Vacuum circuit breaker state online monitoring method and system based on digital twinning

Abstract

The invention discloses a vacuum circuit breaker state online monitoring method and system based on digital twinning, and relates to the field of intelligent operation and maintenance of high-voltage power equipment. According to the method, various sensors are arranged at key parts of the vacuum circuit breaker, breaking current, shielding case potential and vibration signals are collected in real time, and a digital twinborn model is constructed by combining multi-source data preprocessing, three-dimensional modeling, behavior rule definition and fault simulation analysis; and on-line identification and trend prediction of electrical wear, vacuum degree deterioration and mechanical abnormity are realized. The system has the functions of real-time monitoring, three-dimensional visualization, model self-learning and intelligent early warning, can be widely applied to intelligent monitoring and state evaluation of circuit breaker equipment in a power system, and remarkably improves the operation safety and operation and maintenance efficiency of the equipment.

Description

Technical Field

[0001] The present invention relates to the technical field of circuit breaker status monitoring, and in particular to a method and system for online monitoring of the status of a vacuum circuit breaker based on digital twins. Background Art

[0002] Existing vacuum circuit breaker monitoring relies primarily on offline testing, which is unable to achieve real-time, accurate operational status assessments. This leads to delayed equipment maintenance and difficulty in timely predicting vacuum level drops, electrical wear, and mechanical failures. Traditional vacuum level detection methods, such as the power frequency withstand voltage method and the laser-induced plasma method, suffer from complex detection processes, high equipment costs, and the inability to perform long-term online monitoring. Invention patent CN109585217A introduces an online vacuum level monitoring device for vacuum circuit breakers in high-voltage switchgear. This device combines a coupling capacitor method with a high-frequency discharge pulse method to simultaneously detect electric field changes and partial discharge signals, enabling online vacuum level monitoring. However, in practical applications, it still faces challenges such as high equipment costs and complex detection.

[0003] Furthermore, the assessment of electrical contact wear typically relies on empirical judgment. Patent CN113161164B proposes a method for measuring contact wear in vacuum circuit breakers. This method uses a high-precision pressure-sensitive sensor installed on the pull rod to measure contact pressure changes and calculate contact wear. However, this method still relies on empirical judgment and lacks efficient data analysis.

[0004] Mechanical fault detection lacks efficient data analysis methods. Invention patent: CN112378633B discloses a method for mechanical fault diagnosis based on time-frequency analysis. This method requires manual presetting of the theoretical fault frequency. However, the actual fault frequency may shift due to load changes and component aging, resulting in missed detections.

[0005] Existing methods monitor single-point faults such as electrical wear and vacuum degree, but lack systematic support for multi-source data fusion, evolution trend prediction, and the linkage between equipment status and operation and maintenance strategies. Big data methods based on operation similarity have not yet been used in predictive maintenance of vacuum circuit breakers, nor have they achieved the linkage between equipment status and planning and scheduling. To solve the above problems, the present invention proposes a method and system for online monitoring of vacuum circuit breaker status based on digital twins. By constructing a virtual simulation model and combining it with real-time data acquisition, online monitoring, prediction and fault warning of the vacuum circuit breaker operating status are achieved, thereby improving the safety and reliability of power equipment operation. Summary of the Invention

[0006] The present invention introduces a device operation similarity model, combined with a database of historical circuit breaker operation states, to achieve matching and trend deduction of the current device operation state. The purpose of the present invention is to provide a method and system for online monitoring of the status of vacuum circuit breakers based on digital twins. This method uses a coupled capacitance method to monitor the shield potential, a weighted accumulation method for breaking current to assess contact electrical wear, and a slime mold optimization algorithm (SMA) to optimize the key parameters of the variational mode decomposition (VMD) algorithm. A method (SMA-VMD-EE) that processes vibration signals to detect mechanical faults, using a method that optimizes variational mode decomposition parameters and extracts energy entropy (EE) as a fault eigenvalue, is used to detect mechanical faults. This method enables accurate status monitoring and fault prediction of vacuum circuit breakers, thereby improving the stability of power grid operation.

