Dynamic current limiting and overload adaptive protection method and system for circuit breaker

Abstract

The present application relates to the technical field of circuit breaker protection, and provides a method and system for dynamic current limiting and overload adaptive protection of circuit breakers. The method includes: according to the online monitoring network, real-time monitoring of circuit breakers to obtain status and environmental perception data; dividing the prediction time zone according to the predetermined time zone window, using the analytical algorithm and perception data to predict the fault node and determine the node; performing dynamic current limiting response compensation on the circuit breaker to obtain the current limiting mechanism; performing overload risk prediction and determining the risk node; performing overload protection response compensation on the circuit breaker to generate a protection mechanism; protecting the circuit breaker based on the current limiting and overload protection mechanism. The present application solves the technical problem that the circuit breaker lacks predictive protection before the fault occurs, and cannot dynamically adjust the current limiting and overload protection strategies in a timely and accurate manner, making it impossible to effectively deal with potential faults, and realizes the adaptive adjustment of the dynamic current limiting and overload protection of the circuit breaker, and improves the predictability and accuracy of the protection.

Description

Technical Field

[0001] The present application relates to the field of electrical protection technology, specifically to the field of circuit breaker protection technology, and in particular to a dynamic current limiting and overload adaptive protection method and system for a circuit breaker. Background Art

[0002] With the rapid development of the power industry and the continuous expansion of the power grid, the operational safety and stability of power equipment have become key issues that cannot be ignored in the power system. As a vital protection and control device in the power system, the performance of the circuit breaker is directly related to the safe operation of the power grid and the rapid isolation of faults. However, in the process of power transmission and distribution, due to the influence of various factors such as load fluctuations, short-circuit faults and extreme operating conditions, the traditional circuit breaker protection method has gradually revealed its limitations. Traditional circuit breaker protection methods, such as current fixed value protection or time-current characteristic protection, are often difficult to adapt to the higher requirements of modern power systems for protection accuracy, response speed and adaptive capabilities. These methods may malfunction or refuse to operate when faced with complex and changeable fault conditions, resulting in untimely or excessive protection actions, which in turn affects the stability and reliability of the power system. Especially in scenarios where dynamic current changes rapidly and load characteristics are complex and changeable, such as high-power motor starting, grid fault instant, etc., traditional protection methods often cannot accurately judge the type and severity of the fault, and it is difficult to achieve accurate current limiting and overload protection. This may not only aggravate equipment damage, but also trigger a chain reaction and expand the scope of the fault. Summary of the invention

[0003] The present application provides a dynamic current limiting and overload adaptive protection method and system for a circuit breaker, aiming to solve the technical problem that the circuit breaker lacks predictive protection before a fault occurs, cannot dynamically adjust the current limiting and overload protection strategies in a timely and accurate manner, and cannot effectively respond to potential faults.

[0004] In view of the above problems, the present application provides a method and system for dynamic current limiting and overload adaptive protection of a circuit breaker.

[0005] The first aspect disclosed in the present application provides a method for dynamic current limiting and overload adaptive protection of a circuit breaker, the method comprising: monitoring the circuit breaker in real time according to an online monitoring network to obtain circuit breaker state perception data and circuit breaker environment perception data; obtaining a future prediction time zone, and dividing the future prediction time zone according to a predetermined time zone window to obtain a plurality of prediction window nodes; predicting a short-circuit fault node of the circuit breaker according to a short-circuit fault node analysis algorithm, combining the plurality of prediction window nodes, the circuit breaker state perception data and the circuit breaker environment perception data, and determining the short-circuit fault node; performing dynamic current limiting response compensation on the circuit breaker based on the short-circuit fault node to obtain a dynamic current limiting mechanism; based on the plurality of prediction window nodes, performing overload risk prediction on the circuit breaker according to the circuit breaker state perception data and the circuit breaker environment perception data, and determining an overload risk node; performing overload protection response compensation on the circuit breaker based on the overload risk node to generate an overload protection mechanism; and performing feature protection on the circuit breaker based on the dynamic current limiting mechanism and the overload protection mechanism.

[0006] Another aspect disclosed in the present application provides a dynamic current limiting and overload adaptive protection system for a circuit breaker, the system comprising: a circuit breaker monitoring module, the circuit breaker monitoring module is used to monitor the circuit breaker in real time according to an online monitoring network, and obtain circuit breaker state perception data and circuit breaker environment perception data; a future prediction time zone division module, the future prediction time zone division module is used to obtain a future prediction time zone, and divide the future prediction time zone according to a predetermined time zone window to obtain a plurality of prediction window nodes; a short-circuit fault node prediction module, the short-circuit fault node prediction module is used to predict the short-circuit fault node of the circuit breaker according to a short-circuit fault node analysis algorithm, in combination with the plurality of prediction window nodes, the circuit breaker state perception data and the circuit breaker environment perception data, and determine the short-circuit fault node. A fault node; a dynamic current limiting response compensation module, the dynamic current limiting response compensation module is used to perform dynamic current limiting response compensation on the circuit breaker based on the short-circuit fault node to obtain a dynamic current limiting mechanism; an overload risk prediction module, the overload risk prediction module is used to perform overload risk prediction on the circuit breaker based on the multiple prediction window nodes and according to the circuit breaker state perception data and the circuit breaker environment perception data to determine the overload risk node; an overload protection response compensation module, the overload protection response compensation module is used to perform overload protection response compensation on the circuit breaker based on the overload risk node to generate an overload protection mechanism; a feature protection module, the feature protection module is used to perform feature protection on the circuit breaker based on the dynamic current limiting mechanism and the overload protection mechanism.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0008] The above-mentioned dynamic current limiting and overload adaptive protection method for circuit breakers is based on an online monitoring network and collects detailed operating status data of the circuit breaker and information about its environment in real time. These data provide a comprehensive perspective on the current status of the circuit breaker. Subsequently, the future time period is divided into multiple prediction windows for more detailed prediction and analysis. In each prediction window, the circuit breaker status data and environmental data are analyzed using the short-circuit fault node parsing algorithm to predict the possible short-circuit fault location. Once the short-circuit fault node is determined, the dynamic current limiting response mechanism is started to quickly respond by adjusting the current limiting parameters of the circuit breaker to prevent the fault from expanding. At the same time, the overload risk of the circuit breaker is evaluated in multiple prediction windows. By analyzing the status and environmental data, it is predicted at which time points or conditions the circuit breaker may face overload risks, and the specific nodes of overload risks are determined. Based on these predictions, the corresponding overload protection response strategy is formulated and the overload protection mechanism is generated to ensure that the circuit breaker can be protected in a timely and effective manner when an overload occurs. Finally, a comprehensive feature protection solution is provided for the circuit breaker by combining the dynamic current limiting mechanism and the overload protection mechanism. This solution can automatically adjust the protection strategy according to the real-time status and prediction results of the circuit breaker to ensure that the circuit breaker can operate safely and stably under various working conditions.

