Self-adaptive safety protection circuit breaker system and method based on multi-parameter fusion

Abstract

The invention discloses a self-adaptive safety protection circuit breaker system and method based on multi-parameter fusion, and the system comprises a multi-parameter fusion sensing module which is used for collecting and fusing the operation state data of a circuit breaker, and outputting a real-time state evaluation result of the circuit breaker; the predictive maintenance module is used for predicting the health state and the residual life of the circuit breaker based on historical data and the real-time state evaluation result, and generating health state information; and the self-adaptive protection decision module is used for collecting power grid operation state parameters, and dynamically adjusting a protection constant value and generating a protection instruction in combination with the real-time state evaluation result of the circuit breaker and the health state information. The contradiction between a fixed circuit breaker protection setting value and the requirement for accurate and sensitive protection action in a complex operation environment is solved, a dynamic self-adaptive adjustment protection strategy is realized, and the safety of a power system is remarkably improved.

Description

Technical Field

[0001] This application relates to the field of power technology, and in particular to an adaptive safety protection circuit breaker system and method based on multi-parameter fusion. Background Technology

[0002] With the rapid advancement of smart grid construction, circuit breakers, as critical protection devices in power systems, are undergoing a significant transformation from traditional mechanical switches to intelligent units. Current development of intelligent circuit breaker technology mainly revolves around three directions: multi-parameter fusion sensing, adaptive protection decision-making, and predictive maintenance. However, related technical solutions generally suffer from problems such as single-function limitations and system fragmentation. That is, individual circuit breaker products typically only possess one functional module, and there is a lack of effective coordination mechanisms between modules, making it difficult to meet the comprehensive intelligent requirements of new power systems.

[0003] At the level of specific technical implementation, the technical solutions in the above three directions have significant shortcomings. First, circuit breakers with multi-parameter fusion sensing capabilities often limit their data fusion to simple fusion of similar data, lacking cross-domain correlation, leading to inaccurate judgments in complex operating environments. Second, circuit breakers with adaptive protection decision-making capabilities often focus on adjusting parameters of a single protection function, lacking coordinated optimization of the entire protection system. Finally, circuit breakers with predictive maintenance capabilities often rely on building predictive models from single types of data, failing to fully consider the combined impact of multiple factors such as electrical stress, mechanical wear, and environmental aging on equipment lifespan.

[0004] These shortcomings collectively make it difficult for circuit breaker systems to achieve closed-loop control from comprehensive perception to intelligent decision-making, thus limiting their applicability in complex power grid environments. Therefore, it is urgent to construct an intelligent circuit breaker system architecture that integrates multi-parameter fusion perception, adaptive protection decision-making, and predictive maintenance, thereby breaking down the technical barriers between functional modules and achieving comprehensive security protection from data acquisition to protection execution. Summary of the Invention

[0005] This application provides an adaptive safety protection circuit breaker system and method based on multi-parameter fusion, which resolves the contradiction between fixed circuit breaker protection settings and the requirement for precise and sensitive protection actions in complex operating environments. It realizes a protection strategy that can be dynamically and adaptively adjusted, significantly improving the safety of the power system.

[0006] This application provides an adaptive safety protection circuit breaker system and method based on multi-parameter fusion, the system comprising: The multi-parameter fusion sensing module is used to collect and fuse the circuit breaker's operating status data and output the circuit breaker's real-time status assessment results. The predictive maintenance module is used to predict the health status and remaining life of the circuit breaker based on historical data and the real-time status assessment results, and generate health status information. The adaptive protection decision module is used to collect power grid operating status parameters, and combine the real-time status assessment results of the circuit breaker and the health status information to dynamically adjust the protection settings and generate protection commands.

[0007] Optionally, the multi-parameter fusion sensing module includes an uncertain information fusion unit, which performs uncertain information fusion processing on multi-sensor data and outputs the real-time state assessment result; wherein, the uncertain information fusion processing includes: assigning confidence levels to data from different sensors, and performing fusion calculation on the confidence levels based on information fusion rules.

[0008] Optionally, the multi-parameter fusion sensing module includes a multi-dimensional sensing network, which comprises at least three of the following: current sensor, voltage sensor, temperature sensor, vibration sensor, arc light sensor, humidity sensor, and partial discharge sensor.

