Fault risk assessment method and device for disconnecting switch, equipment and storage medium
By applying the sequence Monte Carlo method and particle filtering technology in the isolation switch fault risk assessment, the problem that the existing technology is difficult to deal with high-dimensional and nonlinear systems is solved, and a more accurate and reliable fault risk assessment is achieved.
Patent Information
- Application Number
- CN202411740026.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-29
- Publication Date
- 2025-03-21
- Estimated Expiration
- 2044-11-29
AI Technical Summary
The existing isolation switch failure risk assessment methods are difficult to deal with high-dimensional, multivariable and nonlinear systems, resulting in inaccurate and reliable evaluation results, and it is difficult to deal with uncertainty and randomness problems.
The sequential Monte Carlo method is used, based on particle filtering technology, using historical operation data and real-time collected data, fault prediction is carried out through state transfer model and likelihood function, and fault prediction results are generated.
It improves the accuracy of fault prediction results, can handle nonlinear and dynamic changes of the isolating switch mechanical system more accurately, and enhances the reliability of fault risk assessment.
Smart Images

Figure CN119669973A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of fault detection, and in particular to a fault risk assessment method and device, equipment and storage medium for an isolating switch. Background Art
[0002] Isolators play an important role in protection and isolation in power systems, and their failure may lead to serious power system accidents. As the scale and complexity of power systems continue to increase, the reliability and safety of isolators become particularly important. Based on this, risk assessment methods can be used to monitor isolators to ensure their normal operation.
[0003] Traditional risk assessment methods include qualitative assessment and quantitative assessment. Qualitative assessment relies on the subjective judgment of experts, is easily affected by human factors, and lacks scientific basis; quantitative assessment is usually based on methods such as statistical analysis and fault tree analysis, but these methods are often unable to cope with high-dimensional, multivariable and nonlinear systems. Therefore, existing risk assessment methods are usually based on historical data and expert experience, which makes it difficult to deal with uncertainty and randomness, making the assessment results inaccurate and unreliable. Summary of the invention
[0004] Based on this, it is necessary to propose a fault risk assessment method, device, equipment and storage medium for disconnectors to address the above problems in order to improve the accuracy of fault risk assessment results.
[0005] To achieve the above-mentioned object, the present application provides, in a first aspect, a method for evaluating the fault risk of an isolating switch, the method comprising:
[0006] Obtain historical operating data of the disconnector;
[0007] Determine, based on the historical operation data, a priori distribution that the operation data of the disconnector obeys in an initial state;
[0008] Collecting actual operation data of the isolating switch in real time;
[0009] Using the Sequential Monte Carlo method, fault prediction is performed according to the prior distribution and the actual operation data to obtain a fault prediction result of the disconnector;
[0010] The fault prediction result is compared with a preset fault threshold to determine whether the isolating switch is currently faulty.
[0011] Furthermore, the sequential Monte Carlo method is used to perform fault prediction according to the prior distribution and the actual operation data to obtain the fault prediction result of the disconnector, which specifically includes:
[0012] Randomly draw N particle samples from the prior distribution;
[0013] All particle samples are updated using the state transition model to generate the predicted operation data corresponding to each particle sample;
[0014] A fault analysis is performed based on the gaps between all the predicted operating data and the actual operating data to obtain a fault prediction result.
[0015] Furthermore, the state transition model is represented by the following formula:
[0016] x t =f(x t-1 )+v t
[0017] In the formula, x t is the predicted running data corresponding to the particle sample at time t, f(x t-1 ) is the preset system function of the isolating switch, v t is the preset process noise.
[0018] Further, the fault analysis is performed according to the gap between all the predicted operation data and the actual operation data to obtain the fault prediction result, which specifically includes:
[0019] Using a preset likelihood function, a target weight of each particle sample is obtained by performing calculations based on each of the predicted operation data and the actual operation data;
[0020] The sum of the target weights of all particle samples is calculated to obtain a total weight; the total weight is the fault prediction result.
