Main and auxiliary equipment linkage early warning method and device in transformer substation

Abstract

The invention provides a main and auxiliary equipment linkage early warning method and device in a transformer substation, and relates to the technical field of intelligent early warning, and the method comprises the steps: building a historical work data set; establishing a linkage fitness function, a prediction and early warning fitness function and a fault positioning fitness function; positioning the main and auxiliary equipment; establishing an initial early warning identification result; taking the initial early warning identification result as a central solution, constructing N-1 candidate solutions, and establishing an initial solution set; establishing a comprehensive fitness function, performing fitness analysis on each solution in the initial solution set, and executing iterative updating of the solutions; and establishing an optimal linkage early warning strategy, and carrying out linkage early warning on the main and auxiliary equipment. Through the method and the device, the technical problem of insufficient fault early warning efficiency and accuracy caused by dependence on a single equipment monitoring means and a fault analysis method in the prior art can be solved, and the technical effects of effectively improving early warning accuracy and response speed and ensuring stable operation of substation equipment are achieved.

Description

Technical Field

[0001] The present application relates to the field of intelligent early warning technology, and in particular to a method and device for early warning linkage of main and auxiliary equipment in a substation. Background Art

[0002] With the continuous development of the power system and the increasing scale of substations, the operating environment and operating load of equipment in substations are becoming more and more complex. In this case, fault warning of substation equipment has become particularly important. Traditional equipment fault detection and warning systems often rely on a single equipment monitoring method and fault analysis method, which has certain lags and inaccuracy problems. Especially in the complex linkage relationship between main and auxiliary equipment, the failure of a single device may trigger a chain reaction of other equipment, leading to a large-scale systemic failure. Therefore, how to achieve more accurate and efficient fault warning in substations, especially the linkage warning between main and auxiliary equipment, has become a technical problem that needs to be solved urgently in the power industry.

[0003] In summary, the prior art has a technical problem of insufficient efficiency and accuracy of fault warning due to reliance on a single equipment monitoring means and fault analysis method. Summary of the invention

[0004] The purpose of this application is to provide a method and device for linkage early warning of main and auxiliary equipment in a substation, so as to solve the technical problem in the prior art of insufficient efficiency and accuracy of fault early warning due to reliance on a single equipment monitoring means and fault analysis method.

[0005] In view of the above problems, the present application provides a method and device for early warning linkage of main and auxiliary equipment in a substation.

[0006] In the first aspect, the present application provides a method for linkage early warning of main and auxiliary equipment in a substation, wherein the method for linkage early warning of main and auxiliary equipment in the substation is implemented by a linkage early warning device of main and auxiliary equipment in the substation, wherein the method for linkage early warning of main and auxiliary equipment in the substation comprises: traversing the historical working data of the equipment in the substation to establish a historical working data set; performing data cleaning of the historical working data set, and performing data identification based on the data cleaning result, and establishing a linkage fitness function, a prediction and early warning fitness function and a fault location fitness function according to the data identification; establishing the equipment health status of the substation through the historical working data set, and after real-time monitoring data collection of the equipment, locating the main and auxiliary equipment based on the real-time monitoring data set; locating the main and auxiliary equipment according to the located main and auxiliary equipment The equipment initializes an early warning identification channel, performs early warning identification on the real-time monitoring data set and the equipment health status based on the early warning identification channel, and establishes an initial early warning identification result; takes the initial early warning identification result as the central solution, constructs N-1 candidate solutions, and establishes an initial solution set with the central solution and N-1 candidate solutions, where N is an integer greater than 1; establishes a comprehensive fitness function based on the linkage fitness function, the prediction and early warning fitness function, and the fault location fitness function, performs fitness analysis on each solution in the initial solution set through the comprehensive fitness function, and performs iterative update of the solution using the fitness analysis results; establishes an optimal linkage early warning strategy based on the iterative update results, and performs linkage early warning of primary and auxiliary equipment according to the optimal linkage early warning strategy.

[0007] In the second aspect, the present application also provides a main and auxiliary equipment linkage early warning device in a substation, which is used to execute the main and auxiliary equipment linkage early warning method in the substation as described in the first aspect, wherein the main and auxiliary equipment linkage early warning device in the substation includes: a working data acquisition module, which is used to traverse the historical working data of the equipment in the substation and establish a historical working data set; a fitness function construction module, which is used to perform data cleaning of the historical working data set, and perform data identification based on the data cleaning result, and establish a linkage fitness function, a prediction and early warning fitness function and a fault location fitness function according to the data identification; a main and auxiliary equipment positioning module, which is used to establish the equipment health status of the substation through the historical working data set, and after real-time monitoring data collection of the equipment, locate the main and auxiliary equipment based on the real-time monitoring data set; an early warning identification module, which is used to locate the main and auxiliary equipment according to the located The primary and secondary equipment initializes an early warning identification channel, performs early warning identification on the real-time monitoring data set and the equipment health status based on the early warning identification channel, and establishes an initial early warning identification result; an initial solution set establishment module is used to use the initial early warning identification result as the central solution, construct N-1 candidate solutions, and establish an initial solution set with the central solution and N-1 candidate solutions, where N is an integer greater than 1; an iterative update module is used to establish a comprehensive fitness function based on the linkage fitness function, the prediction and early warning fitness function, and the fault location fitness function, perform fitness analysis on each solution in the initial solution set through the comprehensive fitness function, and use the fitness analysis results to perform iterative updates of the solutions; a linkage early warning module is used to establish an optimal linkage early warning strategy based on the iterative update results, and perform linkage early warning on the primary and secondary equipment according to the optimal linkage early warning strategy.

[0008] One or more technical solutions provided in this application have at least the following technical effects or advantages: The historical working data of the equipment of the substation is traversed to establish a historical working data set; data cleaning of the historical working data set is performed, and data identification is performed based on the data cleaning result, and a linkage fitness function, a prediction and warning fitness function and a fault location fitness function are established according to the data identification; the equipment health status of the substation is established through the historical working data set, and after real-time monitoring data of the equipment is collected, the main and auxiliary equipment are located based on the real-time monitoring data set; the warning identification channel is initialized according to the located main and auxiliary equipment, and the real-time monitoring data set and the equipment health status are identified based on the warning identification channel to establish an initial warning identification result; the initial warning identification result is used as the central solution to construct N-1 candidate solutions, and an initial solution set is established with the central solution and the N-1 candidate solutions, where N is an integer greater than 1; a comprehensive fitness function is established according to the linkage fitness function, the prediction and warning fitness function and the fault location fitness function, and the fitness analysis of each solution in the initial solution set is performed through the comprehensive fitness function, and the solution is iteratively updated using the fitness analysis result; an optimal linkage warning strategy is established based on the iterative update result, and linkage warning of the main and auxiliary equipment is performed according to the optimal linkage warning strategy. By traversing and cleaning the historical working data of substation equipment, combining real-time monitoring data with equipment health status analysis, locating main and auxiliary equipment, establishing linkage fitness function, prediction and early warning fitness function and fault location fitness function, and establishing the optimal linkage early warning strategy through iterative optimization and comprehensive fitness analysis, the technical effect of effectively improving the accuracy of early warning and response speed, and ensuring the stable operation of substation equipment is achieved.