[0009] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. 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 creative work.

[0011] Figure 1 The figure is a flow chart of a method for dynamic current limiting and overload adaptive protection of a circuit breaker in one embodiment.

[0012] Figure 2 FIG. 1 is an architecture diagram of a dynamic current limiting and overload adaptive protection system for a circuit breaker in one embodiment.

[0013] Explanation of the accompanying drawings: circuit breaker monitoring module 1, future prediction time zone division module 2, short circuit fault node prediction module 3, dynamic current limiting response compensation module 4, overload risk prediction module 5, overload protection response compensation module 6, feature protection module 7. DETAILED DESCRIPTION

[0014] The embodiments of the present application provide a method and system for dynamic current limiting and overload adaptive protection of a circuit breaker to solve the technical problem that the circuit breaker lacks predictive protection before a fault occurs, cannot dynamically adjust the current limiting and overload protection strategies in a timely and accurate manner, and cannot effectively deal with potential faults.

[0015] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0016] It should be noted that the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules that are not explicitly listed or inherent to these processes, methods, products or devices.

[0017] Embodiment 1, as Figure 1 As shown, the present application provides a dynamic current limiting and overload adaptive protection method for a circuit breaker, the method comprising:

[0018] According to the online monitoring network, the circuit breaker is monitored in real time to obtain the circuit breaker status perception data and the circuit breaker environment perception data.

[0019] In an embodiment of the present application, the system terminal performs uninterrupted real-time monitoring of the circuit breaker through an online monitoring network. During this process, the monitoring network collects and transmits detailed data about the current working status of the circuit breaker, and organizes it into circuit breaker status perception data. These data include key operating parameters such as current, voltage, and temperature, which directly reflect the health status and working efficiency of the circuit breaker. At the same time, the monitoring network will also pay attention to the environmental conditions of the circuit breaker, such as ambient temperature, humidity, vibration, etc., and obtain the corresponding circuit breaker environmental perception data. These data are crucial for evaluating whether the circuit breaker is in a suitable working environment, because environmental factors may also affect the performance and life of the circuit breaker.

[0020] A future prediction time zone is obtained, and the future prediction time zone is divided according to a predetermined time zone window to obtain a plurality of prediction window nodes.

[0021] In one embodiment, in order to predict the future operation of the circuit breaker, the system terminal first sets a future time period as a prediction time zone. This time zone represents the time range that focuses on the future state of the circuit breaker. Subsequently, according to a predetermined time window size, this future prediction time zone is subdivided into multiple smaller time periods, each of which is called a prediction window. The purpose of this is to decompose the prediction task of a long time period into multiple short-term predictions to improve the accuracy and operability of the prediction. By dividing the prediction time zone, the system terminal obtains multiple prediction window nodes, which represent the prediction focus points at different time points in the future, which helps to predict the future state of the circuit breaker more finely.

[0022] According to the short-circuit fault node analysis algorithm, the short-circuit fault node of the circuit breaker is predicted in combination with the multiple prediction window nodes, the circuit breaker state perception data and the circuit breaker environment perception data to determine the short-circuit fault node.

[0023] In one embodiment, in order to predict the short-circuit fault that may occur in the circuit breaker and determine the short-circuit fault node, the system terminal will use the short-circuit fault node analysis algorithm to comprehensively consider the circuit breaker state perception data and the circuit breaker environment perception data to make a prediction. In the process of using the short-circuit fault node analysis algorithm, the system terminal uses the pre-built circuit breaker short-circuit fault probability prediction channel to predict the short-circuit fault probability of the circuit breaker according to the circuit breaker state perception data and the circuit breaker environment perception data of each prediction window node, and then performs time-related feature correction on the prediction result to obtain the final node short-circuit fault probability coefficient. Subsequently, the system terminal will obtain the node short-circuit fault probability coefficient and the short-circuit fault probability threshold for size judgment, and add the prediction window node that meets the short-circuit fault probability threshold requirements to the short-circuit fault node. In this way, measures can be taken in advance before the fault occurs to prevent the fault from expanding and ensure the safe and stable operation of the power system.

[0024] Further, the present application provides a method for predicting short-circuit fault nodes of the circuit breaker according to a short-circuit fault node analysis algorithm, combining the multiple prediction window nodes, the circuit breaker state perception data and the circuit breaker environment perception data, and determining the short-circuit fault node, including:

[0025] According to the multiple prediction window nodes, an nth prediction window node is extracted, where n is a positive integer; based on the nth prediction window node, a short-circuit fault probability of the circuit breaker is predicted according to the circuit breaker state perception data and the circuit breaker environment perception data, and an nth short-circuit fault probability coefficient is determined.

[0026] Preferably, after the division of the future prediction time zone and the acquisition of multiple prediction window nodes, the system terminal extracts the nth prediction window node from the time zone according to the order of the time zone, wherein n is a positive integer representing the current specific time point. For this nth prediction window node, the system terminal uses the M circuit breaker short circuit fault probability prediction models in the circuit breaker short circuit fault probability prediction channel to predict the short circuit fault probability of the circuit breaker state perception data and the circuit breaker environment perception data, and obtains M predicted short circuit fault probabilities. Subsequently, the accuracy of each circuit breaker short circuit fault probability prediction model is used to assign weights to the corresponding prediction results, and the M predicted short circuit fault probabilities are weighted and summed according to these weights to obtain the nth short circuit fault probability coefficient. The size of this coefficient directly reflects the possibility of a short circuit fault occurring in the circuit breaker at this time point, and provides an important reference for subsequent prevention.

[0027] Based on the nth prediction window node, the nth short-circuit fault probability coefficient is corrected for time-related characteristics to obtain the nth node short-circuit fault probability coefficient; determine whether the nth node short-circuit fault probability coefficient is greater than / equal to a short-circuit fault probability threshold; if the nth node short-circuit fault probability coefficient is greater than / equal to the short-circuit fault probability threshold, add the nth prediction window node to the short-circuit fault node.

[0028] Preferably, for the nth prediction window node, the system terminal uses the time correlation strength identification to correct the time correlation characteristics of its short-circuit fault probability coefficient. The purpose of this step is to more accurately reflect the short-circuit fault risk of the node under different time backgrounds, taking into account the influence of the time factor on the fault probability. During the correction period, the system terminal continuously monitors the nth prediction window node that does not meet the preset time correlation strength, and performs the aforementioned short-circuit fault probability prediction again based on the data of continuous detection, obtains a new nth node short-circuit fault probability coefficient, and then corrects the time correlation characteristics. This process will continue until the nth prediction window node that meets the preset time correlation strength is found. After the correction is completed, the system terminal obtains the nth node short-circuit fault probability coefficient, which is a quantitative indicator used to measure the possibility of the node having a short-circuit fault in the current prediction window. Subsequently, the system terminal compares the nth node short-circuit fault probability coefficient with a preset short-circuit fault probability threshold. This short-circuit fault probability threshold is determined based on historical experience and safety annotations, and is used to determine whether the fault risk reaches a critical value that requires attention or preventive measures. If the short-circuit fault probability coefficient of the nth node is greater than or equal to this threshold, it means that the short-circuit fault risk of the node is high and needs special attention or corresponding measures. Therefore, this node is added to the short-circuit fault node for subsequent multiple analyses.