[0021] Furthermore, the weight calculation formula of the particle sample is as follows:
[0022]
[0023] In the formula, is the target weight of the i-th particle sample, y t is the actual operation data, is the i-th particle sample, is the predicted running data corresponding to the i-th particle sample, σ w is the preset discreteness of the observation noise distribution.
[0024] Furthermore, the use of a preset likelihood function to calculate according to each of the predicted operation data and the actual operation data to obtain a target weight for each particle sample specifically includes:
[0025] Based on a preset likelihood function, calculation is performed according to the predicted operation data and the actual operation data to obtain an initial weight of each particle sample;
[0026] The effective number of particles is calculated based on the initial weights of all particle samples;
[0027] When the effective particle number is less than the preset particle number threshold, resampling the particle samples to generate new particle samples, and continuing to perform the operation of updating all particle samples using the state transition model to generate predicted operation data corresponding to each particle sample until the effective particle number is not less than the preset particle number threshold;
[0028] When the effective particle number is not less than the preset particle number threshold, the initial weight is used as the target weight, and the step of calculating the sum of the target weights of all particle samples to obtain the total weight is continued.
[0029] Furthermore, the effective number of particles is calculated by the following formula:
[0030]
[0031] In the formula, is the target weight of the i-th particle sample, and N is the total number of particle samples.
[0032] To achieve the above-mentioned purpose, the second aspect of the present application provides a fault risk assessment device for an isolating switch, the device comprising a data acquisition module, a data analysis module and a fault prediction module;
[0033] The data acquisition module is used to obtain historical operation data of the disconnector;
[0034] Determine, based on the historical operation data, a priori distribution that the operation data of the disconnector obeys in an initial state;
[0035] Collecting actual operation data of the isolating switch in real time;
[0036] The data analysis module is used to use the sequential Monte Carlo method to perform fault prediction according to the prior distribution and the actual operation data to obtain a fault prediction result of the disconnector;
[0037] The fault prediction module is used to determine whether the isolating switch is currently faulty based on the comparison between the fault prediction result and a preset fault threshold.
[0038] To achieve the above-mentioned purpose, the third aspect of the present application provides a computer-readable storage medium storing a computer program, characterized in that when the computer program is executed by a processor, the processor executes the steps of the method described in the first aspect.
[0039] To achieve the above-mentioned purpose, the fourth aspect of the present application provides a computer device, including a memory and a processor, characterized in that the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the method described in the first aspect.
[0040] The embodiments of the present invention have the following beneficial effects:
[0041] The embodiment of the present invention proposes a fault risk assessment method for an isolating switch, the method comprising: obtaining historical operating data of the isolating switch; determining the prior distribution that the operating data of the isolating switch obeys in the initial state based on the historical operating data; collecting the actual operating data of the isolating switch in real time; using the sequential Monte Carlo method to perform fault prediction based on the prior distribution and the actual operating data to obtain a fault prediction result of the isolating switch; and comparing the fault prediction result with a preset fault threshold to determine whether the isolating switch is currently faulty. The present invention adopts the sequential Monte Carlo method and effectively processes the nonlinear and non-Gaussian distribution of the mechanical system of the isolating switch based on the particle filtering technology to improve the accuracy of the fault prediction result. In addition, the system state is updated based on the time series, and the fault risk assessment result obtained by analyzing the updated system state is more reliable. BRIEF DESCRIPTION OF THE DRAWINGS
[0042] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0043] in:
[0044] Figure 1 A schematic diagram of a flow chart of a method for evaluating the fault risk of an isolating switch according to an embodiment of the present invention;
[0045] Figure 2 It is a structural block diagram of a fault risk assessment device for an isolating switch according to an embodiment of the present invention;
[0046] Figure 3 1 is a diagram showing the internal structure of a computer device in an embodiment of the present invention. DETAILED DESCRIPTION
[0047] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0048] Failures in the mechanical system of disconnectors usually involve multiple physical processes, such as increased friction, wear, poor lubrication, etc. The mechanical resistance changes caused by these failure factors are not linear. Traditional fault assessment methods are difficult to detect the nonlinear changes caused by mechanical system failures, and thus it is difficult to conduct accurate fault analysis.