[0009] Optionally, the predictive maintenance module includes a health prediction model, which constructs a multi-level health index system including electrical life, mechanical life, and insulation status based on the historical data and the real-time status assessment results; it uses a hybrid model combining temporal convolutional networks and long short-term memory networks to predict the remaining service life of the circuit breaker; and it generates the health status information based on the multi-level health index system and the remaining service life.

[0010] Optionally, the adaptive protection decision module includes a power grid parameter processing unit and a protection function coordination matrix. The power grid parameter processing unit calculates the protection setting based on the power grid operating status parameters. The protection function coordination matrix uses the real-time status assessment results of the circuit breaker and the health status information as correction factors to perform sensitivity correction on the protection setting.

[0011] Optionally, the adaptive protection decision module uses a fuzzy logic decision system to implement the sensitivity correction; the input variables of the fuzzy logic decision system include the power grid operating status parameters, the real-time status assessment results of the circuit breaker, and the health status information.

[0012] Furthermore, to achieve the above objectives, embodiments of the present invention also provide an adaptive safety protection circuit breaker operation method based on multi-parameter fusion, the method comprising the following steps: The circuit breaker's operating status data is collected through a multi-sensor array, including electrical data, mechanical data, environmental data, and fault characteristic data. The operational status data is subjected to uncertainty information fusion processing to output the real-time status assessment result of the circuit breaker; Historical data of the circuit breaker is acquired, and combined with the real-time status assessment results of the circuit breaker, the health status and remaining life of the circuit breaker are predicted, and health status information is generated; wherein, the historical data includes the historical operating characteristic parameters, historical operation records and historical maintenance records of the circuit breaker. The system collects power grid operating status parameters and, in conjunction with the real-time status assessment results of the circuit breaker and the health status information, dynamically adjusts protection settings and generates protection commands.

[0013] Optionally, the step of performing uncertainty information fusion processing on the operating status data and outputting the real-time status assessment result of the circuit breaker includes: Feature extraction is performed on the operating status data to obtain multi-dimensional features characterizing the operating status of the circuit breaker; Based on the evidence theory algorithm, the multi-dimensional adjustments are fused to obtain the real-time status assessment result of the circuit breaker.

[0014] Optionally, the step of acquiring historical data of the circuit breaker, combining it with the real-time status assessment results of the circuit breaker, predicting the health status and remaining life of the circuit breaker, and generating health status information includes: A multi-level health index system is constructed, and a comprehensive health index is output; wherein, the multi-level health index system includes electrical life, mechanical life and insulation status dimensions; A hybrid model combining temporal convolutional networks and long short-term memory networks is used to perform sequence analysis on the historical data and the real-time status assessment results of the circuit breaker to predict the remaining service life of the circuit breaker. Based on the multi-level health index system and the remaining service life, a personalized maintenance strategy containing early warning information and maintenance suggestions is generated as the health status information.

[0015] Optionally, the step of collecting power grid operating status parameters, combining the real-time status assessment results of the circuit breaker and the health status information, and dynamically adjusting the protection settings and generating protection commands includes: Based on real-time power flow calculation, the power grid operating status parameters are obtained, including power grid short-circuit capacity, voltage stability and distributed generation penetration rate. The power grid operating status parameters, the real-time status assessment results of the circuit breaker, and the health status information are jointly input into the fuzzy logic decision system, and the protection settings are dynamically adjusted through fuzzy rules. The protection command is generated by coordinating the logical coordination of distance protection, overcurrent protection and zero-sequence protection based on the adjusted protection settings using a preset protection function coordination matrix.

[0016] One or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages: 1. By using a multi-parameter fusion sensing module to perform uncertainty fusion processing on the electrical, mechanical, environmental, and fault characteristic quantities of the circuit breaker, real-time status assessment results of the circuit breaker are obtained. This solves the problem of the one-sidedness of monitoring by a single sensor or homogeneous data source. Through collaborative analysis of heterogeneous data, comprehensive perception of the circuit breaker's operating status is achieved.

[0017] 2. The predictive maintenance module constructs a multi-level health index system and uses a hybrid model combining temporal convolutional networks and long short-term memory networks for sequence analysis to generate circuit breaker health status information. This upgrades health assessment from a static snapshot of the current state to dynamic tracking and prediction of performance degradation processes, enabling proactive planning and operational decision-making from a reactive approach.