[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 according to the contents of the specification, and 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 specifically cited below. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easy to understand through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the technical solutions in the present application or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0011] Figure 1 The figure is a flow chart of the method for early warning of linkage between main and auxiliary equipment in the substation of the present application.

[0012] Figure 2 This is a structural diagram of the main and auxiliary equipment linkage early warning device in the substation of this application.

[0013] Explanation of the reference numerals: working data acquisition module 11, fitness function construction module 12, main and auxiliary equipment positioning module 13, early warning identification module 14, initial solution set establishment module 15, iterative update module 16, linkage early warning module 17. DETAILED DESCRIPTION

[0014] This application solves the technical problem of insufficient fault warning efficiency and accuracy in the prior art due to reliance on a single equipment monitoring method and fault analysis method by providing a method and device for linkage early warning of main and auxiliary equipment in a substation. By traversing and cleaning the historical working data of substation equipment, combining real-time monitoring data with equipment health status analysis, locating main and auxiliary equipment, establishing linkage fitness functions, prediction and early warning fitness functions, and establishing the optimal linkage early warning strategy through iterative optimization and comprehensive fitness analysis, the technical effect of effectively improving the accuracy of early warning and the speed of response, and ensuring the stable operation of substation equipment is achieved.

[0015] Below, the technical solutions in the present application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments of the present application. It should be understood that the present application is not limited to the example embodiments described herein. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application. It should also be noted that, for the convenience of description, only the parts related to the present application are shown in the accompanying drawings, rather than all of them.

[0016] For example, please refer to the attached Figure 1 The present application provides a main and auxiliary equipment linkage early warning method in a substation, wherein the main and auxiliary equipment linkage early warning method in a substation is applied to a main and auxiliary equipment linkage early warning device in a substation, and the main and auxiliary equipment linkage early warning method in a substation specifically includes the following steps: Step 1: Traverse the historical working data of the equipment in the substation and establish a historical working data set.

[0017] Specifically, by traversing the historical working data of various types of equipment in the substation, the performance data of these devices under different operating conditions can be collected, including the equipment's working time, load conditions, environmental conditions, maintenance records, fault events, and alarm records.

[0018] In the process of data traversal, it is important to ensure the integrity and accuracy of the data, so it is necessary to classify and organize it for each type of equipment. For example, for transformers, it may be necessary to include multiple parameters such as its current, voltage fluctuation, oil temperature, temperature control, etc., while for circuit breakers, it may be necessary to include data such as the number of switching times, operating frequency, and degree of wear. By collecting and processing data item by item for each device, this information can be built into a comprehensive historical working data set.

[0019] Step 2: Perform data cleaning of the historical work data set, and perform data identification based on the data cleaning result, and establish a linkage fitness function, a prediction and warning fitness function, and a fault location fitness function according to the data identification.

[0020] Specifically, after the historical working data set is established, data cleaning is the next key step. Since noise, missing values, abnormal data and other problems may occur during the data collection process, these problems need to be solved through data cleaning. The purpose of data cleaning is to ensure that the data used is of high quality, accuracy and consistency, thereby providing a reliable basis for subsequent analysis and modeling. The process of data cleaning includes processing duplicate records, abnormal values ​​and missing data in the historical data set. For example, sensors of certain devices may generate abnormal data due to hardware failures or communication problems. These abnormal values ​​need to be detected and eliminated through algorithms. At the same time, for missing data, interpolation or model prediction methods can be used to fill in to ensure the integrity of the data set. The cleaned data is usually standardized or normalized to facilitate subsequent model training and analysis. Data cleaning is a common technical means used by technicians in this field and will not be expanded here.

[0021] Data identification is the next step after data cleaning. By classifying and labeling the data cleaning results, each record is given a specific identification. These identifications are usually based on the status of the equipment, working characteristics or fault type, providing effective information in subsequent predictions and warnings. Based on data identification, the linkage fitness function, prediction and warning fitness function and fault location fitness function will be established next. These fitness functions are tools used to evaluate the health status of substation equipment and play an important role in equipment linkage warning, fault prediction and fault location. Among them, the linkage fitness function is mainly used to evaluate the linkage effect between different devices. For example, when a certain device fails, it may affect other devices that are linked to it, such as the main transformer and standby transformer in the substation. The linkage fitness function analyzes the interaction between devices and calculates the fault correlation between different devices through the equipment failure conditions in historical data, which is used as the linkage fitness function.

[0022] The prediction and warning fitness function is used to analyze the time from the occurrence of equipment failure to the triggering of warning through the historical data of the equipment. Specifically, the historical data can be analyzed to establish the time relationship from the occurrence of equipment failure to the triggering of warning, and the prediction and warning fitness function can be generated. The fault location fitness function is used to analyze the accuracy of fault identification. In a substation, equipment failure may involve multiple components, and the fault location fitness function analyzes the fault identification accuracy corresponding to different warning results by analyzing the historical data of the equipment, combining the characteristics of the fault, and thereby establishing the relationship between the warning result and the fault identification accuracy, and generating the fault location fitness function.

[0023] Step 3: Establish the equipment health status of the substation through the historical working data set, and after collecting real-time monitoring data of the equipment, locate the main and auxiliary equipment based on the real-time monitoring data set.

[0024] Specifically, the health status of the equipment usually includes multi-dimensional indicators such as the equipment temperature, load, current, and voltage of the equipment under normal working conditions. Specifically, the data under normal working conditions can be extracted from the historical working data set as the health status of the equipment. Next, the equipment is monitored in real time for data collection. In the substation, various sensors are usually installed to obtain the working parameters of the equipment in real time, such as equipment temperature, load, current, and voltage, so as to form a real-time monitoring data set and locate the main and auxiliary equipment. The main equipment is usually the core equipment, such as a transformer or a circuit breaker, and the auxiliary equipment is the auxiliary equipment, such as the cooling equipment that assists the transformer or the circuit breaker. The real-time monitoring data set provides the current monitored equipment operating parameters, which can be combined with the monitoring function requirements to locate the main and auxiliary equipment. For example, it is now necessary to monitor and warn a line. The substation equipment in the line is the main equipment, and the auxiliary equipment that assists the substation equipment is the auxiliary equipment, thus obtaining the main and auxiliary equipment.