[0049] Based on this, the embodiment of the present invention proposes a fault risk assessment method for disconnectors based on the sequential Monte Carlo method, which can be referred to Figure 1 , Figure 1 The figure is a flow chart of a method for fault risk assessment of an isolating switch in an embodiment of the present invention, and the method comprises:
[0050] Step 110: Obtain historical operation data of the disconnector.
[0051] In the embodiment of the present invention, before diagnosing the fault of the mechanical system of the disconnector, it is necessary to evaluate the initial state of the mechanical system. Specifically, historical operation data of the mechanical system of the disconnector is obtained to evaluate its initial state based on the historical operation data.
[0052] The mechanical system of the isolating switch involves complex dynamics, especially the interaction with the drive motor, and the drive motor power will change greatly with the change of the mechanical state. Therefore, in the embodiment of the present invention, the operating data can be the drive motor power of the isolating switch, so as to perform a fault risk assessment on the mechanical system of the isolating switch based on the drive motor power of the isolating switch.
[0053] Step 120: Determine the prior distribution that the operating data of the disconnector obeys in the initial state based on the historical operating data.
[0054] In the embodiment of the present invention, the initial state of the operating data of the isolating switch obeys a certain prior distribution. Therefore, the prior distribution obeyed by the operating data of the isolating switch in the initial state can be obtained by analyzing the historical operating data of the isolating switch in the initial state. The prior distribution obeyed by the operating data of the isolating switch in the initial state can be expressed by the following formula:
[0055] x0~p(x0)
[0056] In the formula, x0 is the particle sample under the initial state, and p(x0) is the prior distribution obeyed by the running data under the initial state.
[0057] Step 130: Collect actual operation data of the disconnector in real time.
[0058] Specifically, through real-time or online monitoring applications, the operating data of the disconnector, such as the real-time power of the drive motor, is collected in real time. By continuously tracking and updating the system status, the fault prediction and update of the system status of the disconnector are carried out in a timely manner.
[0059] Step 140: Using the Sequential Monte Carlo method, fault prediction is performed according to the prior distribution and actual operation data to obtain a fault prediction result of the disconnector.
[0060] Taking into account the influence of adverse factors that cause the disconnector to malfunction, such as increased friction and mechanical resistance, the operating data of the disconnector will change abnormally. For example, when insufficient lubricant causes increased friction, the power of the drive motor of the disconnector may show a nonlinear increase instead of a linear increase; when mechanical components are stuck, the power output of the drive motor of the disconnector may fluctuate sharply, and this fluctuation is not necessarily a linear relationship. It can be seen that the complex dynamic behavior of the mechanical system makes the relationship between the power characteristics and the fault state nonlinear. Therefore, the embodiment of the present invention uses a sequential Monte Carlo method to process the nonlinear and non-Gaussian distribution of the operating data of the disconnector (i.e., the power of the drive motor) caused by the fault state, thereby enhancing the ability to process the nonlinear and dynamic systems of the disconnector, thereby improving the accuracy of the fault prediction results.
[0061] Step 150: Determine whether the disconnector is currently faulty based on the comparison between the fault prediction result and the preset fault threshold.
[0062] In an embodiment of the present aspect, a fault threshold is preset, which represents the minimum fault prediction result of a fault. When the fault prediction result is less than the fault threshold, it is considered that the isolating switch has not failed; when the fault prediction result is not less than the fault threshold, it is considered that the isolating switch has failed. At this time, a fault warning is triggered so that the fault can be handled in time.