[0029] Further, the present application provides a method for predicting the short-circuit fault probability of the circuit breaker based on the nth prediction window node and according to the circuit breaker state sensing data and the circuit breaker environment sensing data, and determining the nth short-circuit fault probability coefficient, including:

[0030] Based on the nth prediction window node, the circuit breaker state perception data is trend predicted to obtain the nth node predicted state data; based on the nth prediction window node, the circuit breaker environment perception data is trend predicted to obtain the nth node predicted environment data.

[0031] Optionally, the system terminal collects historical state perception data and historical environment perception data from the online monitoring network, and divides the historical state perception data and historical environment perception data according to random time points to obtain multiple sample prediction window nodes, multiple sample state perception data, multiple sample prediction state data, multiple sample environment perception data, and multiple sample prediction environment data, wherein the sample state perception data / sample environment perception data is the historical state perception data / historical environment perception data within a preset time range before the sample prediction window node, and the sample prediction state data is the historical state perception data / historical environment perception data within a preset time range after the sample prediction window node. Subsequently, the system terminal selects a suitable prediction model according to the complexity of the problem and the characteristics of the data, and constructs a state perception trend prediction model. Common prediction models include time series analysis models (such as ARIMA, LSTM, etc.), machine learning models (such as support vector machines, neural networks, etc.), and deep learning models. Taking the LSTM model as an example, the system terminal uses multiple sample state perception data and multiple sample prediction state data to train the LSTM model. During the training process, the parameters and structure of the LSTM model are continuously adjusted through forward propagation, loss calculation, and back propagation to optimize the prediction performance of the LSTM model. Similarly, the system terminal uses the same method to build an environmental perception trend prediction model. After that, the system terminal collects the nth state perception data and nth environmental perception data within the preset time range before the node from the online monitoring network according to the nth prediction window node, and inputs the nth state perception data and nth environmental perception data into the state perception trend prediction model and the environmental perception trend prediction model for trend prediction, and obtains the nth node prediction state data and the nth node prediction environment data. These data have important reference value for evaluating the predicted potential faults of the circuit breaker.

[0032] Activate a circuit breaker short circuit fault probability prediction channel, wherein the circuit breaker short circuit fault probability prediction channel includes M circuit breaker short circuit fault probability prediction models, wherein M is a positive integer greater than 1.

[0033] Optionally, after obtaining the nth node prediction state data and the nth node prediction environment data, the system terminal activates a pre-built circuit breaker short circuit fault probability prediction channel, which contains M different circuit breaker short circuit fault probability prediction models, where M is a positive integer greater than 1. Each prediction model is based on a different algorithm, data structure or analysis method to evaluate the risk of a circuit breaker short circuit failure. By running multiple prediction models in parallel, the system terminal can more comprehensively capture the various factors that affect the circuit breaker state and evaluate the failure risk from different angles.

[0034] In the process of building the short-circuit fault probability prediction channel of the circuit breaker, the system terminal first divides the sample prediction state data, the sample prediction environment data and the sample prediction short-circuit fault probability of the corresponding sample prediction window node recorded in the online monitoring network into a training set, a validation set and a test set. Subsequently, the system terminal determines M suitable prediction models according to the complexity of the problem and the characteristics of the data, including support vector machine, random forest, gradient boosting tree, neural network, LSTM, etc., and constructs M circuit breaker short-circuit fault probability prediction models based on these model algorithms, for example, setting the kernel function, regularization parameter C, gamma value, etc. of the support vector machine, the number of trees, the depth of the tree, the maximum number of features considered when splitting, the minimum number of samples, etc. of the random forest, the number of layers of the neural network, the number of neurons in each layer, the activation function, etc. After that, each model is trained, and the model parameters are adjusted using the training set data to minimize the loss function. During the training process, the validation set is used to evaluate the performance of the model, and hyperparameters are tuned as needed. After the training is completed, the model is evaluated using the test set, the accuracy of each model is calculated, and it is judged whether the accuracy of each model meets expectations. If satisfied, the system terminal connects the M circuit breaker short-circuit fault probability models in parallel to obtain a circuit breaker short-circuit fault probability prediction channel, and embeds a weighted sum function in the output layer of the circuit breaker short-circuit fault probability prediction channel to integrate the outputs of the M circuit breaker short-circuit fault probability prediction models and calculate the final short-circuit fault probability coefficient.

[0035] The nth node prediction state data and the nth node prediction environment data are input into the M circuit breaker short-circuit fault probability prediction models to obtain M predicted short-circuit fault probabilities; M short-circuit fault prediction accuracy parameters corresponding to the M circuit breaker short-circuit fault probability prediction models are used for proportion calculation to obtain M fault prediction accuracy incentive coefficients; and the M predicted short-circuit fault probabilities are weightedly calculated according to the M fault prediction accuracy incentive coefficients to generate the nth short-circuit fault probability coefficient.

[0036] Optionally, the system terminal inputs the predicted state data and predicted environment data of the nth node into the activated circuit breaker short-circuit fault probability prediction channel, and the circuit breaker short-circuit fault probability prediction channel transmits the predicted state data and predicted environment data of the nth node to each circuit breaker short-circuit fault probability prediction model. Each circuit breaker short-circuit fault probability prediction model generates an independent short-circuit fault probability prediction value based on these data. In this way, the system terminal obtains M different predicted short-circuit fault probabilities. Subsequently, each circuit breaker short-circuit fault probability prediction model transmits its own predicted short-circuit fault probability to the output layer of the circuit breaker short-circuit fault probability prediction channel, and the system terminal calculates the importance of the prediction accuracy of each model in the overall prediction based on the accuracy shown by each circuit breaker short-circuit fault probability prediction model in the previous test process. This calculation process is completed by converting the prediction accuracy of each model into a proportion, that is, the fault prediction accuracy incentive coefficient. The higher the fault prediction accuracy of the model, the larger its corresponding incentive coefficient, indicating that the prediction result of the model contributes more to the final result. Afterwards, the system terminal uses M fault prediction accuracy incentive coefficients and combines the weighted sum function in the output layer to perform weighted calculations on the M predicted short-circuit fault probabilities, generating an n-th node short-circuit fault probability coefficient that combines the prediction results of multiple models. This coefficient not only takes into account the prediction results of different models, but also emphasizes the importance of models with higher prediction accuracy through the form of incentive coefficients, making the final prediction results more reliable and accurate.