[0025] Step 4: Initialize the early warning identification channel according to the located primary and secondary devices, perform early warning identification on the real-time monitoring data set and the health status of the device based on the early warning identification channel, and establish an initial early warning identification result.

[0026] Specifically, the purpose of initializing the early warning identification channel is to establish an independent monitoring channel for each device or each type of device (main device and auxiliary device). These channels can receive the monitoring data of the device in real time, perform necessary data processing, and perform early warning identification according to the set rules and models. Each identification channel is usually personalized according to the characteristics of the device and the working environment, so as to facilitate targeted identification of the health status of the device. Different early warning identification strategies and early warning thresholds are configured in each identification channel. For example, the temperature, load, current and other data of the transformer will be monitored in the early warning identification channel of the main device. When these data exceed the preset threshold, an early warning signal will be triggered, indicating that the device may have a risk of failure. The real-time monitoring data set is collected and input into the corresponding early warning identification channel. Each early warning identification channel analyzes the input data characteristics to determine whether to trigger an early warning, thereby generating an initial early warning identification result. The initial early warning identification result may include: early warning level, fault type, fault location, etc. The initial early warning identification result is a direct reflection of real-time monitoring and health status assessment, which can provide a decision basis for subsequent maintenance, adjustment and fault handling.

[0027] Step 5: Using the initial warning identification result as the central solution, construct N-1 candidate solutions, and establish an initial solution set using the central solution and N-1 candidate solutions, where N is an integer greater than 1.

[0028] Specifically, the initial warning identification result is the warning information obtained after analyzing the real-time monitoring data and the equipment health status, which usually includes the preliminary warning identification of the main equipment and auxiliary equipment, such as the fault type, fault level and fault location. The initial warning identification result is used as the central solution to construct N-1 candidate solutions. The purpose of constructing N-1 candidate solutions is to further evaluate and optimize the current warning results through multiple possible solutions. N-1 candidate solutions are several different solutions or warning results related to the central solution, which are usually generated based on assumptions about factors such as equipment status changes, environmental conditions or load fluctuations. Each candidate solution represents a possible change scenario, which can be a small adjustment based on the central solution or a different assumption about the status of certain equipment. For example, if the central solution indicates that the main transformer is at risk of overheating, the candidate solutions may include the following situations: Candidate solution 1: The transformer temperature rises to the limit, the standby transformer is started, and the load distribution is readjusted. Candidate solution 2: The transformer load is stable, but the temperature continues to be high, which may cause the equipment to reduce the load. Candidate solution 3: The standby transformer has been started and carries part of the load, but there is still an overload risk. By taking the central solution as the core and combining N-1 candidate solutions, an initial solution set is constructed. The initial solution set is a collection of multiple possible solutions, which aims to cover various potential failure modes or warning situations, so as to facilitate a more comprehensive assessment of the health status and failure risks of the equipment in the future, and select the most appropriate response strategy to improve the safety and fault prevention capabilities of the substation.

[0029] Step six: Establish a comprehensive fitness function based on the linkage fitness function, the prediction and warning fitness function and the fault location fitness function, perform fitness analysis on each solution in the initial solution set through the comprehensive fitness function, and use the fitness analysis results to perform iterative update of the solution.

[0030] Specifically, in order to effectively optimize and iteratively update each solution, it is necessary to establish a comprehensive fitness function, which comprehensively considers multiple fitness evaluation criteria, such as linkage fitness function, prediction and warning fitness function and fault location fitness function. Specifically, the linkage fitness function, prediction and warning fitness function and fault location fitness function can be weighted, that is, a comprehensive fitness function, wherein the weights of the linkage fitness function, prediction and warning fitness function and fault location fitness function are set by professional and technical personnel in this field in combination with needs. For example, if the prediction and warning fitness function is considered to be more important, its corresponding weight is set to the maximum. Through this comprehensive fitness function, the performance of each solution can be comprehensively analyzed, thereby providing guidance for subsequent iterative updates, that is, the fitness of each solution in the initial solution set is calculated, and the fitness of each solution in the current iteration is obtained. According to the fitness, the initial solution set is iteratively updated. In short, for solutions with lower fitness values, they may be improved through optimization strategies such as genetic algorithms and simulated annealing; while solutions with higher fitness values ​​may be retained or further optimized as new initial solutions to make the warning more accurate.

[0031] Step 7: Establish an optimal linkage warning strategy based on the iterative update result, and perform linkage warning for primary and secondary equipment according to the optimal linkage warning strategy.

[0032] Specifically, based on the iterative update results, the optimal linkage warning strategy will eventually be generated. This strategy is the optimal solution obtained after multiple rounds of iterative optimization, that is, the solution with the largest fitness value. According to this strategy, linkage warning of main and auxiliary equipment is carried out. The establishment and application of the optimal linkage warning strategy will greatly improve the accuracy and timeliness of warnings, thereby effectively improving the safety and reliability of substations.

[0033] Further, step 4 of this application includes: The sub-channels in the early warning identification channel are called by utilizing the located main and auxiliary devices to generate sub-channel calling results, wherein the sub-channel calling results include a main identification sub-channel and at least one auxiliary identification sub-channel. After the early warning identification channel is identified and corrected based on the health status of the device, the real-time monitoring data set is correspondingly input into the main identification sub-channel and the auxiliary identification sub-channel for joint early warning identification to establish an initial early warning identification result.

[0034] Specifically, the early warning identification channel usually includes multiple sub-channels, each sub-channel is used to monitor and warn a device, and each sub-channel is based on an existing machine learning model, such as a neural network model. Specifically, the historical fault warning data of the device can be collected as a sample for model training, which is a common technical means for those skilled in the art and will not be expanded here. According to the positioning information of the main and auxiliary devices obtained, the early warning identification channel usually calls the sub-channel of the corresponding device to generate a sub-channel call result, and the sub-channel call result includes a main identification sub-channel and at least one auxiliary identification sub-channel. The main identification sub-channel is responsible for monitoring the status of the main device, and the auxiliary identification sub-channel is responsible for monitoring the working status of the auxiliary device. After the sub-channel is called, the early warning identification channel needs to be calibrated to ensure the accuracy of the identification result. The health status of the device is used as the basis for calibration to set the health threshold (such as the safe range of parameters such as temperature, load, and current). Specifically, the threshold and identification strategy of the early warning identification channel are adjusted according to the health status of the device. For example, if the health status of the main device is good, but the real-time monitoring data shows a certain unstable trend (such as a slight increase in temperature), the threshold of the identification channel may be adjusted to allow fluctuations within a certain range to avoid false alarms. For auxiliary devices with poor health status, the warning trigger threshold will be lowered to enhance the sensitivity of identifying its failure risk. The corrected identification channel can better adapt to the current operating status of the device and improve the accuracy of the warning.