[0007] In order to achieve the above technical objectives, the present invention adopts the following technical solutions:

[0008] A method for online monitoring of vacuum circuit breaker status based on digital twins includes the following steps:

[0009] Step 1: Current sensors, coupling capacitor sensors, and vibration acceleration sensors are placed at key locations on the vacuum circuit breaker to collect the circuit breaker's breaking current data, shielding cover potential change data, and vibration signals during operation.

[0010] Step 2: Filter, denoise, normalize, and remove outliers on the collected raw data to obtain high-quality data that can be used for modeling;

[0011] Step 3: Build a 3D geometric model based on the circuit breaker structural drawings and measured parameters, setting the material, electrical, and mechanical properties. The model includes the circuit breaker body, contacts, arc extinguishing chamber, and spring mechanism. Parameters such as conductor conductivity, contact wear coefficient, spring rebound force, and vacuum sealing performance are set. A multi-physics mathematical model is constructed and solved to generate a dynamically responsive digital twin model.

[0012] Step 4: Based on the working principle of the circuit breaker, establish the functional relationship between contact electrical wear and breaking current, potential change and vacuum degree, vibration spectrum and mechanical state, and define the state judgment and warning rules;

[0013] Step 5: Run the digital model in the simulation platform, input historical or real-time data to predict the circuit breaker status, and iteratively modify the model by comparing the monitoring data with the simulation results;

[0014] Step 6: Based on the digital twin model, the circuit breaker status is displayed in the form of a three-dimensional graph and parameter list, allowing users to interactively view device status, trend prediction, and fault diagnosis;

[0015] Step 7: Regularly update the behavior model and rule base based on on-site operation and maintenance data and user feedback to continuously optimize the prediction and diagnosis accuracy of the monitoring system.

[0016] Furthermore, vibration signals of the circuit breaker under normal conditions, including core jamming, spring fatigue, and base loosening, were collected and processed for noise reduction to construct a database. A sticky mushroom optimization algorithm (SMA) was used to determine the optimal parameters for variational modal decomposition (VMD), extract intrinsic modal components, and construct a feature matrix. An evaluation model was established using radial basis functions and corrected based on database training to determine the current state of the circuit breaker.

[0017] Furthermore, the preprocessing of the vibration signal includes extracting impact data and reducing noise through high-pass filtering, and the obtained clean data is used for database construction.

[0018] Furthermore, the slime mold optimization algorithm iteratively searches for the optimal number of decomposition layers and penalty factors, and combines it with the VMD algorithm to obtain the decomposition parameters with the best fitness.

[0019] Furthermore, VMD decomposition extracts intrinsic modal components based on the optimal parameters, combines frequency band information with frequency domain iteration, extracts stable modes and constructs a feature matrix.

[0020] Furthermore, the relevant components are screened based on the correlation coefficient between the natural mode and the original signal and the energy entropy (EE), and a feature matrix is constructed and input into the evaluation model to obtain the current status of the circuit breaker; when the status is abnormal, maintenance or replacement is prompted.

[0021] Furthermore, the vacuum degree is evaluated by measuring the shielding cover potential change using the coupling capacitance method, extracting the voltage signal from the real-time collected data, and combining it with the established "shielding cover potential-relative dielectric constant-vacuum degree" response model. By collecting the shielding cover potential change, the vacuum degree of the vacuum interrupter is inferred.

[0022] Furthermore, the monitoring system extracts the usage records of the circuit breaker, including the current and arcing time of each interruption, and updates them in combination with the real-time collected values; the cumulative value of contact wear is calculated using the weighted accumulation method of the interruption current; based on the usage records and combined with the real-time collected data, the electrical wear of the contacts of the current circuit breaker is obtained, and a wear life prediction model is established.

[0023] Furthermore, the monitoring system has an early warning function. When the vacuum level is lower than a set threshold, the contact wear exceeds the life limit, or the vibration frequency deviates abnormally, the system automatically issues an alarm message.