[0037] Further, the present application provides a method for performing time-correlated feature correction on the nth short-circuit fault probability coefficient based on the nth prediction window node to obtain the nth node short-circuit fault probability coefficient, including:

[0038] Obtain a real-time monitoring time node of the circuit breaker; based on the real-time monitoring time node, identify the time correlation strength of the nth prediction window node to determine the time correlation strength of the nth node; determine whether the time correlation strength of the nth node is less than a preset time correlation strength; if the time correlation strength of the nth node is less than the preset time correlation strength, obtain continuous state perception data and continuous environment perception data of the circuit breaker according to the online monitoring network; based on the nth prediction window node, predict the short-circuit fault probability of the circuit breaker according to the continuous state perception data and the continuous environment perception data, and generate the short-circuit fault probability coefficient of the nth node.

[0039] Optionally, the system terminal first determines the real-time monitoring time node of the circuit breaker, which provides a basis for time-related analysis and prediction. Subsequently, based on this real-time monitoring time node, the time correlation strength of the nth prediction window node is identified. Specifically, the system terminal obtains the state data and environmental data of the circuit breaker during the real-time monitoring time node and the nth prediction window node from the online monitoring network. Then sort the data under these two time nodes. For each variable, such as current, voltage, temperature, etc., sort them according to the numerical value, and record the position of each data point in the new sorted list, that is, the rank. For each corresponding data point under the real-time monitoring time node and the nth prediction window node, calculate the difference in their ranks in their respective sorted lists. The rank difference can be a positive number or a negative number, which represents the difference in the relative position of the data points under the two time nodes. Subsequently, the sum of the squares of the rank differences is calculated using the Spearman rank correlation coefficient calculation formula, and it is normalized to obtain a value between -1 and 1, that is, the time correlation strength of the nth node. The larger the absolute value of the time correlation strength of the nth node, the stronger the time correlation strength between the real-time monitoring time node and the nth prediction window node. After determining the time correlation strength of the nth node, the system terminal compares the time correlation strength of the nth node with the preset time correlation strength. This preset value is used to determine whether the correlation between the current time node and the prediction window node is strong enough to make an accurate prediction. If the time correlation strength of the nth node is less than the preset time correlation strength, it means that the correlation between the current time node and the prediction window node is not strong enough, and may be affected by some external factors or internal state changes. In this case, the system terminal will obtain the continuous state perception data and continuous environment perception data of the circuit breaker through the online monitoring network. These data are real-time and can reflect the current operating status of the circuit breaker and changes in the surrounding environment. Based on these continuous perception data, the short-circuit fault probability of the circuit breaker at the nth prediction window node is re-predicted, and the process is the same as above. Then, the system terminal will obtain an updated short-circuit fault probability coefficient of the nth node. However, if the time correlation strength of the nth node is greater than or equal to the preset time correlation strength, the system terminal will directly use the short-circuit fault probability coefficient of the nth node as the prediction result of the current time node without performing additional data acquisition and prediction processes. The purpose of this is to reduce unnecessary calculations and resource consumption.

[0040] Dynamic current limiting response compensation is performed on the circuit breaker based on the short-circuit fault node to obtain a dynamic current limiting mechanism.

[0041] In one embodiment, when a short-circuit fault node is identified, the system terminal will immediately perform dynamic current limiting response compensation for possible fault conditions. This includes evaluating the response time of the circuit breaker, etc. Through dynamic evaluation, the system terminal can have a more comprehensive understanding of the fault situation to correct the identified short-circuit fault node. Subsequently, the system terminal combines the corrected short-circuit fault node with the set dynamic current limiting strategy to generate a dynamic current limiting mechanism to ensure that the fault current can be effectively limited while reducing the impact on the normal operation of the power grid.

[0042] Furthermore, the present application provides a method for performing dynamic current limiting response compensation on the circuit breaker based on the short-circuit fault node to obtain a dynamic current limiting mechanism, including:

[0043] Obtain a dynamic current limiting strategy for the circuit breaker; perform control response verification according to the dynamic current limiting strategy to determine a confident dynamic current limiting response duration; perform a protection loss analysis on the circuit breaker according to the confident dynamic current limiting response duration to obtain a dynamic current limiting response protection loss factor.

[0044] Preferably, after identifying the short-circuit fault node, the system terminal first obtains a dynamic current limiting strategy based on the fault characteristics and the operation status of the power grid. The core goal of this strategy is to quickly cut off the fault source by disconnecting the circuit of the circuit breaker, thereby effectively limiting the rise of the short-circuit current. After obtaining the dynamic current limiting strategy, the system terminal performs control response verification. This step is to confirm the effectiveness and reliability of the strategy. Through simulation testing, the system terminal verifies whether the circuit breaker can quickly and accurately perform the disconnection operation after receiving the current limiting instruction, and limit the further rise of the short-circuit current. The result of the control response verification will determine whether the strategy needs to be adjusted or optimized to ensure that the expected current limiting effect can be achieved in actual applications. After completing the control response verification, the system terminal will determine the confidence dynamic current limiting response time according to the verification result. This time refers to the time required from the detection of the short-circuit fault to the successful disconnection of the circuit breaker and the limit of the rise of the short-circuit current. The confidence dynamic current limiting response time is an important indicator for measuring the efficiency and effect of the current limiting strategy. A shorter response time means that the system terminal can cut off the fault source faster, thereby reducing damage to the power grid and equipment. Subsequently, the system terminal defines the goal and framework of the protection loss analysis, that is, determines the type of protection loss to be evaluated, such as power grid stability loss, equipment life loss, user power supply interruption loss, etc., and sets the corresponding evaluation indicators. Then, according to the confidence dynamic current limiting response time, key data is collected for the set evaluation indicators to obtain key data such as power grid parameters, equipment characteristics, and power supply information. After that, the power system simulation software is used to simulate the dynamic response process of the power grid and equipment after the circuit breaker performs the disconnection operation within the confidence dynamic current limiting response time. By comparing the power grid parameters before and after the fault, the changes in stability indicators such as voltage fluctuation and frequency offset are evaluated, and the corresponding losses are calculated to obtain the power grid stability loss. According to the equipment life model, combined with factors such as thermal stress and mechanical stress that the equipment bears during the circuit breaker disconnection process, the reduction in equipment life is calculated to obtain the equipment life loss. Among them, the equipment life model is constructed based on a neural network, and the training steps are similar to those mentioned above. Then, according to the user power supply information and power outage time, the user power interruption duration is calculated to obtain the power interruption loss. Then, the calculated grid stability loss, equipment life loss, and power outage loss are standardized according to the maximum-minimum method, and different weights are assigned to each loss according to actual needs and expert advice to reflect its importance in the overall protection loss. Finally, each loss is multiplied by the corresponding weight and summed to obtain the dynamic current limiting response protection loss factor, which is used to quantitatively describe the potential impact of current limiting operations on the power grid and equipment. This information is of great significance for optimizing current limiting strategies and improving power grid reliability.