[0018] 3. The adaptive protection decision module incorporates power grid operating status parameters and combines them with the circuit breaker's own status and health trends. It utilizes a fuzzy logic decision system to dynamically adjust protection settings and generate protection commands. This overcomes the limitations of static protection setting and enables the protection system to adapt to changes in power grid operating modes and equipment degradation. Attached Figure Description

[0019] Figure 1 This is a system architecture diagram for this application; Figure 2 This is a flowchart illustrating the working method of the robot control method of this application, which is based on the adaptive safety protection circuit breaker operation method of multi-parameter fusion. Detailed Implementation

[0020] To address the shortcomings of existing circuit breaker systems in complex power grid environments, such as insufficient protection performance and lack of foresight due to functional isolation and data fragmentation, this application proposes a circuit breaker system integrating multi-parameter fusion sensing, predictive maintenance, and adaptive protection decision-making. The multi-parameter fusion sensing module integrates electrical, mechanical, environmental, and fault characteristics to achieve highly reliable condition assessment; the predictive maintenance module predicts health trends and remaining lifespan based on historical and real-time data; and the adaptive protection decision-making module dynamically adjusts protection settings and strategies by combining the above information with real-time power grid operating parameters. This system transforms the circuit breaker from passive interruption to proactive intelligent protection, significantly improving the safety, reliability, and operational adaptability of the power system.

[0021] To better understand the above technical solutions, exemplary embodiments of this application will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of this application are shown in the drawings, it should be understood that this application can be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of this application and to fully convey the scope of this application to those skilled in the art.

[0022] To better understand the above technical solutions, the following will provide a detailed explanation of the technical solutions in conjunction with the accompanying drawings and specific implementation methods.

[0023] Example 1 In this embodiment, an adaptive safety protection circuit breaker system based on multi-parameter fusion is provided.

[0024] Reference Figure 1 This embodiment provides an adaptive safety protection circuit breaker system based on multi-parameter fusion, including: The multi-parameter fusion sensing module is used to collect and fuse the circuit breaker's operating status data and output the circuit breaker's real-time status assessment results. In this embodiment, the multi-parameter fusion sensing module includes a multi-dimensional sensing network, which includes at least three of the following: current sensor, voltage sensor, temperature sensor (monitoring the temperature of contacts and terminals), vibration sensor (detecting the status of mechanical operating mechanism), arc light sensor (identifying fault arc), humidity sensor (monitoring condensation risk), and partial discharge sensor (assessing insulation status).

[0025] Optionally, the multi-parameter fusion sensing module also has an adaptive dual-mode sampling mechanism with intelligent switching logic: in normal operation, it adopts a conventional monitoring mode with a low-frequency sampling frequency of 1kHz to track the trend changes of parameters; when any parameter is detected to exceed a preset threshold or multiple parameters show a coordinated abnormal trend, it automatically switches to a precision diagnostic mode with a high-frequency sampling frequency of 10kHz to accurately capture transient features and detailed information.

[0026] Optionally, after entering the precision diagnostic mode, once the parameters return to normal, continuous monitoring can continue for 3-5 cycles before automatically returning to the regular monitoring mode, thus achieving a dynamic balance between power consumption optimization and monitoring accuracy.

[0027] In the data feature extraction stage, the raw sensor data is transformed into feature indicators with clear physical meaning. A multi-scale feature extraction method is employed, mining features from three dimensions: time domain, frequency domain, and time-frequency features. In the time-frequency domain, parameters such as mean, variance, and kurtosis are extracted to characterize the basic amplitude characteristics and statistical distribution of the signal. In the frequency domain, indicators such as power spectral density and bandwidth ratio are calculated to reveal the energy distribution patterns of different frequency components in the signal. Simultaneously, time-frequency analysis algorithms such as wavelet transform and empirical mode decomposition are used to obtain the signal's features in the joint time-frequency domain, thereby effectively capturing transient components and dynamic changes in non-stationary signals.

[0028] The data acquisition and preprocessing unit performs preprocessing such as cleaning, filtering and standardization on the acquired multi-source data, and then extracts valuable time-domain and frequency-domain features.