[0035] After the sub-channel call and health status correction, the real-time monitoring data set is input into the main identification sub-channel and the auxiliary identification sub-channel for joint early warning identification, and the early warning identification results of the main equipment and auxiliary equipment are obtained to form the initial early warning identification results. Through this sub-channel call and joint early warning identification, the substation can more accurately capture the linkage risks between equipment and identify potential fault problems in advance, thereby providing more scientific and timely early warning responses for operation and maintenance personnel.

[0036] Furthermore, step five of this application includes: Obtain the warning confidence of the initial warning identification result, and establish discrete constraints based on the warning confidence; perform extreme warning analysis based on the real-time monitoring data set to establish warning limit constraints; within the warning limit constraint range, use the discrete constraints to randomly construct candidate solutions to establish N-1 candidate solutions.

[0037] Specifically, the initial warning recognition result is generated based on the real-time monitoring data and health status of the equipment. The warning trust is an indicator to measure the credibility of the warning result, indicating the reliability of the current warning result. First, the current monitoring data is compared with the historical data to evaluate the probability of similar failures. If the current state of a device is similar to the historical failure mode, the warning trust is high; otherwise, the trust is low. In other words, the similarity between the current monitoring data and the historical failure mode is used as the warning trust. Similarity analysis is a common technical means used by technicians in this field, including methods such as cosine similarity, which will not be expanded here.

[0038] Discrete constraints are a kind of constraint condition in the process of generating candidate solutions. They are usually used to screen and restrict possible fault scenarios based on the warning confidence. Discrete constraints help optimize the solutions in the solution set and avoid low-confidence candidate solutions from interfering with the final warning decision. The establishment of discrete constraints is usually based on the following rules: High-confidence restriction: When the warning confidence of the initial warning recognition result is high, the range of candidate solutions is narrower, that is, only the fault modes and fault types with high confidence are considered. At this time, it is more inclined to generate candidate solutions that are highly consistent with the current state to reduce the risk of false alarms; Low-confidence expansion: When the warning confidence is low, the range of candidate solutions will be wider, considering more possible fault modes. At this time, the search range is expanded to cover possible changes in device status to increase the probability of discovering potential faults. For example, assuming that the warning confidence of a certain device is 0.8, it means that the failure risk of the device is high. At this time, the discrete constraint will limit the generation of candidate solutions, reduce the adjustment range, and avoid overly dispersed fault modes. If the warning confidence is 0.3, the space of candidate solutions may be expanded, the adjustment range may be increased, and the response of the device in other fault scenarios may be considered. In simple terms, the construction rules of candidate solutions are determined according to the warning confidence. Candidate solutions are small adjustments based on the central solution. If the confidence is high, the adjustment range of candidate solutions can be reduced. On the contrary, if the confidence is low, the adjustment range can be increased. Specifically, we can set the corresponding adjustment range for different warning confidences based on experience.

[0039] Limit warning analysis is to predict the operating risk of the equipment under extreme conditions by performing limit value analysis on the real-time monitoring data set of the equipment, so as to further constrain the generation of failure modes. Limit warning analysis usually considers the equipment's maximum load, minimum voltage, maximum temperature and other extreme conditions. Specifically, by performing limit value analysis on various parameters (such as temperature, load, current, etc.) in the real-time monitoring data set, the operating status of the equipment under extreme conditions can be identified. For example, the maximum temperature of the transformer is 90°C. If the monitoring data shows that the temperature is close to or exceeds this value, the equipment is judged to be in an extreme condition.

[0040] Within the warning limit constraint range, discrete constraints are used to randomly construct candidate solutions to ensure that the generated candidate solutions are limited to failure modes under extreme conditions. In these extreme situations, the range of candidate solutions is controlled according to the level of warning confidence. In this way, the generation process of candidate solutions will be more in line with the extreme conditions in actual equipment operation, while avoiding low-confidence failure modes from entering the solution set. The generation process of candidate solutions is usually a randomized process, which generates N-1 candidate solutions by randomly selecting possible equipment failure modes and state changes, thereby providing multi-angle evaluation for subsequent solution set optimization and improving the accuracy and reliability of warning identification.

[0041] Furthermore, step six of this application includes: Before executing iteration, an evaluation space is established for each solution in the initial solution set, and the evaluation space is used to store the fitness value of the corresponding solution in each round of iteration; before each round of iteration, a backtracking self-check of the fitness value in the space is performed based on the evaluation space, and a backtracking self-check result is established, including: a: performing fitness rating on the fitness value in the evaluation space, and establishing a rating result; b: using the rating result to perform trend analysis, and establishing a trend analysis result; c: using the rating result to perform frequency identification of taboos and incentives, and establishing a frequency identification result; d: generating a backtracking self-check result based on the trend analysis result and the frequency identification result; creating a taboo constraint or an incentive constraint according to the backtracking self-check result, and performing iterative update according to the taboo constraint or the incentive constraint.

[0042] Specifically, the initial solution set contains multiple candidate solutions. In order to optimize these solutions during the iteration process, an evaluation space needs to be established for each solution. The evaluation space is a storage area that records the fitness value of each solution in different iteration rounds, which is used to track the performance of the solution and the optimization process. In each iteration, the solution is evaluated according to its fitness, and the fitness value is stored in the evaluation space of the corresponding solution. For example, for solution A, the fitness value calculated in the first iteration is 0.8, and the fitness value calculated in the second iteration is 0.85. These fitness values ​​will be recorded in the evaluation space of solution A. Backtracking self-checking means that before each iteration, based on the fitness value of the current solution, self-checking is performed in the evaluation space to evaluate the performance of each solution in the historical iteration process. This process involves the following key steps, which are described in detail below.

[0043] a: Perform fitness rating on the fitness values ​​recorded in the evaluation space for each solution, that is, classify and sort these values. The ratings are usually divided into different levels, such as: Excellent: the fitness value is close to the maximum value, indicating that the solution has a high quality; Good: the fitness value is medium, indicating that the solution may need further optimization; Poor: the fitness value is low, indicating that the quality of the solution is poor and may need to be eliminated. The fitness rating process helps identify which solutions perform well in the iteration process and which solutions may need further adjustment or elimination.

[0044] b: Trend analysis. After the fitness rating is performed, the rating results are used for trend analysis, that is, to identify the performance trend of the solution in multiple rounds of iterations to predict the possible direction of the solution in future iterations. For example, the upward or downward trend of the fitness value can be evaluated to determine whether a solution is converging to the optimal solution, or whether there is a trend of fluctuation and instability, to identify whether the solution has entered a local optimal solution, or whether premature convergence has occurred, and then help formulate the next optimization strategy as the trend analysis result.