[0063] The present invention utilizes the sequential Monte Carlo method and particle filtering technology to effectively process the nonlinear and non-Gaussian distribution of the disconnector mechanical system to improve the accuracy of the fault prediction results. In addition, the system state is updated based on the time series, and the fault risk assessment result obtained by analyzing the updated system state is more reliable.
[0064] In one embodiment of the present invention, step 140, using the Sequential Monte Carlo method to perform fault prediction according to the prior distribution and the actual operation data, to obtain the fault prediction result of the disconnector, specifically includes:
[0065] Step 410, randomly select N particle samples from the prior distribution.
[0066] In an embodiment of the present invention, each particle in the prior distribution represents a possible initial state of the disconnector mechanical system, and N particle samples are randomly selected from the prior distribution to gradually update and evaluate the state of the mechanical system based on parallel simulation of multiple particle samples.
[0067] The particle sample extraction process is expressed by the following formula:
[0068]
[0069] In the formula, is the i-th particle sample in the initial state, and t=0 is the initial state.
[0070] Step 420, using the state transfer model to update all particle samples, and generate the predicted operation data corresponding to each particle sample.
[0071] In the embodiment of the present invention, considering the influence of time series on particle state changes, a nonlinear state transition model is introduced, and the current state of the particle sample is updated based on the state transition model to obtain the predicted operation data corresponding to each particle sample at different times. By processing the correlation in time series, the system states at different time points can influence and update each other, so as to improve the fault prediction accuracy of the disconnector mechanical system.
[0072] Specifically, the change of the particle sample over time t can be described by a state transition model, which reflects the change of the mechanical system from time t-1 to t. In one embodiment of the present invention, the state transition model can be expressed by the following state transition equation:
[0073] x t =f(x t-1 )+v t
[0074] In the formula, x t is the particle state at time t, f(x t-1 ) is the system function of the preset disconnect switch, v t is the preset process noise, which follows a normal distribution. Used to model the uncertainty of mechanical systems.
[0075] Particle prediction is achieved by updating particle samples through the state transition model. Each particle sample generates a new predicted state according to the state transition equation, thereby simulating the possible state of the system at time t. The predicted operation data corresponding to each particle sample at different times is generated according to the updated system state.
[0076] The formula for predicting the state of a particle can be expressed as:
[0077]
[0078] In the formula, is the particle state of the ith particle at time t, System function for the preset disconnect switch, is the preset process noise, which follows a normal distribution.
[0079] The embodiment of the present invention reflects the change law of the state of the mechanical system over time through the state transfer equation, and takes into account the factor of random disturbance, so as to obtain more accurate predicted operation data under the current state.
[0080] Step 430, perform fault analysis based on the gap between all predicted operation data and actual operation data to obtain fault prediction results.
[0081] Under ideal conditions, the predicted operating data and the observed actual operating data should be the same or similar. However, if a mechanical failure occurs in the disconnector, there will be an additional difference between the observed actual operating data and the predicted operating data. Therefore, by analyzing the gap between the predicted operating data and the actual operating data, it is determined whether the disconnector has a fault.
[0082] In one embodiment of the present invention, Step 430, performing fault analysis based on the gap between all predicted operation data and actual operation data to obtain a fault prediction result, specifically includes:
[0083] Step 431, using the preset likelihood function, calculate according to each predicted operation data and actual operation data to obtain the target weight of each particle sample.
[0084] In an embodiment of the present invention, an observation model is introduced to map the current state of particle samples into the observation space for observation. The observation model describes the relationship between the system state under normal conditions and the actual operating state, that is, the relationship between the predicted operating data and the actual operating data.
[0085] When constructing the observation model, the errors and external noise of the sensors that collect the operating data are taken into account. The observation equation can be expressed as:
[0086] y t =□(x t)+w t
[0087] Among them, y t is the actual operation data, □(x t ) is the predicted operation data, w t is the observation noise, which follows a normal distribution.