[0029] In the multi-source data fusion stage, after obtaining multi-dimensional features, the Dempster-Shafer evidence theory can be used to fuse uncertain information. First, the confidence level of each sensor in supporting different state propositions (such as "normal", "overheating", "insulation fault") is quantified through a basic probability assignment function. Then, based on specific information fusion rules, data from different sensors are synthesized. These rules can effectively handle and resolve inconsistencies and conflicts between sensor data by calculating the degree of data conflict, ultimately outputting a comprehensive and highly reliable state assessment result.

[0030] Specifically, the Dempster-Shafer evidence theory is used to fuse uncertainty information from multi-source data. The credibility of each piece of evidence (i.e., different sensors) is quantified through a basic probability assignment function. The information fusion rules are as follows: in, To improve the credibility of proposition A after fusion, The basic probability assignment for different sensors is given, where K is the conflict factor.

[0031] Furthermore, the multi-parameter fusion sensing module also includes edge computing and real-time response stages. To meet the real-time requirements of the protection system, a hybrid architecture of "edge computing + cloud collaboration" is adopted. A lightweight AI model is deployed locally on the circuit breaker, responsible for completing more than 90% of the basic analysis and fusion computing tasks, ensuring that critical protection commands can be issued within 30 milliseconds. At the same time, only high-value feature data and diagnostic results that have been selected are uploaded to the cloud for deeper model verification, optimization, and historical data analysis, thereby achieving continuous improvement in the overall intelligence level while ensuring rapid response.

[0032] The predictive maintenance module is used to predict the health status and remaining life of the circuit breaker based on historical data and the real-time status assessment results, and generate health status information. In this embodiment, the predictive maintenance module uses big data analysis and machine learning algorithms to assess the health status of circuit breakers and predict their remaining lifespan.

[0033] Optionally, the predictive maintenance module receives the real-time condition assessment results generated by the multi-parameter fusion sensing module and acquires relevant historical data of the circuit breaker. Based on the real-time condition assessment results and historical data, it constructs a multi-level health index system comprising three dimensions: electrical life (e.g., contact wear), mechanical life (e.g., operating mechanism wear), and insulation condition (e.g., partial discharge activity). Through feature layer fusion and the adoption of Dempster-Shafer evidence theory, multi-source monitoring information from different dimensions is synthesized into a unified health index. The health index ranges from 0 to 1, where 1 represents a healthy state and 0 represents complete failure.

[0034] Historical data refers to the historical records of the circuit breaker's operating status and the historical archives of maintenance and events.

[0035] The historical distance of the main body's operating status refers to the data collected and processed by the multi-parameter fusion sensing module in the past, such as historical temperature trends (average and maximum temperature records of circuit breaker contacts per week in the past three months), historical vibration characteristics (the change curves of the mean, variance, and kurtosis of vibration amplitude during each opening and closing operation), historical insulation indicators (monthly statistical values ​​of partial discharge activity and historical records of ambient humidity), and historical operation counts (cumulative number of opening and closing operations and number of fault current interruptions).

[0036] Maintenance and event history records are event records related to the circuit breaker's lifecycle, such as historical maintenance records (date of the last maintenance, parts replaced, and parameters adjusted, etc.) and historical fault records (types of faults that have occurred in the past, the number of times, and electrical parameters before and after the fault, etc.).

[0037] Optionally, for the remaining service life of the circuit breaker, a hybrid model combining temporal convolutional networks and long short-term memory networks can be used to perform sequence analysis on historical and real-time monitoring data of the equipment, predicting the changing trend and remaining service life over a future period, such as 12-24 hours. The temporal convolutional network captures long-term dependencies through dilated causal convolutions, while the long short-term memory network processes the temporal relationships of the sequence data. The model expression is as follows: in, This is the predicted remaining lifespan. Let be the eigenvector at time t. These are the model parameters.

[0038] Optionally, the health status information is generated by combining a multi-level health index system and remaining service life. Specifically, based on the multi-level health index system and predicted remaining service life, and considering factors such as equipment importance and maintenance resources, a personalized maintenance strategy can be generated using a multi-objective optimization algorithm, such as continue operation, planned maintenance, or immediate shutdown. Simultaneously, the equipment health status, predicted trends, and maintenance recommendations can be displayed through a visual interface, such as "The contact is expected to require maintenance after 200 operations."

[0039] The adaptive protection decision module is used to collect power grid operating status parameters, and combine the real-time status assessment results of the circuit breaker and the health status information to dynamically adjust the protection settings and generate protection commands.