[0024] The present invention also provides an online monitoring system for the status of a vacuum circuit breaker constructed based on the above method, comprising: a sensor module for collecting the operating parameters of the vacuum circuit breaker, an edge computing and communication module for performing preliminary processing of the data and transmitting it to the cloud platform, a digital twin modeling module for building a virtual model of the circuit breaker and performing simulation, a data analysis and fault identification module for judging the current equipment status and predicting fault trends, and a human-computer interaction and visualization module for displaying the model operation results and equipment status, supporting interactive operations and operation and maintenance management.

[0025] Furthermore, vibration signals of the circuit breaker are collected under various conditions, including normal, core jamming, spring fatigue, and loose base. After noise reduction processing such as high-pass filtering, clean data is generated to construct a historical database for model training. A slime mold optimization algorithm is used to iteratively search for the optimal number of decomposition layers and penalty factors for VMD decomposition to improve the effectiveness of subsequent feature extraction. The VMD algorithm is used for signal decomposition, combined with the Laplace algorithm and frequency domain iteration methods to extract stable intrinsic modal components. Modes strongly correlated with the vibration state are screened, their energy entropy is calculated, and a feature matrix is constructed. An evaluation model is established using radial basis functions. The model is fitted using a training set divided by the database, and a validation set is used to evaluate its accuracy, and the model parameters are continuously adjusted. The current vibration signal is input into the evaluation model to identify the current state of the circuit breaker. If the state is abnormal (such as core jamming, spring fatigue, or loose base), maintenance is prompted; otherwise, normal monitoring is maintained.

[0026] Furthermore, the acquisition of historical data includes collecting operating vibration signals of the circuit breaker under different typical states (normal, core stuck, spring fatigue, loose base), extracting impact characteristics, and performing noise reduction processing through high-pass filtering and other methods to form high-quality clean data to build a state database.

[0027] Furthermore, the slime mold optimization algorithm automatically obtains the optimal decomposition parameters through a combined search of multiple sets of decomposition levels and penalty factors, combined with the fitness value of the VMD decomposition result for iterative update, thereby improving the accuracy and efficiency of modal extraction.

[0028] Furthermore, the extraction of inherent modal components is based on the optimized VMD algorithm, combined with the variational model and the frequency domain iteration method to obtain the stable modes and their center frequencies; then, using correlation analysis and energy entropy calculation, the modes closely related to the vibration state are screened to construct the characteristic matrix.

[0029] Furthermore, the evaluation model is constructed using a radial basis function neural network. The model structure is fitted through the training set, and the model performance is evaluated using the validation set. The kernel function parameters and penalty coefficients are adjusted to improve the classification accuracy and generalization ability.

[0030] Furthermore, the monitoring system inputs the currently collected vibration signal into the evaluation model to determine the status of the circuit breaker; if it identifies a fault state such as core jamming, spring fatigue or loose base, it triggers an alarm to prompt maintenance suggestions; otherwise, it maintains online monitoring and data updates.

[0031] Furthermore, the present invention constructs a virtual simulation model based on the collected data, and the digital twin system includes the following modules: a physical entity layer of the vacuum circuit breaker body, sensors, data acquisition devices, etc.; a data transmission layer that uses the Internet or wireless communication technology to realize real-time data transmission; a virtual model of the vacuum circuit breaker is constructed in a simulation environment, and a digital twin model layer that combines historical data and real-time monitoring data to achieve accurate simulation; and an intelligent analysis and prediction layer that uses machine learning algorithms and signal processing technology to perform real-time analysis of vacuum degree, electrical wear, and mechanical vibration signals, and predict potential faults.

[0032] Furthermore, based on the digital twin model and data analysis, the system monitors the operating status of the vacuum circuit breaker in real time and sets an early warning mechanism: when the shielding cover potential measured by the coupling capacitance method changes abnormally and the calculated vacuum degree is lower than the set threshold, the system will issue an alarm; when the contact wear value calculated by the weighted accumulation algorithm reaches the preset life threshold, the system will automatically remind you to replace the contact; if the characteristic frequency of the vibration signal shows an abnormal deviation, the SMA-VMD-EE analysis will identify abnormal conditions such as mechanical jamming and spring failure, and the system will trigger an alarm.

