Intelligent Recommendation Method, Device and Medium for Load Transfer Scheme Based on Machine Learning

Abstract

The present invention discloses an intelligent recommendation method, device and medium for load transfer schemes based on machine learning. The steps of the method include: Step S01. Obtain the characteristic parameters of multiple different load transfer schemes and construct a multi-dimensional feature set; Step S02. Calculate the performance scores of the load transfer schemes to be evaluated; Step S03. After combining every two load transfer schemes, splice the feature vectors to form combined feature vectors; Step S04. Determine the classification labels of the classification model according to the magnitude relationship of the performance scores of every two load transfer schemes; Step S05. Train the classification model to obtain a load transfer scheme classification model; Step S06. Obtain multiple load transfer schemes to be recommended, input them into the trained classification model, sort each scheme, and generate a recommendation result. The present invention has the advantages of simple implementation method, low cost, high intelligence level and fast response speed, etc.

Description

Technical Field

[0001] The present invention relates to the technical field of smart grids, and particularly to an intelligent recommendation method, device and medium for load transfer schemes based on machine learning. Background Art

[0002] A load transfer scheme is a technical solution in a power system to transfer the load of one power user to another. By adjusting the load distribution, it realizes the balance and optimization of the power system. As a part of power dispatching, distribution network dispatching needs to perform accurate load forecasting and formulate load transfer schemes to ensure the stable operation of the power system. Therefore, formulating an efficient and reliable load transfer scheme is crucial for ensuring the safe and stable operation of the power grid.

[0003] Currently, the structure of the new power system is becoming more and more complex, and the uncertainty and dynamics of the system are further increasing. Therefore, the load transfer arrangement or path is not pre-fixed or completely determined during actual operation. Instead, the system needs to dynamically select and adjust the path of power transfer from one power supply line or device to another according to factors such as real-time demand, network status, and equipment conditions. Thus, an indeterminate load transfer scheme is formed. For the construction of an indeterminate load transfer scheme, there are mainly the following methods in the prior art: The first is to optimize the load transfer path based on the topological structure, so as to identify backup power sources and possible load transfer paths in the power grid to reduce network losses and improve the system recovery speed; the second is a multi-stage dynamic reconstruction model based on a snowflake-shaped distribution network, which constructs a flexible load transfer scheme with the goal of minimizing the load balance degree and active power loss; the third is an auxiliary decision-making method based on an expert system, which evaluates the economy and risk of the load transfer scheme by analyzing the number of switch operations, impact current, and system security. However, the construction of the above-mentioned indeterminate load transfer scheme usually relies on empirical rules or static models, which not only have high computational complexity and low efficiency, but also are not flexible and accurate enough when dealing with large-scale and dynamically changing power system problems, and are difficult to meet the requirements of the new power system. Summary of the Invention

[0004] The technical problem to be solved by the present invention is: aiming at the technical problems existing in the prior art, the present invention provides an intelligent recommendation method, device and medium for load transfer schemes based on machine learning with a simple implementation method, low cost, high intelligence level and fast response speed, which can accurately model and predict the dynamic changes of the power system, thereby improving the efficiency and accuracy of the load transfer scheme, and optimizing and enhancing the power grid resource allocation efficiency and corresponding speed.

[0005] To solve the above technical problems, the technical solution proposed by the present invention is:

[0006] An intelligent recommendation method for load transfer schemes based on machine learning, the steps include:

[0007] Step S01. Obtain the characteristic parameters of multiple different load transfer schemes, and construct a multi-dimensional feature set of the load transfer schemes. The characteristic parameters include line transfer index parameters and the change amount of the line transfer index parameters before and after transfer.

[0008] Step S02. Calculate the performance scores of the load transfer schemes to be evaluated according to the characteristic parameters of each load transfer scheme in the multi-dimensional feature set of the load transfer schemes. The load transfer schemes include individual load transfer schemes and combined schemes, and the combined schemes are formed by combining two or more individual load transfer schemes.

[0009] Step S03. Construct the feature vectors of each load transfer scheme according to the multi-dimensional feature set of the load transfer schemes, and combine every two load transfer schemes to be evaluated and compared to form a combined sample. Concatenate the feature vectors of the combined sample to form a combined feature vector, and construct a feature database from all the combined feature vectors.

[0010] Step S04. Determine the classification labels of all the features in the feature database according to the magnitude relationship of the performance scores of every two load transfer schemes, and form a label database. Set the difference greater than zero as the preferred class to train the model to distinguish the relative pros and cons between the schemes.

[0011] Step S05. Construct a training database from the feature database and the label database, and use the training database to train a classification model. After the training is completed, a transfer scheme classification model is obtained.

[0012] Step S06. Obtain multiple load transfer schemes to be recommended, input them into the trained transfer scheme classification model, and compare and re-arrange all the schemes pairwise according to the classification results of the transfer scheme classification model.

[0013] Further, the line transfer index parameters include any one or more of line current value, number of important users on the line, number of dual-power users on the line, line load rate, and main transformer load rate of the line.