[0045] c: Taboo and incentive frequency identification, that is, identifying which solutions frequently show poor fitness (taboo solutions) in past iterations, and which solutions show strong improvement potential (incentive solutions). Specifically, if the fitness value of a solution is always low in multiple iterations and fails to improve effectively, the solution is marked as a taboo solution, and the frequency of fitness change with the number of iterations is counted; if a solution shows a trend of continuous improvement during the iteration process, the solution is marked as an incentive solution, and the frequency of fitness change with the number of iterations is also counted, thereby obtaining the frequency identification result.

[0046] d: Based on the trend analysis result and the frequency identification result, a backtracking self-check result is generated, which summarizes the performance of each solution in past iterations and determines whether a taboo constraint or an incentive constraint needs to be imposed on the solution.

[0047] Finally, if the backtracking self-checking results show that the fitness value of a solution is always low in multiple iterations, and the trend analysis results show that it is trapped in a local optimal solution, then these solutions are imposed with taboo constraints to restrict their continued appearance in future iterations, avoid repeated attempts at invalid solutions, and reduce invalid expansion of the search space. If the backtracking self-checking results show that a solution shows a trend of continuous improvement in multiple iterations, the solution is marked as an incentive solution, and incentive constraints are imposed, so that more attention is paid to the optimization direction of the solution in subsequent iterations. For example, the fitness of solution B is 0.6 in the first round of iteration, and the fitness is increased to 0.75 in the second round of iteration, and the improvement trend is relatively stable. Solution B is marked as an incentive solution, and further exploration of the optimization path of the solution is encouraged.

[0048] Finally, according to the taboo constraints and incentive constraints, the solutions in the initial solution set are iteratively updated. The specific steps are as follows: For taboo solutions, they are excluded from the solution set according to the taboo constraints, and no subsequent iterative optimization is performed. The removal of taboo solutions can help reduce unnecessary calculations and improve iteration efficiency. For incentive solutions, more attention is paid to the solution, and its optimization process may be further accelerated by adjusting the search strategy, optimization algorithm or resource allocation, so as to converge to the optimal solution as soon as possible. In the iterative process, each round of updates is based on the fitness analysis and backtracking self-check results of the previous round, and the overall quality of the solution set is gradually improved. As the iteration proceeds, the solution set is continuously optimized, and finally the optimal main and auxiliary equipment linkage warning strategy is selected.

[0049] By creating an evaluation space for the initial solution set before each round of iteration, performing backtracking self-checking on the fitness value, and imposing taboo constraints and incentive constraints based on the backtracking self-checking results, the solution set can be continuously optimized, the accuracy and efficiency of early warning identification can be improved, the dilemma of local optimal solutions can be effectively avoided, and the solution set can be promoted to gradually converge to the optimal solution during the iteration process, thereby improving the intelligent early warning capability.

[0050] Furthermore, the present application also includes the following steps: Obtain the current number of iterations, and perform an iterative state self-check based on the fitness analysis results of the solutions of the previous iteration, and construct an exploration constraint based on the iterative state self-check results and the current number of iterations; obtain the constraint state of each solution, and if an incentive constraint exists, generate a joint analysis instruction; based on the joint analysis instruction, perform a joint analysis based on the incentive constraint and the fitness analysis results of the solutions of the previous iteration, and establish a following constraint; perform iterative update based on the exploration constraint and the following constraint.

[0051] Specifically, the current iteration number is the core timeline of the entire optimization process, which is used to control the update progress of the solution and the fitness convergence. In each round of iteration, the current iteration number will be recorded and combined with the fitness analysis results of the solution in the previous round of iteration to perform iteration status self-check. The main contents of the self-check include: Fitness distribution evaluation: Analyze whether the fitness values ​​of the solutions in the previous round of iteration are evenly distributed. For example, if the fitness values ​​of all solutions are concentrated in a certain range, it may mean that they are trapped in a local optimal solution; Solution diversity evaluation: Check whether the solution set maintains sufficient diversity. If the parameters of the solution tend to be homogenized, it may make it difficult for subsequent iterations to break through the existing solution space; Convergence speed evaluation: Evaluate the improvement speed of the fitness value. If the fitness improvement tends to be flat or stagnant in multiple rounds of iterations, it may be necessary to adjust the constraint strategy to explore new solution spaces. Based on the self-check results, an iteration status self-check report is generated to guide the construction of subsequent exploration constraints and following constraints.

[0052] The exploration constraint is a solution expansion constraint constructed to guide the optimization direction of the solution set in the current iteration. Especially when the number of iterations is low, the exploration constraint can encourage the solution set to expand to a wider solution space to avoid falling into the local optimal solution. If the current number of iterations is low, set a loose exploration constraint to allow the solution set to explore in a larger search space. For example, the amplitude of random mutation can be increased or more candidate solutions can be introduced. If the current number of iterations is close to the middle value, the exploration constraint will gradually narrow, focusing on the solution space around the high fitness solution for local optimization. If the number of iterations is close to the upper limit, the exploration constraint will be further tightened, focusing on optimizing the current optimal solution and concentrating resources on fine-tuning these solutions. For example, in the early iterations, the exploration constraint may allow solutions with large differences in fitness values ​​to enter the next round, while in the later iterations, the exploration constraint will strictly limit solutions with lower fitness values ​​and retain more promising solutions.

[0053] Furthermore, the constraint state of each solution is obtained, which may include incentive constraints or taboo constraints. If an incentive constraint is found in the constraint state of the solution, a joint analysis instruction is generated. The purpose is to combine the incentive constraint with the fitness analysis results to conduct an in-depth evaluation of the optimization direction of the solution. For example, if solution A shows a continuously improving fitness value, the joint analysis instruction may require solution A to be combined and crossed with other solutions to explore a better solution. The results of the joint analysis are used to construct a following constraint. The following constraint is a dynamic constraint. The purpose is to allow the solution set to follow the solution that shows strong improvement potential during the optimization process, thereby improving the overall quality of the solution set. The establishment of the following constraint includes the following steps: Prioritized optimization of high-fitness solutions: According to the results of the joint analysis, high-fitness solutions are retained or optimized first, and the solution space is expanded with these solutions as the center; Auxiliary adjustment of low-fitness solutions: For solutions with low fitness values ​​but still potential, the following constraint will adjust them to variants of high-fitness solutions to continue exploring potential optimization paths. For example, if the fitness value of solution B shows a large room for improvement in the joint analysis, following the constraint may guide the generation of a new solution B' that is similar to solution B but with slightly adjusted parameters to further optimize the potential direction of solution B.