[0033] Compared with the prior art, the present invention has the following technical advantages:

[0034] 1. A coupled capacitance method is used to monitor shield potential changes in real time. By establishing a "shield potential-dielectric constant-vacuum level" response model, changes in vacuum level can be inferred. This replaces the traditional power-frequency withstand voltage testing method that relies on power outages. Compared to traditional offline testing, this method overcomes the limitations of traditional offline testing, enabling continuous online assessment of vacuum level and ensuring full-cycle status control during equipment operation.

[0035] 2. Utilizing the weighted accumulation method of interrupting current, the current value and arcing time of each interruption are weighted and accumulated to form an electrical wear assessment. This method also establishes a life prediction model. Compared to traditional manual judgment, this method offers the advantages of quantification, traceability, and predictability, supporting a more scientific preventive maintenance strategy.

[0036] 3. Introducing the SMA-VMD-EE algorithm combination. The Slime Mold Optimization Algorithm (SMA) is used to determine the optimal VMD decomposition parameters, improving signal decomposition accuracy. Energy Entropy (EE) is used to analyze the extracted modal eigenvalues to identify the vibration state of the equipment. This method can extract weak vibration characteristics even under complex background noise, effectively identifying typical mechanical faults such as core jamming, spring fatigue, and base loosening, enabling early detection and classification of faults.

[0037] 4. Build a three-dimensional digital twin model that couples multiple physical fields, integrating electrical, thermal, and mechanical behaviors into a unified model. Combined with real-time / historical data, it performs operational simulation and model correction. This can dynamically respond to changes in the circuit breaker's operating status and improve the predictive model's adaptability and accuracy to complex operating conditions.

[0038] 5. The system has a self-learning mechanism and behavioral rule library update function, which can continuously correct the model and judgment logic based on on-site feedback; and supports interactive monitoring and operation of operation and maintenance personnel through a three-dimensional visualization platform, building a "virtual and real" operation and management system with self-evolution and intelligent decision-making capabilities, improving overall operation and maintenance efficiency and intelligence level. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 : Schematic diagram of mechanical fault monitoring process;

[0040] Figure 2 : Schematic diagram of the contact electrical wear monitoring process;

[0041] Figure 3 :Schematic diagram of vacuum degree monitoring process of vacuum interrupter;

[0042] Figure 4 : physical entity diagram;

[0043] Figure 5 : Virtual object functional diagram;

[0044] Figure 6 : How the specific diagram of the overall architecture of the digital twin system reflects the digital twin diagram. DETAILED DESCRIPTION

[0045] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0046] A method for online monitoring of vacuum circuit breaker status based on digital twins includes the following steps:

[0047] Step 1: Arrange a current sensor, a coupling capacitor sensor, and a vibration acceleration sensor at the location of the vacuum circuit breaker to respectively collect the circuit breaker's breaking current data, shielding cover potential change data, and vibration signals during operation;

[0048] Step 2: Filter, denoise, normalize and remove outliers on the collected data to obtain high-quality data that can be used for modeling;

[0049] Step 3. Establish a three-dimensional geometric model based on the circuit breaker structural drawings and measured parameters, set the material, electrical, and mechanical properties, and construct a three-dimensional geometric model in the simulation platform. The model includes key components such as the circuit breaker body, contacts, arc extinguishing chamber, and spring mechanism; set the physical properties of each component in the model, including the electrical conductivity of the conductor material, the wear coefficient of the contact material, the rebound force parameters of the spring assembly, and the vacuum sealing performance of the arc extinguishing chamber; in addition, considering the multi-physical field coupling factors, establish mathematical models of multi-physical fields such as electric field, magnetic field, thermal field, and mechanical motion, and define electrical behavior, mechanical motion behavior, and insulation characteristics; by setting initial conditions and boundary conditions, use simulation software to solve the model and construct a digital twin model with dynamic response capabilities for subsequent operation simulation and state prediction.

