the times of the electrified switching-on and switching-off actions of the high-voltage circuit breaker before the faults are N, and taking periodic detection data as reference data and correlating the periodic detection data with the times of the electrified switching-on and switching-off actions corresponding to the tests;

c) Comparing the periodic detection data of the high-voltage circuit breaker with the reference data to obtain the number of times N corresponding to the reference data closest to the periodic detection data, and taking N/N as a life cycle evaluation result of the high-voltage circuit breaker;

d) Correlating the reference data with the type of the faults to form sample data, and training a fault early warning neural network model by using the sample data;

Step D) further comprises: the detection data of M times before the occurrence of the fault is associated with the corresponding fault type to form fault sample data, and the fault sample data is used for training a fault judging neural network model; using detection data of the times (N-M) before occurrence of faults, correlating with the types of the faults, and forming a sample data training fault early warning neural network model; step E further comprises: inputting the detection data of the high-voltage circuit breaker into the fault judging neural network model obtained in the step D), and giving out fault alarm if the fault judging neural network model outputs the fault type;

the method for establishing the high-voltage circuit breaker fault research model comprises the following steps:

d21 All detection data are obtained, and the detection data are associated with the corresponding fault types to be used as sample data;

d22 Preprocessing, normalizing and training the sample data, and taking the trained neural network model as a fault studying and judging model;

in step D21), the method of associating the detection data with the corresponding fault type includes:

d211 Obtaining the detection data in step B11) as historical detection data;

d212 Comparing the detection data obtained by the detection in the step B12) and the step B13) with the historical detection data in sequence, and if the difference between the detection data and the historical detection data is larger than a preset threshold value, associating the detection data with the maintenance requirement which does not reach the standard;

D213 Comparing the plurality of groups of detection data obtained by the plurality of mechanical characteristic tests in the step B14) with the historical detection data, and if the difference between the detection data and the historical detection data is larger than a preset threshold value, associating the group of detection data with the jamming fault of the mechanical part; the method for judging whether the difference between the detection data and the historical detection data is larger than a preset threshold value comprises the following steps:

d31 Carrying out segmentation processing on the numerical value in the detection data and the historical detection data, and converting the numerical value into a state quantity by taking a segmentation interval as a name;

d32 State quantities in the detection data and the history detection data are converted into boolean quantities, and {0,1} is used to represent false and true, respectively; d32 The Boolean amount of the processed historical detection data is regarded as a numerical value to be averaged, the average value is rounded to be an integer, the obtained integer is regarded as the Boolean amount again, and the processed detection data and the historical detection data are sequenced according to the setting to respectively form a detection vector and a historical detection vector;

d33 Calculating the distance between the detection vector and the historical detection vector, comparing the distance with a preset distance threshold, if the distance is larger than the preset distance threshold, judging that the difference between the detection data corresponding to the detection vector and the historical detection data is larger than the preset threshold, otherwise, judging that the difference between the detection data corresponding to the detection vector and the historical detection data is not larger than the preset threshold;

E) Inputting the detection data of the high-voltage circuit breaker to be evaluated and pre-warned into the fault pre-warning neural network model obtained in the step D), and if the fault pre-warning neural network model outputs a fault type, sending out fault pre-warning, wherein the corresponding fault is the fault type output by the fault pre-warning neural network model.

2. The method for evaluating the life cycle and pre-warning faults of a high voltage circuit breaker according to claim 1, wherein,

in the step B), the method for artificially setting the fault source comprises the following steps:

b11 Detecting the high-voltage circuit breaker for a plurality of times;

b12 According to the maintenance requirement of the high-voltage circuit breaker, sequentially selecting one maintenance requirement to ensure that the maintenance requirement does not reach the standard, and detecting for a plurality of times after carrying out a plurality of electrified opening and closing actions;

b13 Sequentially selecting two maintenance requirements to ensure that the maintenance requirements do not reach the standard, and detecting for a plurality of times after carrying out a plurality of charged opening and closing actions; b14 Using liquid nitrogen or dry ice to rapidly cool the high-voltage breaker, performing mechanical property tests for a plurality of times, and obtaining detection data of the mechanical property tests.