[0088] The observation equation is used to compare the predicted operation data with the actual operation data to determine the rationality of the particles. It can be understood that when the system fails, the relationship between the predicted operation data and the actual operation data cannot satisfy the above observation equation. Therefore, the error between the actual operation data and the predicted operation data is analyzed to determine whether the mechanical system fails.
[0089] Specifically, the state of the particle sample is corrected according to the actual operation data actually collected. First, a target weight is assigned to each particle sample. The target weight is calculated based on the likelihood function of the actual operation data and the predicted operation data. The weight of the particle sample represents the degree of match between the predicted operation data and the actual operation data. The higher the weight, the closer the predicted operation data corresponding to the particle sample is to the actual operation data.
[0090] In one embodiment of the present invention, the weight calculation formula of the particle sample is as follows:
[0091]
[0092] In the formula, is the target weight of the i-th particle sample, y t For actual operation data, is the i-th particle sample, is the predicted running data corresponding to the i-th particle sample, σ w is the preset discreteness of the observation noise distribution.
[0093] In another embodiment of the present invention, in order to prevent the target weight of the particle sample from being overly concentrated on a few particles and causing particle degradation, an adaptive resampling strategy can be introduced. When the particle sample does not meet the preset conditions, the particle sample is resampled to improve the quality of the particle sample. Based on this, Step 431, using the preset likelihood function, calculates according to each predicted operation data and actual operation data to obtain the target weight of each particle sample, specifically including:
[0094] Step 4311, based on the preset likelihood function, calculate according to the predicted operation data and the actual operation data to obtain the initial weight of each particle sample.
[0095] In the embodiment of the present invention, the initial weight of each particle sample is calculated based on the predicted operation data and the actual operation data. The initial weight can be calculated by the following formula:
[0096]
[0097] In the formula, is the initial weight of the i-th particle sample, y t For actual operation data, is the i-th particle sample, is the predicted running data corresponding to the i-th particle sample, σ w is the preset discreteness of the observation noise distribution.
[0098] Step 4312, calculate the effective number of particles based on the initial weights of all particle samples.
[0099] Specifically, the effective sample size (ESS) reflects the diversity of the particle sample set. The larger the effective particle number, the more uniform the particle sample weights and the higher the particle quality, and the more accurate the results obtained based on the particle sample set analysis. The smaller the effective particle number, the more weights are concentrated on a few particles. At this time, the particle quality is lower, and the results obtained based on the particle sample set analysis are less robust.
[0100] In one embodiment of the present invention, the effective particle number is calculated by the following formula:
[0101]
[0102] In the formula, is the target weight of the i-th particle sample, and N is the total number of particle samples.
[0103] Step 4313, when the number of valid particles is less than the preset particle number threshold, resample the particle samples to generate new particle samples, and continue to use the state transition model to update all particle samples to generate the predicted running data corresponding to each particle sample until the number of valid particles is not less than the preset particle number threshold; when the number of valid particles is not less than the preset particle number threshold, use the initial weight as the target weight, and continue to calculate the sum of the target weights of all particle samples to obtain the total weight.
[0104] Specifically, when the number of valid particles is lower than the preset particle number threshold, a resampling operation needs to be performed. New particle samples are resampled from high-weight particle samples. By introducing an adaptive resampling strategy, whether to resample is determined based on the dynamic changes in the number of valid particles. While preventing particle degradation, unnecessary calculations are avoided, so that computing resources can be more concentrated on high-weight particles, thereby effectively reducing the number of invalid particles, greatly improving computing efficiency, and improving the real-time performance of fault risk assessment.
[0105] The process of extracting new particle samples can be referred to the following formula:
[0106]
[0107] In the formula, are the i-th and j-th particle samples respectively, is the Dirac function.
[0108] After re-extracting and generating new particle samples, repeat the steps of Step 420-Step 430 until the number of valid particles is not less than the particle number threshold. The initial weights of all particle samples at this time are used as the target weights so that subsequent fault risk assessment can be performed based on the target weights.