[0045] According to the dynamic current limiting response protection loss factor, a dynamic current limiting pre-execution threshold is set, and the short-circuit fault node is corrected according to the dynamic current limiting pre-execution threshold to obtain an optimized short-circuit fault node; the optimized short-circuit fault node and the dynamic current limiting strategy are integrated to generate the dynamic current limiting mechanism.

[0046] Preferably, the system terminal sets the calculated dynamic current limiting response protection loss factor as the dynamic current limiting pre-execution threshold. This dynamic current limiting pre-execution threshold represents the degree of loss suffered by the power system under different current limiting response times. By setting this threshold, it can be ensured that when a short-circuit fault occurs, the system terminal can respond more quickly and accurately, thereby effectively controlling the fault range and reducing protection losses. Subsequently, the processing logic of the short-circuit fault node is corrected using this pre-execution threshold. When the system terminal identifies a short-circuit fault node, a similar protection loss analysis method as described above is used to evaluate the impact of the current fault, and the evaluation result is compared with the dynamic current limiting pre-execution threshold to determine whether the dynamic current limiting strategy needs to be started in advance. If necessary, that is, the evaluation result is greater than or equal to the dynamic current limiting pre-execution threshold, the system terminal will correct the short-circuit fault node, that is, the nth prediction window node corresponding to the short-circuit fault node is moved forward by m confidence dynamic current limiting response times to obtain an optimized short-circuit fault node to adjust the operating state of the power grid in advance. Wherein, m is an integer greater than 0, usually 1, and can be adjusted according to actual needs. Afterwards, the system terminal combines the optimized short-circuit fault node with the dynamic current limiting strategy to generate a complete dynamic current limiting mechanism. This mechanism not only includes the original dynamic current limiting strategy, but also incorporates the corrected short-circuit fault node based on the protection loss factor setting, thereby achieving rapid and effective control of short-circuit faults.

[0047] Based on the multiple prediction window nodes, an overload risk prediction is performed on the circuit breaker according to the circuit breaker state perception data and the circuit breaker environment perception data to determine an overload risk node.

[0048] In one embodiment, in order to predict the possible overload risk of the circuit breaker and determine the overload risk node, the system terminal will use the overload risk node analysis algorithm to comprehensively consider the circuit breaker state perception data and the circuit breaker environment perception data to make a prediction. In the process of using the overload risk algorithm, the system terminal uses the pre-built circuit breaker overload risk prediction channel to predict the overload risk of the circuit breaker according to the circuit breaker state perception data and the circuit breaker environment perception data of each prediction window node, and then performs time-related feature correction on the prediction result to obtain the final node overload risk coefficient. Subsequently, the system terminal will obtain the node overload risk coefficient and the overload risk threshold for size judgment, and add the prediction window node that meets the overload risk threshold requirements to the overload risk node. In this way, measures can be taken in advance before the overload occurs to reduce or avoid the failure of the circuit breaker under the overload risk node.

[0049] Further, the present application provides a method for predicting an overload risk of the circuit breaker based on the multiple prediction window nodes and according to the circuit breaker state sensing data and the circuit breaker environment sensing data, and determining an overload risk node, including:

[0050] Activate a circuit breaker overload risk prediction channel, wherein the circuit breaker overload risk prediction channel includes Q circuit breaker overload risk prediction models, wherein Q is a positive integer greater than 1; based on the circuit breaker overload risk prediction channel, perform overload risk prediction on the nth prediction window node according to the circuit breaker state perception data and the circuit breaker environment perception data, and determine the nth overload risk coefficient; based on the nth prediction window node, perform time-correlated feature correction on the nth overload risk coefficient to obtain the nth node overload risk coefficient; determine whether the nth node overload risk coefficient is greater than / equal to an overload risk threshold; if the nth node overload risk coefficient is greater than / equal to the overload risk threshold, add the nth prediction window node to the overload risk node.

[0051] Preferably, when predicting the overload risk, the system terminal first activates the circuit breaker overload risk prediction channel, which consists of Q independent overload risk prediction models, where Q is a positive integer greater than 1. These models are similar to the aforementioned M circuit breaker short circuit fault probability prediction models, and are all constructed and trained based on different algorithms such as support vector machines, random forests, gradient boosting trees, neural networks, LSTM, etc., aiming to comprehensively evaluate the overload risk of circuit breakers from different angles. Subsequently, for the nth prediction window node, the system terminal uses the Q circuit breaker overload risk prediction models in the circuit breaker overload risk prediction channel to predict the overload risk of the circuit breaker state perception data and the circuit breaker environment perception data, and obtains Q predicted short circuit fault probabilities. Afterwards, the accuracy of each circuit breaker overload risk prediction model is used to assign weights to the corresponding prediction results, and the Q predicted short circuit fault probabilities are weighted and summed according to these weights to obtain the nth overload risk coefficient. The size of this coefficient directly reflects the possibility of overload of the circuit breaker at this point in time, and provides an important reference for subsequent prevention. In order to improve the accuracy of the prediction, the system terminal will also perform time-related feature correction on the calculated nth overload risk coefficient. For the nth overload risk coefficient, the system terminal uses time-related strength identification to correct the time-related features of its overload risk coefficient. The purpose of this step is to more accurately reflect the overload risk of the node under different time backgrounds, taking into account the impact of time factors on overload risk. During the correction period, the system terminal continuously monitors the nth prediction window nodes that do not meet the preset time-related strength, and performs the above-mentioned overload risk prediction again based on the continuously detected data to obtain a new nth overload risk coefficient, and then corrects the time-related features. This process will continue until the nth prediction window node that meets the preset time-related strength is found. After the correction is completed, the system terminal obtains the nth node overload risk coefficient, which is a quantitative indicator used to measure the probability of the node being overloaded in the current prediction window. Then, the system terminal compares the nth node overload risk coefficient with a preset overload risk threshold. This overload risk threshold is determined based on historical experience and safety annotations, and is used to determine whether the overload risk has reached a critical value that requires attention or preventive measures. If the overload risk coefficient of the nth node is greater than or equal to this threshold, it means that the overload risk of the node is high and needs special attention or corresponding measures. Therefore, this node is added to the overload risk node for subsequent multiple analyses.

[0052] An overload protection response compensation is performed on the circuit breaker based on the overload risk node to generate an overload protection mechanism.