3. The method for evaluating the life cycle and pre-warning faults of a high voltage circuit breaker according to claim 1, wherein,

the method for judging whether the difference between the detection data and the historical detection data is larger than a preset threshold value comprises the following steps:

D41 State quantities in the detection data and the history detection data are converted into boolean quantities, and {0,1} is used to represent false and true, respectively; d42 Normalizing the detection data and the numerical value in the history detection data to obtain the minimum value and the maximum value of each item of the normalized history detection data respectively, arranging the processed Boolean quantity and the numerical value according to a set sequence, ordering the detection data to form a detection vector, ordering each minimum value of the history detection data to form a history detection left vector, and ordering each maximum value of the history detection data to form a history detection right vector;

d43 The distance between the detection vector and the historical detection left vector as well as the distance between the detection vector and the historical detection right vector are calculated respectively, compared with a preset distance threshold, if the distance between the detection vector and the historical detection left vector as well as the distance between the detection vector and the historical detection right vector are larger than the preset distance threshold, the difference between the detection data corresponding to the detection vector and the historical detection data is judged to be larger than the preset threshold, otherwise, the difference between the detection data corresponding to the detection vector and the historical detection data is judged to be not larger than the preset threshold.

4. The method for evaluating the life cycle and the fault pre-warning of the high voltage circuit breaker according to claim 1, wherein the method for judging whether the difference between the detected data and the historical detected data is larger than a preset threshold value comprises the following steps:

D51 Carrying out segmentation processing on the numerical value in the detection data and the historical detection data, and converting the numerical value into a state quantity by taking a segmentation interval as a name;

d52 State quantities in the detection data and the history detection data are converted into boolean quantities, and {0,1} is used to represent false and true, respectively; d52 Deleting the Boolean amounts with different values in the processed historical detection data, sorting the residual Boolean amounts of the historical detection data according to the setting to respectively form historical detection vectors, selecting the Boolean amounts corresponding to the historical detection vectors in the detection data, and sorting according to the setting to form detection vectors;

d53 Calculating the distance between the detection vector and the historical detection vector, comparing the distance with a preset distance threshold, if the distance is larger than the preset distance threshold, judging that the difference between the detection data corresponding to the detection vector and the historical detection data is larger than the preset threshold, otherwise, judging that the difference between the detection data corresponding to the detection vector and the historical detection data is not larger than the preset threshold.

5. The method for life cycle evaluation and fault pre-warning of high voltage circuit breaker according to claim 4, wherein,

in step D51), the method for performing the segmentation processing on the numerical value amounts in the detection data and the history detection data includes:

D511 Selecting a numerical value to obtain all numerical values of the numerical value in the historical detection data, and sequentially arranging the numerical values according to the numerical values, wherein the numerical values are marked as a set Ki, and the minimum value in the set Ki is k _min And a maximum value of k _max ；

D512 To partition start point k _s Giving an initial value of k _min Partition endpoint k _e Giving an initial value of k _max Let the value k be examined _m ＝k _s +n×Δk, Δk is a step size manually set, n is a positive integer, and n is 1 as an initial value;

d513 N is continuously added with 1, if the value k is examined _m The following conditions are satisfied:

wherein the function N (x, y) represents the set Ki, and the number of data values in the numerical interval (x, y) is (2 k) _m -k _s ) As the interval dividing points and adding the dividing point set Km, the value of (2 k _m -k _s ) Assignment of the value of k to _s Continuously adding 1 to n until k _m >k _max ；

D514 To k) _min And k _max Adding a set Km, and dividing the numerical value data into numerical value intervals by using values in the Km as dividing points;

d515 Selecting the next numerical value, and repeating the steps D511) to D514) until all the numerical values are divided into interval segments;

d516 The detection data is divided into sections of a numerical value corresponding to the history detection data.

6. The method for evaluating the life cycle and early warning faults of a high voltage circuit breaker according to claim 4 or 5, wherein in the step D51), the method for converting the numerical value into the state value by using the section as the name comprises the following steps:

D511 Dividing the numerical data into a plurality of intervals, [ n ] _m(1) ,n _m(2) ],[n _m(2) ,n _m(3) ]…[n _m(k-1) ,n _m(k) ]Wherein n is _m(1) And n _m(k) Respectively a starting point and an ending point of a numerical value interval, n _m(2) ～n _m(k-1) Dividing the middle of the numerical interval into pointsRespectively used as state names of corresponding numerical value intervals;

d512 If the data of the historical detection numerical value falls into the interval [ n ] _m(d) ,n _m(d+1) ],d∈[1,k-1]The state name is thenAs the value of the numerical quantity, conversion of the numerical quantity data into state quantity data is completed.

7. The method for evaluating life cycle and early warning faults of a high voltage circuit breaker according to claim 4 or 5, wherein in the step D52), the method for converting the state quantity in the detection data and the historical detection data into the boolean quantity comprises the following steps: d521 Obtaining all state values of the state quantity data;

d522 Splitting the state quantity field into a plurality of fields by taking the state value as a field name;

d523 Setting the field with the same field name and state quantity data value as 1, setting the other split fields as 0, and completing the splitting of the state quantity data into Boolean quantity data.