[0109] By evaluating the effectiveness of particle samples and gradually updating particles, the prediction of failure rate can be dynamically adjusted to ensure the diversity of particle sample sets, so that subsequent analysis can continue to accurately capture the system status. Because the occurrence of disconnector system failures may be gradual and the system status will change over time and operating conditions, the ability to update particles in real time is particularly important for mechanical failure prediction.
[0110] Step 432, calculate the sum of the target weights of all particle samples to obtain the total weight; the total weight is the fault prediction result.
[0111] In the embodiment of the present invention, whether a mechanical failure occurs in the isolating switch can be determined based on the sum of the target weights of all particle samples. First, the total weight of all particle samples is obtained based on the sum of the target weights of all particle samples; second, the total weight is compared with the preset fault threshold. When the total weight is less than the preset fault threshold, it indicates that the isolating switch has not mechanically failed; when the total weight is not less than the preset fault threshold, it indicates that a mechanical failure occurs in the isolating switch. At this time, the system should trigger a fault warning so as to make timely fault adjustments to the isolating switch.
[0112] The adaptive resampling strategy introduced in the embodiment of the present invention effectively avoids the particle degradation problem by adjusting the weight distribution of particles, improves the accuracy of the fault diagnosis process, and can also reduce unnecessary resampling operations if the particles meet the conditions, thereby improving the computing efficiency and achieving more real-time fault assessment.
[0113] The present invention also provides a fault risk assessment device for an isolating switch, which can be referred to Figure 2 , Figure 2 2 is a structural block diagram of a fault risk assessment device for an isolating switch in an embodiment of the present invention. The device includes a data acquisition module 201 , a data analysis module 202 and a fault prediction module 203 .
[0114] The data acquisition module 201 is used to obtain the historical operation data of the disconnector.
[0115] The prior distribution of the operating data of the disconnector in the initial state is determined based on the historical operating data.
[0116] Collect the actual operation data of the disconnector in real time.
[0117] The data analysis module 202 is used to use the sequential Monte Carlo method to perform fault prediction according to the prior distribution and the actual operation data to obtain the fault prediction result of the disconnector.
[0118] The fault prediction module 203 is used to compare the fault prediction result with a preset fault threshold value to determine whether the disconnector is currently faulty.
[0119] The fault risk assessment device for the disconnector in the embodiment of the present invention adopts the sequential Monte Carlo method and is based on the particle filtering technology to effectively process the nonlinear and non-Gaussian distribution of the disconnector mechanical system to improve the accuracy of the fault prediction results. In addition, the system state is updated based on the time series, and the fault risk assessment result obtained by analyzing the updated system state is more reliable.
[0120] Figure 3 FIG. 1 shows an internal structure diagram of a computer device in one embodiment of the present invention. The computer device may be a terminal or a system. Figure 3 As shown, the computer device includes a processor, a memory and a network interface connected via a system bus. The memory includes a non-volatile storage medium and an internal memory. The non-volatile storage medium of the computer device stores an operating system and may also store a computer program. When the computer program is executed by the processor, the processor may implement each step in the above method embodiment. The internal memory may also store a computer program. When the computer program is executed by the processor, the processor may implement each step in the above method embodiment. Those skilled in the art will understand that Figure 3 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.
[0121] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes each step in the above method embodiment.
[0122] In one embodiment, a computer-readable storage medium is proposed, which stores a computer program. When the computer program is executed by a processor, the processor executes each step in the above method embodiment. A person of ordinary skill in the art can understand that all or part of the processes in the above embodiment method can be completed by instructing the relevant hardware through a computer program. The program can be stored in a non-volatile computer-readable storage medium, and the program can include the processes of the embodiments of the above methods when executed. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.
[0123] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.
[0124] The above embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several modifications and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.