[0050] Step 4: Based on the working principle of the circuit breaker, establish the functional relationship between contact electrical wear and breaking current, potential change and vacuum degree, vibration spectrum and mechanical state, and define state judgment and early warning rules;

[0051] Step 5: Run the digital model in the simulation platform, input historical or real-time data to predict the circuit breaker status, and iteratively modify the model by comparing the actual monitoring data with the simulation results;

[0052] Step 6: Based on the digital twin model, the circuit breaker status is displayed in the form of a three-dimensional graph and parameter list, allowing users to interactively view the circuit breaker's electrical and mechanical status information, trend prediction, and fault diagnosis;

[0053] Step 7: Regularly update the behavior model and rule base based on on-site operation and maintenance data and user feedback to continuously optimize the prediction and diagnosis accuracy of the digital twin system.

[0054] In a further implementation, such as Figure 1 As shown in the figure, vibration signals of the circuit breaker in normal, core stuck, spring fatigue and base loose states are collected, and a database is constructed after high-pass filtering and noise reduction processing; the slime mold optimization algorithm (SMA) is used to determine the optimal parameters of the VMD algorithm, and the inherent modal components are extracted to construct a feature matrix; a radial basis function neural network is used to establish an evaluation model and train and correct it to determine the current state of the circuit breaker; the vibration signal impact data is combined with high-pass filtering to obtain clean data; SMA determines the VMD decomposition parameters by searching for the optimal decomposition layer number and penalty factor, and extracts stable modes in combination with frequency domain information; the modes are selected according to the correlation between the inherent modes and the original signal and the energy entropy to construct a feature matrix and input into the model to determine the state of the circuit breaker. If abnormal, maintenance is prompted.

[0055] At the same time, the system extracts the circuit breaker usage records, including each breaking current and arcing time, and combines the real-time collected values with the breaking current weighted accumulation method to calculate the contact wear value and establish a life prediction model, such as Figure 2 shown.

[0056] In terms of vacuum degree evaluation, the coupling capacitance method is used to measure the potential change of the shielding cover, extract the voltage signal and infer the vacuum degree based on the "potential-dielectric constant-vacuum degree" response model, such as Figure 3 shown.

[0057] like Figure 4 As shown in the figure, based on the circuit breaker structural drawings and actual measured parameters, a physical entity layer consisting of components such as the circuit breaker body, sensors, and edge computing units is established. On this basis, the three-dimensional geometric foundation required for the digital twin is constructed based on the actual properties of each component, such as the conductivity of the conductor, the wear coefficient of the contact material, the rebound force of the spring mechanism, and the vacuum sealing performance of the arc extinguishing chamber. Based on the circuit breaker structural drawings and actual measured parameters, a three-dimensional geometric model is constructed in the simulation platform, and the conductivity of the conductor, the wear coefficient of the contact material, the rebound force of the spring mechanism, and the vacuum sealing performance of the arc extinguishing chamber are set.

[0058] Based on the operating principle of the circuit breaker, an empirical functional relationship between contact electrical wear and interrupting current was established. The potential change of the shielding cover was used to reflect the trend of decreasing vacuum, and the modal characteristics of the vibration signal were used to reflect the operating stability of the mechanical device. Furthermore, various fault diagnosis rules were set. When the shielding cover potential measured by the coupling capacitance method changes abnormally and the calculated vacuum level falls below a set threshold, the system issues an alarm. When the contact wear value calculated by the weighted accumulation algorithm reaches the preset lifespan threshold, the system automatically prompts contact replacement. If the characteristic frequency of the vibration signal shows an abnormal deviation, the SMA-VMD-EE analysis will identify abnormal conditions such as mechanical jamming and spring failure, triggering the system alarm.