[0053] In one embodiment, when an overload risk node is identified, the system terminal will immediately perform overload protection response compensation for possible overload conditions. This includes evaluating the response time of the circuit breaker, etc. Through dynamic evaluation, the system terminal can have a more comprehensive understanding of the overload situation to correct the identified overload risk node. Subsequently, the system terminal combines the corrected overload risk node with the set overload protection strategy to generate an overload protection mechanism to ensure that measures can be taken quickly and accurately to protect the circuit breaker when an overload risk occurs.

[0054] Furthermore, the present application provides an overload protection response compensation for the circuit breaker based on the overload risk node to generate an overload protection mechanism, including:

[0055] Based on the circuit breaker overload protection map, an overload protection strategy is generated; according to the overload protection strategy, control response verification is performed to determine the confident overload protection response duration; according to the confident overload protection response duration, protection loss analysis is performed on the circuit breaker to obtain an overload protection response loss factor; according to the overload protection response loss factor, an overload protection pre-execution threshold is set, and the overload risk node is corrected according to the overload protection pre-execution threshold to obtain an optimized overload risk node; the optimized overload risk node and the overload protection strategy are integrated to generate the overload protection mechanism.

[0056] Preferably, the system terminal uses the tool of circuit breaker overload protection map to comprehensively analyze the overload characteristics, historical data and current power grid conditions of the circuit breaker. Through the map, the performance of the circuit breaker under different load conditions is understood, so as to formulate targeted overload protection strategies. These strategies include setting reasonable overload current thresholds, adjusting protection action time, etc. Subsequently, the system terminal uses the same method as the above-mentioned method of obtaining the confidence dynamic current limiting response time to ensure that the overload protection strategy can quickly and accurately trigger the protection action when the circuit breaker is overloaded through simulation testing. Through verification, the confidence overload protection response time is determined, that is, the time required for the strategy to be triggered from the trigger to the actual protection action. Afterwards, the same method as the above-mentioned method of obtaining the dynamic current limiting response protection loss factor is used to collect key data according to the confidence dynamic current limiting response time to obtain key data such as overload current, power grid parameters, equipment characteristics, power supply information, etc. Then, the power system simulation software is used to simulate the dynamic response process of the power grid and equipment after the circuit breaker performs the disconnection operation within the confidence overload protection response time. By comparing the grid parameters before and after the overload, combined with the overload current and equipment characteristics, the degree of damage that the equipment may suffer due to overheating is evaluated according to the thermal stress analysis model, and the equipment damage loss is obtained. Among them, the thermal stress analysis model is constructed based on a neural network, and the training steps are similar to the above. Then, according to the user's power supply information and power outage time, the user's power supply interruption duration is calculated to obtain the power supply interruption loss. Then, the calculated equipment damage loss and power supply interruption loss are standardized according to the maximum-minimum method, and then different weights are assigned to each loss according to actual needs and expert advice to reflect its importance in the overall protection loss. Then, each loss is multiplied by the corresponding weight and summed to obtain the overload protection response loss factor. After obtaining the overload protection response loss factor, the system terminal sets the overload protection pre-execution threshold according to the same correction method as above, and judges whether the overload risk node needs to be optimized based on this set overload protection pre-execution threshold. If necessary, that is, when the evaluation result of the overload risk node is greater than or equal to the overload protection pre-execution threshold, the same correction process as above is performed to obtain the optimized overload risk node. Finally, the optimized overload risk nodes are combined with the overload protection strategy to generate a complete overload protection mechanism. This mechanism not only includes the original overload protection strategy, but also incorporates the modified overload risk nodes based on the protection loss factor setting, thereby achieving rapid and effective control of overload.

[0057] The circuit breaker is subjected to characteristic protection based on the dynamic current limiting mechanism and the overload protection mechanism.

[0058] In one embodiment, the system terminal combines the obtained dynamic current limiting mechanism with the overload protection mechanism to perform characteristic protection on the circuit breaker, so that the circuit breaker can respond quickly and accurately when facing different types of current anomalies. This dual protection mechanism improves the reliability and safety of the circuit breaker and ensures the stable operation of the power grid and equipment.

[0059] In summary, the embodiments of the present application have at least the following technical effects:

[0060] In the embodiment of the present application, the future prediction time zone is divided into multiple prediction window nodes for short-circuit fault node prediction, and the short-circuit fault node is determined by combining the analytical algorithm and the perception data. Subsequently, the circuit breaker is subjected to dynamic current limiting response compensation based on these nodes to generate a dynamic current limiting mechanism. At the same time, overload risk prediction is performed according to the prediction window node, the circuit breaker state and the environmental data to determine the overload risk node, and the overload protection mechanism is generated by response compensation. The fault probability prediction analyzes the output of multiple models and performs weighted calculation in combination with the fault prediction accuracy to improve the accuracy of the prediction. In addition, the method also involves time-related feature correction to ensure the prediction accuracy of short-circuit faults and overload risks. In the circuit breaker protection process, the protection loss is analyzed through the dynamic current limiting strategy and the overload protection strategy, and the execution threshold is adjusted after the control response verification, so as to optimize the short-circuit fault node and the overload risk node, and realize an efficient dynamic current limiting and overload protection mechanism. These technical effects jointly solve the technical problem that the circuit breaker lacks predictive protection before a fault occurs, and cannot dynamically adjust the current limiting and overload protection strategies in a timely and accurate manner, making it impossible to effectively respond to potential faults. They realize adaptive adjustment of the circuit breaker's dynamic current limiting and overload protection, and improve the predictability and accuracy of protection.

[0061] Embodiment 2 is based on the same inventive concept as the dynamic current limiting and overload adaptive protection method of the circuit breaker in the above embodiment. Figure 2 As shown, the present application provides a dynamic current limiting and overload adaptive protection system for a circuit breaker, the system comprising: a circuit breaker monitoring module 1: the circuit breaker monitoring module 1 is used to monitor the circuit breaker in real time according to an online monitoring network, and obtain circuit breaker state perception data and circuit breaker environment perception data;

[0062] Future prediction time zone division module 2: The future prediction time zone division module 2 is used to obtain a future prediction time zone, and divide the future prediction time zone according to a predetermined time zone window to obtain a plurality of prediction window nodes;

[0063] Short-circuit fault node prediction module 3: the short-circuit fault node prediction module 3 is used to predict the short-circuit fault node of the circuit breaker according to the short-circuit fault node analysis algorithm, in combination with the multiple prediction window nodes, the circuit breaker state perception data and the circuit breaker environment perception data, and determine the short-circuit fault node;

[0064] Dynamic current limiting response compensation module 4: the dynamic current limiting response compensation module 4 is used to perform dynamic current limiting response compensation on the circuit breaker based on the short-circuit fault node to obtain a dynamic current limiting mechanism;

[0065] Overload risk prediction module 5: The overload risk prediction module 5 is used to predict the overload risk of the circuit breaker based on the multiple prediction window nodes, according to the circuit breaker state perception data and the circuit breaker environment perception data, and determine the overload risk node;

[0066] Overload protection response compensation module 6: the overload protection response compensation module 6 is used to perform overload protection response compensation on the circuit breaker based on the overload risk node to generate an overload protection mechanism;

[0067] Characteristic protection module 7: The characteristic protection module 7 is used to perform characteristic protection on the circuit breaker based on the dynamic current limiting mechanism and the overload protection mechanism.