[0040] In this embodiment, the adaptive protection decision module dynamically adjusts protection parameters and strategies through real-time system status perception (circuit breaker real-time status assessment results) and intelligent algorithms (health status information) to adapt to the complex operating environment of the power system.

[0041] Optionally, power grid operating status parameters, including short-circuit capacity, voltage stability, and distributed generation penetration, can be obtained based on real-time power flow calculations. Then, combined with network topology identification, an equivalent impedance model of the power grid can be established, and protection settings can be calculated.

[0042] Optionally, after calculating the protection settings, the real-time status assessment results and health status information of the circuit breaker are used as correction factors to perform sensitivity correction on the protection settings, thereby generating the final protection settings.

[0043] A fuzzy logic decision-making system is employed to transform the real-time status assessment results and health status information of the circuit breaker into fuzzy rules, enabling smooth adjustment of protection settings. Taking overcurrent protection as an example, the operating current setting... Based on the distributed power source injection current With load current Dynamic adjustment: in, As a benchmark value, The adjustment coefficients are dynamically updated based on real-time short-circuit capacity and voltage stability. Simultaneously, an adaptive adjustment mechanism for the directional element's operating angle is introduced to dynamically correct the boundary conditions of the protection characteristics, ensuring the correctness and selectivity of protection operation under complex fault conditions.

[0044] Optionally, the adaptive protection decision module also includes a multi-protection function coordination unit. Based on grid operating status parameters, circuit breaker real-time status assessment results, and health status information, the adaptive protection decision module calculates adaptive protection settings for overcurrent, distance, and zero-sequence protection functions. The multi-protection function coordination unit then performs coordinated decision-making and logical management on these adjusted protection functions, ensuring they work collaboratively as a whole. When a specific operating mode is detected, such as microgrid islanded operation, the corresponding protection group and logic are automatically activated to avoid conflicts and blind spots in protection functions.

[0045] The adaptive safety protection circuit breaker system based on multi-parameter fusion also includes an intelligent execution layer.

[0046] In this embodiment, the intelligent execution layer (such as a trip unit) receives protection instructions from the adaptive protection decision module and generates precise operations (such as tripping) to cut off the fault or issue an alarm, thereby completing one protection cycle.

[0047] The adaptive safety protection circuit breaker system based on multi-parameter fusion also includes a cloud collaboration module.

[0048] In this embodiment, the cloud collaboration module adopts a hybrid architecture of "edge-cloud" collaboration. While making real-time decisions at the edge, the predictive maintenance module uploads some anonymized operational data and diagnostic results to the cloud platform as a continuous and asynchronous optimization process. The cloud platform performs in-depth mining and analysis on the received massive amounts of data, and based on this, retrains, validates, and optimizes the hybrid model of the predictive maintenance module deployed at the edge. After optimization, the updated and more powerful model parameters are distributed to the predictive maintenance module. This process enables the circuit breaker system to break through the limitations of data and computing power of a single edge node, forming an intelligent ecosystem that can continuously learn from the collective operational experience and iteratively optimize, thereby achieving continuous evolution of the overall diagnostic accuracy and predictive capabilities of the system.

[0049] In this embodiment, the multi-parameter fusion sensing module collects data through a multi-dimensional sensor network and outputs a highly reliable real-time status assessment result after uncertainty information fusion. The predictive maintenance module predicts equipment lifespan and generates health status information based on historical and real-time data through a health index system and a hybrid model. The adaptive protection decision module integrates the power grid status, real-time equipment status, and health trends, and dynamically adjusts setpoints and generates optimal protection commands through fuzzy logic and a coordination matrix. Finally, the intelligent execution structure precisely executes the commands to complete fault isolation or early warning, ensuring system safety. In addition, this closed-loop collaborative mechanism also realizes the transformation of circuit breakers from passive disconnection to active intelligent protection.

[0050] Based on the same inventive concept, this application also provides the working method corresponding to the system in Embodiment 1, as shown in Embodiment 2.

[0051] Example 2 This embodiment proposes an adaptive safety protection circuit breaker operation method based on multi-parameter fusion, the method including the following steps: Step S100: Collect the operating status data of the circuit breaker through a multi-sensor array. The operating status data includes electrical data, mechanical data, environmental data, and fault characteristic data. In this embodiment, a multi-sensor array is used to synchronously collect operational data of multiple physical quantities from the circuit breaker. Specifically, current and voltage sensors acquire electrical operating parameters, vibration sensors monitor mechanical operating characteristics, temperature and humidity sensors record environmental conditions, and arc flash sensors capture fault characteristics. This multi-dimensional data acquisition method overcomes the limitations of traditional single electrical quantity monitoring, providing comprehensive data support for subsequent in-depth analysis.