[0059] like Figure 5 As shown, the virtual object layer in the digital twin system receives historical and real-time monitoring data. Using a 3D model and simulation platform, it simulates the circuit breaker's operating behavior under different electrical and mechanical conditions. It also predicts and identifies trends for contact electrical wear, vacuum degradation, and mechanical failures. By continuously comparing simulation outputs with field measurements, the behavioral model is dynamically modified to improve the virtual model's fit to actual operating conditions and its prediction accuracy. This figure emphasizes the "functions implemented by the virtual object." Historical operating data is input into the digital twin, and the simulation platform simulates the circuit breaker's behavior under various conditions, such as current, vacuum level, and contact wear. By comparing simulation results with field measurements, behavioral model parameters are adjusted to improve its fit to actual operating conditions.

[0060] The system periodically updates models for electrical wear, vacuum degradation, and vibration characteristics based on new data generated during actual circuit breaker operation, ensuring accuracy and timeliness. Any maintenance or component replacement is automatically recorded and the model structure or parameters are updated accordingly.

[0061] like Figure 6 As shown in Figure 1, the overall architecture of the digital twin system consists of a physical entity layer, a data transmission layer, a digital twin model layer, and an intelligent analysis and prediction layer. The physical entity layer is deployed at the vacuum circuit breaker site and primarily includes the circuit breaker itself, vibration sensors, coupling capacitor sensors, current sensors, and edge computing units. It is responsible for collecting key signals during device operation, such as current waveforms, voltage changes, and vibration data. The data transmission layer transmits raw or pre-processed data stably and with low latency to a cloud platform or local simulation platform via wired or wireless communication methods (such as industrial Ethernet, Wi-Fi, and 5G), providing real-time data support for subsequent modeling and analysis. The digital twin model layer constructs a three-dimensional virtual model that maps closely to the physical circuit breaker in a simulation environment. This model integrates multiple physical factors, including electrical characteristics, mechanical motion, thermal field changes, and insulation behavior. A simulation solver dynamically simulates the device's operating status, achieving virtual-real mapping and behavioral synchronization. The intelligent analysis and prediction layer uses machine learning algorithms and signal processing techniques to perform feature extraction, modal identification, and energy entropy analysis on the input data, thereby completing electrical wear life assessment, vacuum degradation prediction, and mechanical fault diagnosis. This layer supports model self-learning and rule base updates, ensuring the system maintains diagnostic accuracy and responsiveness during long-term operation. The entire system uses a data-driven closed-loop mechanism between layers to achieve continuous twin updates and intelligent evolution, enabling three-dimensional visualization of vacuum circuit breaker status, trend identification, anomaly warnings, and intelligent decision support.

[0062] The system's overall architecture consists of a physical layer and a digital virtual layer, enabling bidirectional data flow and dynamic, iterative updates of behavioral models through interactive communication links. The physical layer includes devices such as circuit breakers, sensors, and edge computing units; the virtual layer includes functional modules such as digital modeling, behavioral simulation, and fault analysis and prediction. The system can be deployed in power grid substations or distribution stations to enable real-time visual monitoring of circuit breaker status, anomaly warnings, trend prediction, and intelligent operation and maintenance management. Ultimately, the system will be deployed in power grid substations or distribution stations and integrated with existing monitoring platforms to enable online evaluation of vacuum circuit breaker operating status, predict fault trends and automatically warn, deliver intelligent maintenance recommendations to operators, and optimize overall operation and maintenance plans. The system also supports collecting user feedback and suggestions for subsequent functional improvements and continuous model optimization.

Claims

1. A method for online monitoring of vacuum circuit breaker status based on digital twin, characterized in that: The following steps are involved: Step 1: Current sensors, coupling capacitor sensors, and vibration acceleration sensors are placed at key locations on the vacuum circuit breaker to collect the circuit breaker's breaking current data, shielding cover potential change data, and vibration signals during operation. Step 2: Filter, denoise, normalize, and remove outliers on the collected raw data to obtain high-quality data that can be used for modeling; Step 3: Build a 3D geometric model based on the circuit breaker structural drawings and measured parameters, setting the material, electrical, and mechanical properties. The model includes the circuit breaker body, contacts, arc extinguishing chamber, and spring mechanism. Parameters such as conductor conductivity, contact wear coefficient, spring rebound force, and vacuum sealing performance are set. A multi-physics mathematical model is constructed and solved to generate a dynamically responsive digital twin model. Step 4: Based on the working principle of the circuit breaker, establish the functional relationship between contact electrical wear and breaking current, potential change and vacuum degree, vibration spectrum and mechanical state, and define the state judgment and warning rules; Step 5: Run the digital model in the simulation platform, input historical or real-time data to predict the circuit breaker status, and iteratively modify the model by comparing the monitoring data with the simulation results; Step 6: Based on the digital twin model, the circuit breaker status is displayed in the form of a three-dimensional graph and parameter list, allowing users to interactively view device status, trend prediction, and fault diagnosis; Step 7: Regularly update the behavior model and rule base based on on-site operation and maintenance data and user feedback to continuously optimize the prediction and diagnosis accuracy of the monitoring system.