[0054] Finally, the solution set is iteratively updated in combination with the exploration constraint and the pursuit constraint. The specific operations include: selecting solutions with better performance to enter the next round of iteration according to the fitness value and the pursuit constraint; cross-combining the selected solutions to generate new candidate solutions; introducing appropriate random mutations in the solution set to increase the diversity of solutions; elimination operation: according to the exploration constraint and the pursuit constraint, eliminate solutions with low fitness values ​​or restricted by taboo constraints, and retain more promising solutions. For example, assuming that in a certain round of iterations, solution C shows a trend of continuously improving fitness, the solution C is locally optimized according to the pursuit constraint, and solutions C' and C'' are generated through the crossover operation, and their fitness are evaluated. In the next round of iterations, the solution with higher fitness will be retained, so as to complete the iterative update and obtain the iterative update result.

[0055] By obtaining the fitness analysis results of the current iteration number and the previous iteration, the exploration constraints and following constraints can be dynamically adjusted to ensure that the solution set can both widely explore the solution space and concentrate on optimizing high-potential solutions during the iteration process, thereby improving the accuracy and efficiency of the linkage early warning of the main and auxiliary equipment of the substation.

[0056] Furthermore, the present application also includes the following steps: If there is a taboo constraint, an elimination instruction is generated, and the corresponding solution is subjected to taboo elimination according to the elimination instruction, and a reconstruction instruction is generated; the following constraint is compensated according to the reconstructed new solution, and an iterative update is performed with the compensated following constraint.

[0057] Specifically, the taboo constraint constrains low-quality solutions in the current solution set to prevent them from participating in subsequent iterations, especially when some solutions show a continuous decline in fitness or fall into a local optimum. The role of the taboo constraint is to prevent these unqualified solutions from entering the next round of optimization process. When the fitness value of some solutions fails to improve significantly after multiple rounds of iterations, and the change trend is negative, an elimination instruction is generated according to the taboo constraint to filter it out of the solution set. For example, the fitness value of solution A has changed little in the past few rounds of iterations and has not been effectively improved. The elimination instruction is generated according to the taboo constraint to remove solution A from the current solution set and obtain a reconstructed new solution. The following constraint is compensated according to the reconstructed new solution to ensure that the solution can continue to be optimized in an effective direction under the new solution set. The following constraint is usually used to guide the search and follow of high-fitness solutions in the solution space. The compensated following constraint will adjust the original constraint conditions to adapt to the newly generated solution set. In the compensation process, the goal of the following constraint is to ensure that the new solution can continue to follow the excellent solution and perform effective optimization, rather than deviate from the existing good solution space. The compensated following constraints will dynamically adjust the optimization direction of the solution. For example, in the generated new solutions, it is found that although a solution is generated by the cross-generation of two high-fitness solutions, its performance may not meet expectations. By adjusting the following constraints, the solution is promoted to optimize towards a better solution. The compensation method can include expanding or reducing the optimization space for certain specific solutions, or giving greater weights to certain parameters, thereby starting a new iterative update and further optimizing based on the new solution set. This can significantly improve the reliability of substation equipment fault warnings, reduce false alarms and missed alarms, and improve overall stability and safety.

[0058] Furthermore, the present application also includes the following steps: Perform warning configuration according to the optimal linkage warning strategy, the warning configuration including warning signal feature configuration and warning transmission path configuration; perform warning alarm according to the warning configuration, and generate warning monitoring; if the warning monitoring fails to meet the preset feedback threshold, generate an abnormal report; after dynamically adjusting the warning configuration according to the abnormal report, execute the warning alarm again.

[0059] Specifically, once the optimal linkage warning strategy is established, the next key step is to configure the warning, in order to ensure that appropriate warning signals can be effectively issued and transmitted through effective paths. Among them, the warning signal feature configuration involves the nature, characteristics and content of the warning signal. It is necessary to define different types of warning signals and set different warning signal features for different fault types (such as equipment overload, temperature abnormality, current fluctuation, etc.). Warning transmission path configuration: Once the warning signal is generated, it is crucial to quickly transmit the signal to the relevant personnel. The warning transmission path configuration determines the transmission channel and recipient of the warning signal. Different types of warning signals may need to be transmitted through different channels. For example, some urgent and high-priority warning signals may need to immediately notify the on-duty personnel or maintenance team of the substation, and transmit them through SMS, phone calls or on-site alarms. Relatively minor fault warning signals may be transmitted to equipment maintenance personnel via email or automated monitoring systems for subsequent processing. Once the warning configuration is completed, the warning and monitoring operations are executed, that is, the warning signal is automatically issued according to the optimal linkage warning strategy and the configured warning signal characteristics. The warning process includes signal generation, path selection, and notification issuance. According to the real-time status and health of the equipment, the warning information is generated in time and transmitted to relevant personnel through the preset path. Warning monitoring is the real-time tracking of the issued warning signal. Warning monitoring will continuously check the effectiveness of the warning signal and monitor whether the warning target is achieved in time, for example, tracking whether the warning signal is successfully conveyed to the receiving personnel, whether there are appropriate response measures, or whether the warning signal is misunderstood or ignored.

[0060] Although the initial signal is sent out through the early warning configuration, the early warning process is not always perfect, and there may be some problems or room for improvement. Therefore, it is necessary to establish a feedback mechanism and dynamically adjust the early warning configuration according to the feedback. The preset feedback threshold is a feedback threshold, that is, in order to ensure the effectiveness of the early warning signal, a feedback threshold is set, that is, the early warning signal must meet a certain standard to be considered a successful early warning. For example, the early warning signal may need to be confirmed by the receiving personnel within the specified time, and emergency measures must be taken within the specified time. If the early warning signal does not meet these standards, it is considered a warning failure and subsequent adjustments are required. When the early warning monitoring finds that the early warning signal fails to meet the preset feedback threshold, an abnormal report is generated. This may occur in a variety of scenarios, such as: the early warning signal fails to be delivered in time; the receiving personnel fail to respond to the early warning within the specified time.

[0061] Based on the results of abnormal reports, the early warning configuration is adjusted dynamically, for example, the severity level of the early warning signal is modified, the priority of the early warning is changed, and the early warning transmission path is adjusted. If some transmission paths are unreliable or slow to respond, they can be adjusted to more efficient paths and the early warning is issued again. By dynamically adjusting the early warning configuration, the accuracy of the early warning can be continuously improved, the false alarms and missed alarms can be reduced, the equipment safety can be effectively improved, and the potential risks can be reduced.