[0052] Step S200: Perform uncertainty information fusion processing on the operating status data and output the real-time status assessment result of the circuit breaker; In this embodiment, multi-source heterogeneous sensor data is transformed into highly reliable state assessment results through uncertainty information fusion processing. Evidence theory algorithms are then employed to collaboratively analyze potentially conflicting or uncertain multi-sensor evidence, ultimately outputting a unified diagnostic conclusion.

[0053] As an optional implementation, feature extraction is performed on the operating status data to obtain multi-dimensional features characterizing the operating status of the circuit breaker; based on the evidence theory algorithm, the multi-dimensional adjustments are fused to obtain the real-time status evaluation result of the circuit breaker.

[0054] Optionally, when processing uncertain information fusion, in addition to evidence theory algorithms, methods such as Bayesian inference, fuzzy logic, neural networks, and Kalman filtering, as well as hybrid strategies of two or more of these methods, can be used to determine the real-time state of the circuit breaker. For example, a neural network can be used to reduce the dimensionality and extract features from the original data, and the extracted features can be used as evidence to input into a DS evidence theory or fuzzy logic system for the final state determination.

[0055] Step S300: Obtain historical data of the circuit breaker, combine it with the real-time status assessment results of the circuit breaker, predict the health status and remaining life of the circuit breaker, and generate health status information; wherein, the historical data includes the historical operating characteristic parameters, historical operation records and historical maintenance records of the circuit breaker; In this embodiment, real-time condition assessment results are combined with historical operating data to construct a multi-level health assessment system covering electrical life, mechanical life, and insulation condition. By analyzing historical operating characteristic parameters, operation records, and maintenance files, a full lifecycle health model of the equipment is established to achieve accurate prediction of the circuit breaker's health status and remaining life.

[0056] As an optional implementation, a multi-level health index system is constructed based on the historical data and the real-time status assessment results of the circuit breaker, and a comprehensive health index is output. The multi-level health index system includes electrical life, mechanical life, and insulation status dimensions. A hybrid model combining temporal convolutional networks and long short-term memory networks is then used to perform sequence analysis on the historical data and the real-time status assessment results of the circuit breaker to predict the remaining service life of the circuit breaker. Finally, based on the multi-level health index system and the remaining service life, a personalized maintenance strategy containing early warning information and maintenance suggestions is generated as the health status information.

[0057] For example, based on the circuit breaker's historical operation records, temperature trends, vibration characteristics, and real-time condition assessment results, a multi-level health index system covering electrical life (contact wear), mechanical life (mechanical operating characteristics), and insulation condition (partial discharge activity) was constructed, and a quantified comprehensive health index (HI=0.72) was output. Deep sequence analysis of the aforementioned time-series data was performed using a hybrid model combining temporal convolutional networks and long short-term memory networks, predicting the remaining service life of the circuit breaker to be 385 operations. Based on this, a personalized strategy was generated, including a warning message that "contact wear has reached a critical value" and a maintenance suggestion that "contact inspection and mechanism lubrication are recommended within 30 days," which was then output as the final health status information to the adaptive protection decision module.

[0058] Step S400: Collect power grid operating status parameters, and combine the real-time status assessment results of the circuit breaker and the health status information to dynamically adjust the protection settings and generate protection commands.

[0059] In this embodiment, protection parameters are dynamically adjusted through an intelligent decision-making algorithm, taking into account the overall power grid operating status, circuit breaker real-time status, and health trend information. Power grid operating characteristics are obtained based on real-time power flow calculations, and combined with equipment status assessment results, an optimal protection strategy is generated using a decision model.