8. The method for evaluating the life cycle and early warning faults of a high voltage circuit breaker according to claim 4 or 5, wherein in the step D512), the step size Δk setting method comprises the steps of: and calculating the value data of the numerical value in the set Ki, removing the value data of the numerical value data of the set Ki from the value data of the numerical value data to obtain the value data of the numerical value data, performing absolute value operation on the residual value data, taking the minimum value as the step delta k, and participating in calculation.

9. The method for evaluating the life cycle and pre-warning faults of a high voltage circuit breaker according to claim 7, wherein,

in step D521), the method for obtaining all state values of the state quantity data includes: if the state quantity data is the state of the breaker, all state values comprise all possible values of the state; if the state quantity data is the state quantity data converted from the numerical quantity data, all state values only comprise the values which appear in the history state.

10. The method for evaluating the life cycle and pre-warning faults of a high voltage circuit breaker according to claim 1, wherein,

the detection data comprise closing time, opening time, closing speed, opening speed, three-phase different periods, same-phase different periods, golden short time, no-flow time, moving contact maximum speed, moving contact average speed, moving contact action time, bouncing times, bouncing maximum amplitude, opening and closing stroke, opening and closing process current waveform curve, time speed and stroke dynamic curve in the opening and closing stroke of the moving contact, and contact resistance.

11. The method for evaluating the life cycle of a high voltage circuit breaker and pre-warning faults according to claim 1 or 2, wherein in the step D, before training the fault diagnosis neural network model by using the fault sample data, the normalization processing is performed on the fault sample data, which comprises:

D11 The method comprises the steps of) listing numerical data in fault sample data, obtaining a theoretical boundary value of the numerical data as a boundary value, if the theoretical boundary value does not exist, obtaining a historical boundary value of the data as the boundary value, wherein a left boundary value of the boundary value is regarded as 0, a right boundary value of the boundary value is regarded as 1, and dividing a difference of the numerical data minus the left boundary value by a difference of the right boundary value and the left boundary value to obtain a normalized value of the numerical data;

d12 Splitting the state quantity data into a plurality of boolean data;

d13 Boolean data into numerical values and normalized.

12. The method for evaluating life cycle and early warning faults of a high voltage circuit breaker according to claim 11, wherein in the step D12), the method for splitting the state quantity data into a plurality of boolean data comprises:

d121 Obtaining all state values of the state quantity data;

d122 Splitting the state quantity field into a plurality of fields by taking the state value as a field name;

d123 Setting the field with the same field name and state quantity data value, setting the other split fields to zero, and completing the splitting of the state quantity data into Boolean quantity data.

13. A method for evaluating life cycle and early warning faults of a high voltage circuit breaker according to claim 2 or 3, wherein in the step B), before the high voltage circuit breaker is tested, a non-contact displacement sensor is installed on each mechanical moving part of the high voltage circuit breaker, and displacement data measured by the non-contact displacement sensor is added into the test data of the high voltage circuit breaker.

14. The method for evaluating the life cycle and early warning faults of the high-voltage circuit breaker according to claim 13, wherein in the step E), a non-contact displacement sensor is installed on each mechanical moving part of the high-voltage circuit breaker to be evaluated and early warned, displacement data measured by the non-contact displacement sensor are obtained, the state data of the high-voltage circuit breaker to be evaluated and early warned and the displacement data measured by the non-contact displacement sensor are taken as inputs of a fault judging neural network model, the fault judging neural network model is trained, and fault judging is carried out.

15. The method for evaluating the life cycle and early warning faults of the high-voltage circuit breaker according to claim 13, wherein in the step B), a non-contact displacement sensor is arranged on each mechanical moving part of the normal high-voltage circuit breaker, the switching-on and switching-off test is continuously repeated on the high-voltage circuit breaker under the power-off condition until the mechanical parts of the high-voltage circuit breaker are damaged, the switching-on and switching-off times K in the test process and the displacement data of each mechanical moving part in the switching-on and switching-off process are recorded as historical displacement data;

in step E), if the fault judging result of the high-voltage circuit breaker to be judged is no fault, installing a non-contact displacement sensor on each mechanical moving part of the high-voltage circuit breaker to be judged, performing one-time opening and closing of the high-voltage circuit breaker to be judged to obtain displacement data measured by the non-contact displacement sensor, comparing the displacement data with the historical displacement data to obtain the opening and closing test times K corresponding to the closest historical displacement data, and taking (K-K) as the residual service life of the high-voltage circuit breaker to be judged.