[0068] Furthermore, the short-circuit fault node prediction module 3 is also used to execute the following method:

[0069] According to the multiple prediction window nodes, an nth prediction window node is extracted, wherein n is a positive integer; based on the nth prediction window node, a short-circuit fault probability prediction is performed on the circuit breaker according to the circuit breaker state perception data and the circuit breaker environment perception data, and an nth short-circuit fault probability coefficient is determined; based on the nth prediction window node, a time-correlated feature correction is performed on the nth short-circuit fault probability coefficient to obtain the nth node short-circuit fault probability coefficient; it is determined whether the nth node short-circuit fault probability coefficient is greater than / equal to a short-circuit fault probability threshold; if the nth node short-circuit fault probability coefficient is greater than / equal to the short-circuit fault probability threshold, the nth prediction window node is added to the short-circuit fault node.

[0070] Furthermore, the short-circuit fault node prediction module 3 is also used to execute the following method:

[0071] Based on the nth prediction window node, the circuit breaker state perception data is trend predicted to obtain the nth node prediction state data; based on the nth prediction window node, the circuit breaker environment perception data is trend predicted to obtain the nth node prediction environment data; the circuit breaker short circuit fault probability prediction channel is activated, wherein the circuit breaker short circuit fault probability prediction channel includes M circuit breaker short circuit fault probability prediction models, wherein M is a positive integer greater than 1; the nth node prediction state data and the nth node prediction environment data are input into the M circuit breaker short circuit fault probability prediction models to obtain M predicted short circuit fault probabilities; according to the M short circuit fault prediction accuracy parameters corresponding to the M circuit breaker short circuit fault probability prediction models, a proportion calculation is performed to obtain M fault prediction accuracy incentive coefficients; according to the M fault prediction accuracy incentive coefficients, the M predicted short circuit fault probabilities are weightedly calculated to generate the nth short circuit fault probability coefficient.

[0072] Furthermore, the short-circuit fault node prediction module 3 is also used to execute the following method:

[0073] Obtain a real-time monitoring time node of the circuit breaker; based on the real-time monitoring time node, identify the time correlation strength of the nth prediction window node to determine the time correlation strength of the nth node; determine whether the time correlation strength of the nth node is less than a preset time correlation strength; if the time correlation strength of the nth node is less than the preset time correlation strength, obtain continuous state perception data and continuous environment perception data of the circuit breaker according to the online monitoring network; based on the nth prediction window node, predict the short-circuit fault probability of the circuit breaker according to the continuous state perception data and the continuous environment perception data, and generate the short-circuit fault probability coefficient of the nth node.

[0074] Furthermore, the dynamic current limiting response compensation module 4 is also used to execute the following method:

[0075] Obtain a dynamic current limiting strategy for the circuit breaker; perform control response verification according to the dynamic current limiting strategy to determine a confident dynamic current limiting response duration; perform protection loss analysis on the circuit breaker according to the confident dynamic current limiting response duration to obtain a dynamic current limiting response protection loss factor; set a dynamic current limiting pre-execution threshold according to the dynamic current limiting response protection loss factor, and correct the short-circuit fault node according to the dynamic current limiting pre-execution threshold to obtain an optimized short-circuit fault node; fuse the optimized short-circuit fault node and the dynamic current limiting strategy to generate the dynamic current limiting mechanism.

[0076] Furthermore, the overload risk prediction module 5 is also used to execute the following method:

[0077] Activate a circuit breaker overload risk prediction channel, wherein the circuit breaker overload risk prediction channel includes Q circuit breaker overload risk prediction models, wherein Q is a positive integer greater than 1; based on the circuit breaker overload risk prediction channel, perform overload risk prediction on the nth prediction window node according to the circuit breaker state perception data and the circuit breaker environment perception data, and determine the nth overload risk coefficient; based on the nth prediction window node, perform time-correlated feature correction on the nth overload risk coefficient to obtain the nth node overload risk coefficient; determine whether the nth node overload risk coefficient is greater than / equal to an overload risk threshold; if the nth node overload risk coefficient is greater than / equal to the overload risk threshold, add the nth prediction window node to the overload risk node.

[0078] Furthermore, the overload protection response compensation module 6 is also used to execute the following method:

[0079] Based on the circuit breaker overload protection map, an overload protection strategy is generated; according to the overload protection strategy, control response verification is performed to determine the confident overload protection response duration; according to the confident overload protection response duration, protection loss analysis is performed on the circuit breaker to obtain an overload protection response loss factor; according to the overload protection response loss factor, an overload protection pre-execution threshold is set, and the overload risk node is corrected according to the overload protection pre-execution threshold to obtain an optimized overload risk node; the optimized overload risk node and the overload protection strategy are integrated to generate the overload protection mechanism.

[0080] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. The processes depicted in the accompanying drawings do not necessarily require the specific order and continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0081] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

[0082] This specification and drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. A method for dynamic current limiting and overload adaptive protection of a circuit breaker, characterized in that: The method comprises: According to the online monitoring network, the circuit breaker is monitored in real time to obtain the circuit breaker status perception data and circuit breaker environment perception data; Obtaining a future prediction time zone, and dividing the future prediction time zone according to a predetermined time zone window to obtain a plurality of prediction window nodes; According to the short-circuit fault node analysis algorithm, the short-circuit fault node of the circuit breaker is predicted in combination with the multiple prediction window nodes, the circuit breaker state perception data and the circuit breaker environment perception data to determine the short-circuit fault node; Performing dynamic current limiting response compensation on the circuit breaker based on the short-circuit fault node to obtain a dynamic current limiting mechanism; Based on the multiple prediction window nodes, predict the overload risk of the circuit breaker according to the circuit breaker state perception data and the circuit breaker environment perception data, and determine the overload risk node; Performing overload protection response compensation on the circuit breaker based on the overload risk node to generate an overload protection mechanism; The circuit breaker is subjected to characteristic protection based on the dynamic current limiting mechanism and the overload protection mechanism.