[0060] For example, in the field verification of a smart circuit breaker with a rated current of 63A, the test platform integrated current / voltage transformers, temperature sensors, vibration sensors, and arc sensors, and was equipped with a lightweight AI edge computing unit. System verification was conducted under harsh conditions including a wide temperature range of -25℃ to 70℃, 95% high humidity, and exposure to salt spray, dust, and vibration. After testing and verification, the multi-parameter fusion sensing module, through collaborative analysis and uncertainty fusion of multi-source heterogeneous data, significantly outperformed methods relying on single-parameter monitoring in terms of state recognition accuracy. The adaptive protection decision module, under complex operating scenarios such as frequent switching of distributed power sources and short-circuit faults at different locations, improved the accuracy of protection actions by up to 12% compared to fixed-value protection strategies. The predictive maintenance module, based on historical data and real-time state assessment results, achieved an 87.3% accuracy rate in predicting critical faults such as contact wear and mechanical operating mechanism jamming, providing reliable technical support for the transformation of the operation and maintenance model from "periodic maintenance" to "predictive maintenance."

[0061] As an optional implementation, the power grid operating status parameters are obtained based on real-time power flow calculation. These parameters include power grid short-circuit capacity, voltage stability, and distributed generation penetration. The power grid operating status parameters, the real-time status assessment results of the circuit breakers, and the health status information are input into a fuzzy logic decision system. The protection settings are dynamically adjusted using fuzzy rules. A preset protection function coordination matrix is ​​used to coordinate the logical coordination of distance protection, overcurrent protection, and zero-sequence protection based on the adjusted protection settings, and the protection command is generated.

[0062] For example, when an operating condition is identified where the short-circuit capacity on the grid side drops to 65% of the baseline value and the voltage fluctuation exceeds ±8% due to large-scale distributed photovoltaic grid connection, the fuzzy logic decision system simultaneously receives a "normal insulation temperature rise" warning from the circuit breaker's real-time status assessment and a warning from the predictive maintenance module indicating that the health index has dropped to 0.68. Through multi-dimensional fuzzy rule calculations, the overcurrent protection setting is dynamically lowered by 15%, the distance protection range is corrected to 85% of its original value, and the zero-sequence protection automatic reclosing function is blocked. Based on the above adjustments, the protection function coordination matrix generates an "Enable islanded operation protection group" instruction, ultimately forming a hierarchical protection strategy: immediately triggering overcurrent protection tripping, simultaneously sending an insulation degradation warning to the operation and maintenance platform, and prohibiting automatic reclosing operations after a fault.

[0063] In this embodiment, a leap in the overall performance of the circuit breaker protection system is achieved through multi-source information fusion and collaborative decision-making mechanisms. First, by fusing multi-dimensional sensing and uncertain information, the comprehensiveness and reliability of equipment status perception are significantly improved, resolving the limitations and false alarms inherent in single-sensor monitoring. Second, by combining real-time and historical data to construct a predictive model, accurate assessment of equipment health status and lifespan prediction are achieved, transforming the maintenance model from reactive to predictive. Finally, by integrating grid operating parameters, real-time equipment status, and health trends, an adaptive protection decision-making mechanism is constructed, enabling the circuit breaker system to dynamically optimize protection strategies based on changes in the internal and external environment. This effectively addresses the insufficient adaptability of fixed-set protection in complex grid environments, comprehensively improving the speed, selectivity, and reliability of the protection system.

[0064] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects.

[0065] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0066] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0067] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1The steps of the function specified in one or more boxes.

[0068] It should be noted that any reference signs placed between parentheses in the claims should not be construed as limiting the claims. The word "comprising" does not exclude the presence of components or steps not listed in the claims. The word "a" or "an" preceding a component does not exclude the presence of a plurality of such components. This application can be implemented by means of hardware comprising several different components and by means of a suitably programmed computer. In a unit claim enumerating several means, several of these means may be embodied by the same item of hardware. The use of the words first, second, third, etc., does not indicate any order. These words can be interpreted as names.

[0069] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application.

[0070] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of the invention. Therefore, if these modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include these modifications and variations.

Claims

1. An adaptive safety protection circuit breaker system based on multi-parameter fusion, characterized in that, The circuit breaker system includes: The multi-parameter fusion sensing module is used to collect and fuse the circuit breaker's operating status data and output the circuit breaker's real-time status assessment results. The predictive maintenance module is used to predict the health status and remaining life of the circuit breaker based on historical data and the real-time status assessment results, and generate health status information. The adaptive protection decision module is used to collect power grid operating status parameters, and combine the real-time status assessment results of the circuit breaker and the health status information to dynamically adjust the protection settings and generate protection commands.