[0062] In summary, the main and auxiliary equipment linkage early warning method in the substation provided by the present application has the following technical effects: The historical working data of the equipment of the substation is traversed to establish a historical working data set; data cleaning of the historical working data set is performed, and data identification is performed based on the data cleaning result, and a linkage fitness function, a prediction and warning fitness function and a fault location fitness function are established according to the data identification; the equipment health status of the substation is established through the historical working data set, and after real-time monitoring data of the equipment is collected, the main and auxiliary equipment are located based on the real-time monitoring data set; the warning identification channel is initialized according to the located main and auxiliary equipment, and the real-time monitoring data set and the equipment health status are identified based on the warning identification channel to establish an initial warning identification result; the initial warning identification result is used as the central solution to construct N-1 candidate solutions, and an initial solution set is established with the central solution and the N-1 candidate solutions, where N is an integer greater than 1; a comprehensive fitness function is established according to the linkage fitness function, the prediction and warning fitness function and the fault location fitness function, and the fitness analysis of each solution in the initial solution set is performed through the comprehensive fitness function, and the solution is iteratively updated using the fitness analysis result; an optimal linkage warning strategy is established based on the iterative update result, and linkage warning of the main and auxiliary equipment is performed according to the optimal linkage warning strategy. By traversing and cleaning the historical working data of substation equipment, combining real-time monitoring data with equipment health status analysis, locating main and auxiliary equipment, establishing linkage fitness function, prediction and early warning fitness function and fault location fitness function, and establishing the optimal linkage early warning strategy through iterative optimization and comprehensive fitness analysis, the technical effect of effectively improving the accuracy of early warning and response speed, and ensuring the stable operation of substation equipment is achieved.

[0063] Embodiment 2, based on the same inventive concept as the main and auxiliary equipment linkage early warning method in the substation in the previous embodiment, the present application also provides a main and auxiliary equipment linkage early warning device in the substation, please refer to the attached Figure 2 The main and auxiliary equipment linkage early warning device in the substation includes: The working data acquisition module 11 is used to traverse the historical working data of the equipment in the substation and establish a historical working data set.

[0064] The fitness function building module 12 is used to perform data cleaning of the historical working data set, and to perform data identification based on the data cleaning result, and to establish a linkage fitness function, a prediction and warning fitness function, and a fault location fitness function according to the data identification.

[0065] The main and auxiliary equipment positioning module 13 is used to establish the equipment health status of the substation through the historical working data set, and after collecting real-time monitoring data of the equipment, locate the main and auxiliary equipment based on the real-time monitoring data set.

[0066] The early warning identification module 14 is used to initialize the early warning identification channel according to the located primary and secondary devices, perform early warning identification on the real-time monitoring data set and the device health status based on the early warning identification channel, and establish an initial early warning identification result.

[0067] The initial solution set establishing module 15 is used to construct N-1 candidate solutions using the initial warning identification result as the central solution, and to establish an initial solution set using the central solution and the N-1 candidate solutions, where N is an integer greater than 1.

[0068] The iterative update module 16 is used to establish a comprehensive fitness function based on the linkage fitness function, the prediction and warning fitness function and the fault location fitness function, perform fitness analysis on each solution in the initial solution set through the comprehensive fitness function, and use the fitness analysis results to perform iterative update of the solution.

[0069] The linkage warning module 17 is used to establish an optimal linkage warning strategy based on the iterative update result, and perform linkage warning of the main and auxiliary equipment according to the optimal linkage warning strategy.

[0070] Further, the initial solution set establishment module 15 in the main and auxiliary equipment linkage early warning device in the substation is also used for: Obtain the warning confidence of the initial warning identification result, and establish discrete constraints based on the warning confidence; perform extreme warning analysis based on the real-time monitoring data set to establish warning limit constraints; within the warning limit constraint range, use the discrete constraints to randomly construct candidate solutions to establish N-1 candidate solutions.

[0071] Further, the iterative update module 16 in the primary and auxiliary equipment linkage early warning device in the substation is also used for: Before executing iteration, an evaluation space is established for each solution in the initial solution set, and the evaluation space is used to store the fitness value of the corresponding solution in each round of iteration; before each round of iteration, a backtracking self-check of the fitness value in the space is performed based on the evaluation space, and a backtracking self-check result is established, including: a: performing fitness rating on the fitness value in the evaluation space, and establishing a rating result; b: using the rating result to perform trend analysis, and establishing a trend analysis result; c: using the rating result to perform frequency identification of taboos and incentives, and establishing a frequency identification result; d: generating a backtracking self-check result based on the trend analysis result and the frequency identification result; creating a taboo constraint or an incentive constraint according to the backtracking self-check result, and performing iterative update according to the taboo constraint or the incentive constraint.

[0072] Further, the iterative update module 16 in the primary and auxiliary equipment linkage early warning device in the substation is also used for: Obtain the current number of iterations, and perform an iterative state self-check based on the fitness analysis results of the solutions of the previous iteration, and construct an exploration constraint based on the iterative state self-check results and the current number of iterations; obtain the constraint state of each solution, and if an incentive constraint exists, generate a joint analysis instruction; based on the joint analysis instruction, perform a joint analysis based on the incentive constraint and the fitness analysis results of the solutions of the previous iteration, and establish a following constraint; perform iterative update based on the exploration constraint and the following constraint.

[0073] Further, the iterative update module 16 in the primary and auxiliary equipment linkage early warning device in the substation is also used for: If there is a taboo constraint, an elimination instruction is generated, and the corresponding solution is subjected to taboo elimination according to the elimination instruction, and a reconstruction instruction is generated; the following constraint is compensated according to the reconstructed new solution, and an iterative update is performed with the compensated following constraint.

[0074] Furthermore, the warning identification module 14 in the main and auxiliary equipment linkage warning device in the substation is also used for: The sub-channels in the early warning identification channel are called by utilizing the located main and auxiliary devices to generate sub-channel calling results, wherein the sub-channel calling results include a main identification sub-channel and at least one auxiliary identification sub-channel. After the early warning identification channel is identified and corrected based on the health status of the device, the real-time monitoring data set is correspondingly input into the main identification sub-channel and the auxiliary identification sub-channel for joint early warning identification to establish an initial early warning identification result.

[0075] Furthermore, the linkage warning module 17 in the main and auxiliary equipment linkage warning device in the substation is also used for: Perform warning configuration according to the optimal linkage warning strategy, the warning configuration including warning signal feature configuration and warning transmission path configuration; perform warning alarm according to the warning configuration, and generate warning monitoring; if the warning monitoring fails to meet the preset feedback threshold, generate an abnormal report; after dynamically adjusting the warning configuration according to the abnormal report, execute the warning alarm again.

[0076] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. Figure 1 The main and auxiliary equipment linkage early warning method and specific examples in the substation in the first embodiment are also applicable to the main and auxiliary equipment linkage early warning device in the substation in this embodiment. Through the above detailed description of the main and auxiliary equipment linkage early warning method in the substation, those skilled in the art can clearly know the main and auxiliary equipment linkage early warning device in the substation in this embodiment, so for the sake of brevity of the specification, it will not be described in detail here. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description.