[0014] Further, in step S02, if the load transfer scheme to be evaluated is a single load transfer scheme, calculating the performance score of the load transfer scheme to be evaluated includes: calculating the safety performance score using the load ratios of the transferred lines on the local side and the opposite side, the main transformer load ratios, and the main transformer load ratios of the local side line and the opposite side line; calculating the reliability index score using any one or more of the number of important users, the number of dual-power users, the historical power outage duration, the number of power outages, and the number of affected dual-power users; calculating the cost index score using the line loss rate, the main transformer load balance, and the line load balance; and comprehensively obtaining the performance score of the corresponding load transfer scheme from the safety performance score, the reliability index score, and the cost index score.

[0015] Further, in step S02, if the load transfer scheme to be evaluated is a combined scheme, calculating the performance score of the load transfer scheme to be evaluated includes:

[0016] Checking whether the load transfer scheme to be evaluated contains a sub-scheme that only involves isolation operations. If so, forming a processed combined scheme after removing the sub-scheme involving isolation operations;

[0017] Respectively calculating the safety performance score, the reliability index score, and the cost index score of each sub-scheme in the processed combined scheme, and finally determining the safety performance score, the reliability index score, and the cost index score of the load transfer scheme to be evaluated according to the scoring results of each sub-scheme.

[0018] Further, if the load transfer scheme to be evaluated is a single load transfer scheme, the calculation expression of the safety performance score is:

[0019]

[0020] Wherein, represents the safety performance score, and are the normalized scores of the load ratios of the local side line and the main transformer respectively, and are the normalized scores of the load ratios of the opposite side line and the main transformer respectively;

[0021] The calculation expression of the reliability index score is:

[0022]

[0023] Wherein, represents the reliability index score, , , , , respectively represent the normalized scores of the number of important users, the number of users with dual power supplies, the historical power outage duration, the number of power outages, and the number of affected users with dual power supplies;

[0024] The calculation expression of the cost index score is:

[0025]

[0026] where, is the cost index score, is the normalized score of the line loss rate, is the normalized score of the main transformer load balance, is the normalized score of the line load balance.

[0027] If the load transfer scheme to be evaluated is a combined scheme, the calculation expression of the safety performance score is:

[0028]

[0029]

[0030] where, represents the safety performance score of the combined scheme, is the sub - scheme S i the information entropy weight in terms of safety, represents the sub - scheme S i 's safety performance score, represents the sub - scheme S i 's information entropy, n represents the number of sub - schemes in the combined scheme;

[0031] The calculation expression of the reliability index score is:

[0032]

[0033]

[0034] where, represents the reliability index score of the combined scheme, represents the sub - scheme S i 's reliability index score, is the fuzzy adjustment coefficient used to control the balance between the maximum value and the mean value, is the sub - scheme S i 's reliability membership degree, is the minimum value of the reliability index score, is the maximum value of the reliability index score;

[0035] The calculation expression for the cost index score is as follows:

[0036]

[0037] Wherein, represents the cost index score of the combined scheme, represents the sub-scheme S i 's cost index score, and γ represents a preset coefficient.

[0038] Furthermore, in step S04, according to the magnitude relationship of the performance scores of every two load transfer schemes, the classification labels of all features in the feature database are determined, including:

[0039] Calculate the comprehensive score difference between every two load transfer schemes. The comprehensive score difference is the comprehensive score of the first load transfer scheme minus the comprehensive score of the second load transfer scheme among the two load transfer schemes. The comprehensive score is calculated based on the safety performance score, reliability index score, and cost index score;

[0040] If the score difference is greater than zero, the classification label is set to the first category for marking that the first load transfer scheme is superior to the second load transfer scheme; otherwise, the classification label is set to the second category for marking that the second load transfer scheme is superior to or equivalent to the first load transfer scheme.

[0041] Furthermore, in step S05, a random forest classification model optimized by the sea squirt swarm algorithm to construct the structure is adopted to train the load transfer scheme classification model. The training process of the load transfer scheme classification model includes the following steps:

[0042] Step S501. Initialize the random forest classification model to construct the basic framework of the load transfer scheme classification model;

[0043] Step S502. Construct a sea squirt population. Each individual in the sea squirt population corresponds to a structure initialization scheme of a random forest. The structure initialization scheme defines the training sample subset and feature subset used by each sub-decision tree during the training process to initialize the construction structure of each decision tree in the forest;

[0044] Step S503. Execute the sea squirt swarm structure optimization strategy to iteratively update the position of each individual. Among them, in each round of iteration, the position of the individual is updated according to the classification performance of the random forest model constructed by the corresponding structure scheme of the individual on the validation set;

[0045] Step S504. When the set maximum number of iterations is reached, or the change in the fitness function value is less than the convergence threshold for several consecutive rounds, terminate the iteration and output the structure initialization scheme with the optimal fitness at present;

[0046] Step S505. Apply the optimal structure initialization scheme obtained in step S504 to construct a random forest classification model, and use the complete training set to complete the model training, and finally obtain a transfer supply scheme classification model that meets the preset accuracy requirements.