2. The circuit breaker system as described in claim 1, characterized in that, The multi-parameter fusion sensing module includes an uncertain information fusion unit, which performs uncertain information fusion processing on multi-sensor data and outputs the real-time state assessment result; wherein, the uncertain information fusion processing includes: assigning confidence levels to data from different sensors, and performing fusion calculation on the confidence levels based on information fusion rules.

3. The circuit breaker system as described in claim 1, characterized in that, The multi-parameter fusion sensing module includes a multi-dimensional sensing network, which comprises at least three of the following: current sensor, voltage sensor, temperature sensor, vibration sensor, arc light sensor, humidity sensor, and partial discharge sensor.

4. The circuit breaker system as described in claim 1, characterized in that, The predictive maintenance module includes a health prediction model, which constructs a multi-level health index system including electrical life, mechanical life, and insulation status based on the historical data and the real-time status assessment results; and uses a hybrid model combining temporal convolutional networks and long short-term memory networks to predict the remaining service life of the circuit breaker. The health status information is generated based on the multi-level health index system and the remaining lifespan.

5. The circuit breaker system as described in claim 1, characterized in that, The adaptive protection decision module includes a power grid parameter processing unit and a protection function coordination matrix. The power grid parameter processing unit calculates the protection setting based on the power grid operating status parameters. The protection function coordination matrix uses the real-time status assessment results of the circuit breaker and the health status information as correction factors to perform sensitivity correction on the protection setting.

6. The circuit breaker system as described in claim 5, characterized in that, The adaptive protection decision module uses a fuzzy logic decision system to achieve the sensitivity correction; the input variables of the fuzzy logic decision system include the power grid operating status parameters, the real-time status assessment results of the circuit breaker, and the health status information.

7. A method for operating an adaptive safety protection circuit breaker based on multi-parameter fusion, characterized in that, The method includes the following steps: The circuit breaker's operating status data is collected through a multi-sensor array, including electrical data, mechanical data, environmental data, and fault characteristic data. The operational status data is subjected to uncertainty information fusion processing to output the real-time status assessment result of the circuit breaker; Historical data of the circuit breaker is acquired, and combined with the real-time status assessment results of the circuit breaker, the health status and remaining life of the circuit breaker are predicted, and health status information is generated; wherein, the historical data includes the historical operating characteristic parameters, historical operation records and historical maintenance records of the circuit breaker. The system collects power grid operating status parameters and, in conjunction with the real-time status assessment results of the circuit breaker and the health status information, dynamically adjusts protection settings and generates protection commands.

8. The method as described in claim 7, characterized in that, The step of performing uncertainty information fusion processing on the operating status data and outputting the real-time status assessment result of the circuit breaker includes: Feature extraction is performed on the operating status data to obtain multi-dimensional features characterizing the operating status of the circuit breaker; Based on the evidence theory algorithm, the multi-dimensional adjustments are fused to obtain the real-time status assessment result of the circuit breaker.

9. The method as described in claim 7, characterized in that, The steps of acquiring historical data of the circuit breaker, combining it with the real-time status assessment results of the circuit breaker, predicting the health status and remaining life of the circuit breaker, and generating health status information include: A multi-level health index system is constructed, and a comprehensive health index is output; wherein, the multi-level health index system includes electrical life, mechanical life and insulation status dimensions; A hybrid model combining temporal convolutional networks and long short-term memory networks is used to perform sequence analysis on the historical data and the real-time status assessment results of the circuit breaker to predict the remaining service life of the circuit breaker. Based on the multi-level health index system and the remaining service life, a personalized maintenance strategy containing early warning information and maintenance suggestions is generated as the health status information.

10. The method as described in claim 7, characterized in that, The steps of collecting power grid operating status parameters, combining the real-time status assessment results of the circuit breaker with the health status information, and dynamically adjusting protection settings and generating protection commands include: Based on real-time power flow calculation, the power grid operating status parameters are obtained, including power grid short-circuit capacity, voltage stability and distributed generation penetration rate. The power grid operating status parameters, the real-time status assessment results of the circuit breaker, and the health status information are jointly input into the fuzzy logic decision system, and the protection settings are dynamically adjusted through fuzzy rules. The protection command is generated by coordinating the logical coordination of distance protection, overcurrent protection and zero-sequence protection based on the adjusted protection settings using a preset protection function coordination matrix.

Images