Claims
1. A method for evaluating the failure risk of an isolating switch, characterized in that: The method comprises: Obtain historical operating data of the disconnector; Determine, based on the historical operation data, a priori distribution that the operation data of the disconnector obeys in an initial state; Collecting actual operation data of the isolating switch in real time; Using the Sequential Monte Carlo method, fault prediction is performed according to the prior distribution and the actual operation data to obtain a fault prediction result of the disconnector; The fault prediction result is compared with a preset fault threshold to determine whether the isolating switch is currently faulty.
2. The method according to claim 1, characterized in that The method of using the Sequential Monte Carlo method to perform fault prediction according to the prior distribution and the actual operation data to obtain the fault prediction result of the disconnector specifically includes: Randomly draw N particle samples from the prior distribution; All particle samples are updated using the state transition model to generate the predicted operation data corresponding to each particle sample; A fault analysis is performed based on the gaps between all the predicted operating data and the actual operating data to obtain a fault prediction result.
3. The method according to claim 2, characterized in that The state transition model is expressed by the following formula: x t =f(x t-1 )+v t In the formula, x t is the predicted running data corresponding to the particle sample at time t, f(x t-1 ) is the preset system function of the isolating switch, v t is the preset process noise.
4. The method according to claim 2, characterized in that The performing of fault analysis based on the gap between all the predicted operation data and the actual operation data to obtain a fault prediction result specifically includes: Using a preset likelihood function, a target weight of each particle sample is obtained by performing calculations based on each of the predicted operation data and the actual operation data; The sum of the target weights of all particle samples is calculated to obtain a total weight; the total weight is the fault prediction result.
5. The method according to claim 4, characterized in that The weight calculation formula of the particle sample is as follows: In the formula, is the target weight of the i-th particle sample, y t is the actual operation data, is the i-th particle sample, is the predicted running data corresponding to the i-th particle sample, σ w is the preset discreteness of the observation noise distribution.
6. The method according to claim 4, characterized in that The method of using a preset likelihood function to calculate according to each of the predicted operation data and the actual operation data to obtain a target weight for each particle sample specifically includes: Based on a preset likelihood function, calculation is performed according to the predicted operation data and the actual operation data to obtain an initial weight of each particle sample; The effective number of particles is calculated based on the initial weights of all particle samples; When the effective particle number is less than the preset particle number threshold, resampling the particle samples to generate new particle samples, and continuing to perform the operation of updating all particle samples using the state transition model to generate predicted operation data corresponding to each particle sample until the effective particle number is not less than the preset particle number threshold; When the effective particle number is not less than the preset particle number threshold, the initial weight is used as the target weight, and the step of calculating the sum of the target weights of all particle samples to obtain the total weight is continued.
7. The method according to claim 6, characterized in that The effective number of particles is calculated by the following formula: In the formula, is the target weight of the i-th particle sample, and N is the total number of particle samples.
8. A fault risk assessment device for an isolating switch, characterized in that: The device includes a data acquisition module, a data analysis module and a fault prediction module; The data acquisition module is used to obtain historical operation data of the disconnector; Determine, based on the historical operation data, a priori distribution that the operation data of the disconnector obeys in an initial state; Collecting actual operation data of the isolating switch in real time; The data analysis module is used to use the sequential Monte Carlo method to perform fault prediction according to the prior distribution and the actual operation data to obtain a fault prediction result of the disconnector; The fault prediction module is used to determine whether the isolating switch is currently faulty based on the comparison between the fault prediction result and a preset fault threshold.
9. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the processor is caused to perform the steps of the method according to any one of claims 1 to 7.
10. A computer device comprising a memory and a processor, characterized in that: The memory stores a computer program, and when the computer program is executed by the processor, the processor is caused to perform the steps of the method according to any one of claims 1 to 7.
Citation Information
Patent Citations
Training method of defect prediction model of high-voltage switch cabinet and defect prediction method
CN118568433A
Isolation switch fault risk assessment method and system based on logistic regression, and storage medium
CN118861682A
Reliability assessment method for and planning method for offshore wind farm power collection system
WO2024138866A1