[0077] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

[0078] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application belong to the scope of the present application and its equivalent technology, the present application is also intended to include these modifications and variations.

Claims

1. A method for early warning linkage of main and auxiliary equipment in a substation, characterized in that: The main and auxiliary equipment linkage early warning method in the substation includes: Traverse the historical working data of the equipment in the substation and establish a historical working data set; Execute data cleaning of the historical working data set, perform data identification based on the data cleaning result, and establish a linkage fitness function, a prediction and warning fitness function, and a fault location fitness function according to the data identification; Establishing the equipment health status of the substation through the historical working data set, and locating the main and auxiliary equipment based on the real-time monitoring data set after collecting real-time monitoring data of the equipment; Initialize the early warning identification channel according to the located primary and secondary devices, perform early warning identification on the real-time monitoring data set and the health status of the device based on the early warning identification channel, and establish an initial early warning identification result; Taking the initial warning identification result as the central solution, constructing N-1 candidate solutions, and establishing an initial solution set with the central solution and the N-1 candidate solutions, where N is an integer greater than 1; Establishing a comprehensive fitness function according to the linkage fitness function, the prediction and warning fitness function and the fault location fitness function, performing fitness analysis on each solution in the initial solution set through the comprehensive fitness function, and performing iterative update of the solution using the fitness analysis result; An optimal linkage warning strategy is established based on the iterative update result, and linkage warning of primary and secondary equipment is performed according to the optimal linkage warning strategy.

2. The method for early warning of main and auxiliary equipment linkage in a substation according to claim 1, characterized in that: The initial warning identification result is used as the central solution to construct N-1 candidate solutions, including: Obtaining the warning confidence of the initial warning identification result, and establishing discrete constraints according to the warning confidence; Performing limit warning analysis based on the real-time monitoring data set and establishing warning limit constraints; Within the warning limit constraint range, the discrete constraint is used to randomly construct candidate solutions to establish N-1 candidate solutions.

3. The main and auxiliary equipment linkage early warning method in a substation according to claim 1, characterized in that: The method of performing iterative update of the solution using the fitness analysis result comprises: Before executing the iteration, an evaluation space is established for each solution in the initial solution set, and the evaluation space is used to store the fitness value of the corresponding solution in each round of iteration; Before each iteration, a fitness value backtracking self-check is performed in the evaluation space to establish a backtracking self-check result, including: a: Perform fitness rating on the fitness values ​​in the evaluation space and establish rating results; b: using the rating results to conduct trend analysis and establish trend analysis results; c: using the rating results to identify the frequencies of taboos and incentives, and establishing frequency identification results; d: generating a retrospective self-check result based on the trend analysis result and the frequency identification result; A taboo constraint or an incentive constraint is created according to the backtracking self-check result, and an iterative update is performed according to the taboo constraint or the incentive constraint.

4. The main and auxiliary equipment linkage early warning method in a substation according to claim 3, characterized in that: The main and auxiliary equipment linkage early warning method in the substation also includes: Get the current number of iterations, perform a self-check on the iteration status based on the fitness analysis results of the solution in the previous iteration, and build exploration constraints based on the iteration status self-check results and the current number of iterations; Obtain the constraint status of each solution, and if there is an incentive constraint, generate a joint analysis instruction; Based on the joint analysis instruction, a joint analysis is performed according to the incentive constraint and the fitness analysis result of the solution of the previous iteration to establish a following constraint; An iterative update is performed based on the exploration constraint and the pursuit constraint.

5. The method for early warning of main and auxiliary equipment linkage in a substation according to claim 4, characterized in that: The obtaining of the constraint state of each solution includes: If there is a taboo constraint, an elimination instruction is generated, and the corresponding solution is eliminated according to the elimination instruction, and a reconstruction instruction is generated; The following constraint is compensated according to the reconstructed new solution, and iterative update is performed with the compensated following constraint.

6. The method for early warning of main and auxiliary equipment linkage in a substation according to claim 1, characterized in that: Initializing the early warning identification channel according to the located primary and secondary devices, performing early warning identification on the real-time monitoring data set and the health status of the device based on the early warning identification channel, and establishing an initial early warning identification result, includes: Using the located primary and secondary devices to call the sub-channels in the early warning identification channel, generating a sub-channel calling result, wherein the sub-channel calling result includes a primary identification sub-channel and at least one auxiliary identification sub-channel; After the early warning identification channel is identified and corrected based on the health status of the equipment, the real-time monitoring data set is input into the main identification sub-channel and the auxiliary identification sub-channel for joint early warning identification to establish an initial early warning identification result.

7. The method for early warning of main and auxiliary equipment linkage in a substation according to claim 1, characterized in that: The main and auxiliary equipment linkage early warning method in the substation also includes: Performing early warning configuration according to the optimal linkage early warning strategy, wherein the early warning configuration includes early warning signal feature configuration and early warning transmission path configuration; Issue early warnings according to the early warning configuration and generate early warning monitoring; If the early warning monitoring fails to meet the preset feedback threshold, an abnormality is generated; After dynamically adjusting the warning configuration according to the reported abnormality, the warning alarm is executed again.

8. The main and auxiliary equipment linkage early warning device in the substation is characterized by: Steps for implementing the method for linkage early warning of main and auxiliary equipment in a substation according to any one of claims 1 to 7, wherein the linkage early warning device of main and auxiliary equipment in the substation comprises: The working data acquisition module is used to traverse the historical working data of the equipment in the substation and establish a historical working data set; A fitness function building module, used to perform data cleaning of the historical working data set, and to perform data identification based on the data cleaning result, and to establish a linkage fitness function, a prediction and warning fitness function, and a fault location fitness function according to the data identification; A main and auxiliary equipment positioning module, used to establish the equipment health status of the substation through the historical working data set, and after collecting real-time monitoring data of the equipment, locate the main and auxiliary equipment based on the real-time monitoring data set; An early warning identification module is used to initialize an early warning identification channel according to the located primary and secondary devices, perform early warning identification on the real-time monitoring data set and the health status of the device based on the early warning identification channel, and establish an initial early warning identification result; An initial solution set establishment module, used to construct N-1 candidate solutions using the initial warning identification result as a central solution, and to establish an initial solution set using the central solution and the N-1 candidate solutions, where N is an integer greater than 1; An iterative update module, used to establish a comprehensive fitness function according to the linkage fitness function, the prediction and warning fitness function and the fault location fitness function, perform fitness analysis on each solution in the initial solution set through the comprehensive fitness function, and perform iterative update of the solution using the fitness analysis result; The linkage warning module is used to establish an optimal linkage warning strategy based on the iterative update result, and perform linkage warning of primary and secondary equipment according to the optimal linkage warning strategy.

Images