2. The method for dynamic current limiting and overload adaptive protection of a circuit breaker according to claim 1, characterized in that: According to the short-circuit fault node analysis algorithm, the short-circuit fault node of the circuit breaker is predicted in combination with the multiple prediction window nodes, the circuit breaker state perception data and the circuit breaker environment perception data to determine the short-circuit fault node, including: Extracting an nth prediction window node according to the multiple prediction window nodes, where n is a positive integer; Based on the nth prediction window node, predict the short-circuit fault probability of the circuit breaker according to the circuit breaker state sensing data and the circuit breaker environment sensing data, and determine the nth short-circuit fault probability coefficient; Based on the nth prediction window node, performing time correlation characteristic correction on the nth short circuit fault probability coefficient to obtain the nth node short circuit fault probability coefficient; Determine whether the short circuit fault probability coefficient of the nth node is greater than / equal to a short circuit fault probability threshold; If the short-circuit fault probability coefficient of the n-th node is greater than / equal to the short-circuit fault probability threshold, the n-th prediction window node is added to the short-circuit fault node.

3. The method for dynamic current limiting and overload adaptive protection of a circuit breaker according to claim 2, characterized in that: Based on the nth prediction window node, predicting the short circuit fault probability of the circuit breaker according to the circuit breaker state sensing data and the circuit breaker environment sensing data, and determining the nth short circuit fault probability coefficient, including: Based on the nth prediction window node, performing trend prediction on the circuit breaker state sensing data to obtain nth node prediction state data; Based on the nth prediction window node, performing trend prediction on the circuit breaker environment perception data to obtain nth node prediction environment data; Activating a circuit breaker short circuit fault probability prediction channel, wherein the circuit breaker short circuit fault probability prediction channel includes M circuit breaker short circuit fault probability prediction models, wherein M is a positive integer greater than 1; Inputting the nth node prediction state data and the nth node prediction environment data into the M circuit breaker short circuit fault probability prediction models to obtain M predicted short circuit fault probabilities; Perform a proportion calculation based on the M short-circuit fault prediction accuracy parameters corresponding to the M circuit breaker short-circuit fault probability prediction models to obtain M fault prediction accuracy incentive coefficients; The M predicted short-circuit fault probabilities are weightedly calculated according to the M fault prediction accuracy incentive coefficients to generate the nth short-circuit fault probability coefficient.

4. The method for dynamic current limiting and overload adaptive protection of a circuit breaker according to claim 2, characterized in that: Based on the nth prediction window node, the nth short circuit fault probability coefficient is corrected for time correlation characteristics to obtain the nth node short circuit fault probability coefficient, including: Obtaining a real-time monitoring time node of the circuit breaker; Based on the real-time monitoring time node, identifying the time correlation strength of the nth prediction window node to determine the time correlation strength of the nth node; Determining whether the time correlation strength of the nth node is less than a preset time correlation strength; If the time correlation strength of the nth node is less than the preset time correlation strength, obtaining continuous state perception data and continuous environment perception data of the circuit breaker according to the online monitoring network; Based on the nth prediction window node, a short circuit fault probability prediction is performed on the circuit breaker according to the continuous state perception data and the continuous environment perception data to generate a short circuit fault probability coefficient of the nth node.

5. The method for dynamic current limiting and overload adaptive protection of a circuit breaker according to claim 1, characterized in that: The circuit breaker is dynamically compensated for current limiting response based on the short-circuit fault node to obtain a dynamic current limiting mechanism, including: Obtaining a dynamic current limiting strategy for the circuit breaker; Perform control response verification according to the dynamic current limiting strategy to determine the confidence dynamic current limiting response time; Performing protection loss analysis on the circuit breaker according to the confidence dynamic current limiting response time to obtain a dynamic current limiting response protection loss factor; According to the dynamic current limiting response protection loss factor, a dynamic current limiting pre-execution threshold is set, and according to the dynamic current limiting pre-execution threshold, the short-circuit fault node is corrected to obtain an optimized short-circuit fault node; The optimized short-circuit fault node and the dynamic current limiting strategy are integrated to generate the dynamic current limiting mechanism.

6. The method for dynamic current limiting and overload adaptive protection of a circuit breaker according to claim 1, characterized in that: Based on the multiple prediction window nodes, performing overload risk prediction on the circuit breaker according to the circuit breaker state perception data and the circuit breaker environment perception data, and determining an overload risk node, including: Activating a circuit breaker overload risk prediction channel, wherein the circuit breaker overload risk prediction channel includes Q circuit breaker overload risk prediction models, wherein Q is a positive integer greater than 1; Based on the circuit breaker overload risk prediction channel, an overload risk prediction is performed on the nth prediction window node according to the circuit breaker state perception data and the circuit breaker environment perception data to determine the nth overload risk coefficient; Based on the nth prediction window node, performing time-correlation feature correction on the nth overload risk coefficient to obtain the nth node overload risk coefficient; Determining whether the overload risk coefficient of the nth node is greater than / equal to an overload risk threshold; If the overload risk coefficient of the nth node is greater than / equal to the overload risk threshold, the nth prediction window node is added to the overload risk node.

7. The method for dynamic current limiting and overload adaptive protection of a circuit breaker according to claim 1, characterized in that: Performing overload protection response compensation on the circuit breaker based on the overload risk node to generate an overload protection mechanism includes: Generate overload protection strategy based on circuit breaker overload protection map; Performing control response verification according to the overload protection strategy to determine the confidence overload protection response duration; Performing protection loss analysis on the circuit breaker according to the confident overload protection response time to obtain an overload protection response loss factor; According to the overload protection response loss factor, an overload protection pre-execution threshold is set, and according to the overload protection pre-execution threshold, the overload risk node is corrected to obtain an optimized overload risk node; The optimized overload risk node and the overload protection strategy are integrated to generate the overload protection mechanism.

8. The dynamic current limiting and overload adaptive protection system of the circuit breaker is characterized by: The steps for implementing the dynamic current limiting and overload adaptive protection method for a circuit breaker according to any one of claims 1 to 7 include: Circuit breaker monitoring module: monitors the circuit breaker in real time based on the online monitoring network, and obtains circuit breaker status perception data and circuit breaker environment perception data; Future prediction time zone division module: obtains the future prediction time zone, and divides the future prediction time zone according to the predetermined time zone window to obtain multiple prediction window nodes; Short-circuit fault node prediction module: predicts the short-circuit fault node of the circuit breaker according to the short-circuit fault node analysis algorithm, combining the multiple prediction window nodes, the circuit breaker state perception data and the circuit breaker environment perception data, and determines the short-circuit fault node; Dynamic current limiting response compensation module: performs dynamic current limiting response compensation on the circuit breaker based on the short-circuit fault node to obtain a dynamic current limiting mechanism; An overload risk prediction module: based on the multiple prediction window nodes, performs overload risk prediction on the circuit breaker according to the circuit breaker state perception data and the circuit breaker environment perception data, and determines an overload risk node; An overload protection response compensation module is configured to perform overload protection response compensation on the circuit breaker based on the overload risk node to generate an overload protection mechanism; Characteristic protection module: performs characteristic protection on the circuit breaker based on the dynamic current limiting mechanism and the overload protection mechanism.

Images