2. The method for online monitoring of vacuum circuit breaker status based on digital twin according to claim 1, characterized in that: Vibration signals of the circuit breaker in normal conditions, core jamming, spring fatigue, and base looseness are collected and processed for noise reduction to construct a database. A slime mold optimization algorithm is used to determine the optimal parameters for variational modal decomposition, extract inherent modal components, and construct a characteristic matrix. An evaluation model is established through radial basis function and corrected based on database training to determine the current status of the circuit breaker.

3. The method for online monitoring of vacuum circuit breaker status based on digital twin according to claim 2, characterized in that: The vibration signal preprocessing includes extracting impact data and reducing noise through high-pass filtering, and the obtained clean data is used for database construction.

4. The method for online monitoring of vacuum circuit breaker status based on digital twin according to claim 2, characterized in that: The slime mold optimization algorithm iteratively searches for the optimal number of decomposition layers and penalty factors, and combines the variational mode decomposition algorithm to obtain the decomposition parameters with the best fitness.

5. The method for online monitoring of vacuum circuit breaker status based on digital twin according to claim 2, characterized in that: Variational mode decomposition extracts the intrinsic modal components based on the optimal parameters, combines frequency band information with frequency domain iteration, extracts stable modes and constructs the characteristic matrix.

6. The method for online monitoring of vacuum circuit breaker status based on digital twin according to claim 5, characterized in that: The relevant components are screened based on the correlation coefficient and energy entropy between the inherent mode and the original signal, and a feature matrix is constructed and input into the evaluation model to obtain the current status of the circuit breaker; when the status is abnormal, it prompts maintenance or replacement.

7. The method for online monitoring of vacuum circuit breaker status based on digital twin according to claim 1, characterized in that: The vacuum degree is assessed by measuring the shield cover potential change using the coupling capacitance method, extracting the voltage signal from the real-time data, and combining it with the established "shield cover potential-relative dielectric constant-vacuum degree" response model. By collecting the shield cover potential change, the vacuum degree of the vacuum interrupter is inferred.

8. The method for online monitoring of vacuum circuit breaker status based on digital twin according to claim 1, characterized in that: The monitoring system extracts the usage records of the circuit breaker, including the current and arcing time of each interruption, and updates them in combination with the real-time collected values; the cumulative value of contact wear is calculated using the weighted accumulation method of the interruption current; based on the usage records and combined with the real-time collected data, the electrical wear of the contacts of the current circuit breaker is obtained and a wear life prediction model is established.

9. The method for online monitoring of vacuum circuit breaker status based on digital twin according to claim 1, characterized in that: The monitoring system has an early warning function. When the vacuum level is lower than the set threshold, the contact wear exceeds the life limit, or the vibration frequency deviates abnormally, the system automatically issues an alarm message.

10. An online monitoring system for vacuum circuit breaker status constructed based on the method according to any one of claims 1 to 9, characterized in that: include: A sensor module for collecting the operating parameters of the vacuum circuit breaker, an edge computing and communication module for preliminary data processing and transmission to the cloud platform, a digital twin modeling module for building a virtual model of the circuit breaker and performing simulations, a data analysis and fault identification module for judging the current equipment status and predicting fault trends, a human-computer interaction and visualization module for displaying model operation results and equipment status, and supporting interactive operations and operation and maintenance management.

Images