[0047] Further, step S06 includes:

[0048] Step S601. Use the scheme to be recommended as the current scheme set, and randomly select a scheme from the current scheme set as the reference scheme;

[0049] Step S602. Select a scheme other than the reference scheme from the current scheme set as the current comparison scheme;

[0050] Step S603. Input the current comparison scheme and the reference scheme into the trained classification model to obtain the performance classification result between the current comparison scheme and the reference scheme. Select the next scheme other than the reference scheme from the current scheme set as the current comparison scheme, and return to step S602 until all the schemes in the current scheme set are compared with the reference scheme;

[0051] Step S604. Divide the schemes with performance better than the reference scheme in the current classification result into one group, and the remaining schemes into another group. Use the two divided groups of schemes as the current scheme set respectively, and return to step S602 until all the schemes in the scheme to be recommended are compared to obtain the sorting result of the scheme to be recommended;

[0052] Step S605. Output an intelligent recommendation result according to the sorting result of the scheme to be recommended.

[0053] Step S606. Determine whether the current recommendation result is consistent with the actual working condition. If not, construct a feedback sample according to the actual superiority and inferiority relationship between the schemes to be recommended and add it to the training sample set to update the training sample set.

[0054] An intelligent recommendation device for load transfer supply schemes based on machine learning, including a processor and a memory. The memory is used to store a computer program, and the processor is used to execute the computer program to execute the method as described above.

[0055] A computer-readable storage medium storing a computer program, and the computer program realizes the above method when executed by a processor.

[0056] Compared with the prior art, the beneficial effects of the present invention are as follows: By establishing a multi-dimensional feature set of load transfer schemes, calculating the performance scores of single schemes and combined schemes respectively based on the feature parameters of different load transfer schemes, combining each pair of load transfer schemes to be compared and evaluated to form a combined sample and performing feature vector splicing, a training database is formed to train a classifier model. At the same time, during the training process, classification labels are determined according to the performance scores of each scheme, so that the classifier model can realize the classification of the advantages and disadvantages between single or combined load transfer schemes. When intelligent recommendation of load transfer schemes is required, multiple load transfer schemes (alternative schemes) to be recommended are classified using the trained classifier model to sort each load transfer scheme, and intelligent recommendation of each load transfer scheme can be realized, greatly improving the response speed and intelligence level of the load transfer scheme, effectively realizing the optimization and improvement of the power grid resource allocation efficiency, and solving the problems of untimely response and unreasonable resource allocation in the current power grid dispatching process. Description of the Drawings

[0057] Figure 1 It is a schematic flowchart of the implementation of the intelligent recommendation method for load transfer schemes based on machine learning in this embodiment. Detailed Embodiments

[0058] The present invention will be further described below in conjunction with the accompanying drawings of the specification and specific preferred embodiments, but the protection scope of the present invention is not limited thereby.

[0059] The present invention first establishes a multi-dimensional feature set of load transfer schemes through feature engineering, and calculates the performance scores of single schemes and combined schemes respectively based on the feature parameters of different schemes. Each pair of load transfer schemes to be compared and evaluated is combined to form a training sample and feature vector splicing is performed to form a training database to train a classifier model. At the same time, during the training process, classification labels are determined according to the performance scores of each scheme, so that the classifier model can realize the classification of the advantages and disadvantages between single or combined load transfer schemes. When intelligent recommendation of load transfer schemes is required, multiple load transfer schemes (alternative schemes) to be recommended are classified using the trained classifier model to sort each load transfer scheme, and intelligent recommendation of each load transfer scheme can be realized, greatly improving the response speed and intelligence level of the load transfer scheme, effectively realizing the optimization and improvement of the power grid resource allocation efficiency, and solving the problems of untimely response and unreasonable resource allocation in the current power grid dispatching process.

[0060] As Figure 1 shown, the steps of the intelligent recommendation method for load transfer schemes based on machine learning in this embodiment include:

[0061] Step S01. Obtain the characteristic parameters of multiple different load transfer schemes, and construct a multi-dimensional feature set of the load transfer schemes. The characteristic parameters include line transfer index parameters and the change amounts of the line transfer index parameters before and after the transfer.

[0062] In this embodiment, through feature engineering on the candidate load transfer scheme data, a multi-dimensional feature set is established as the input of the model, so as to realize an accurate comprehensive scoring of the candidate schemes. The line transfer index parameters include line current value, number of important users on the line, number of dual-power users on the line, line load rate, main transformer load rate of the line, etc. By collecting the above line transfer index parameters and the change amounts before and after the transfer, the characteristics of different load transfer schemes can be comprehensively and accurately characterized.

[0063] Specifically, data such as the current value, load rate, and number of important users before and after the transfer can be obtained, and their change amounts can be calculated. For example, subtract the data such as the current value and load rate before and after the transfer, and combine them to form the change characteristics of the current and load rate before and after the transfer, and form a multi-dimensional feature set. As shown in Table 1, each feature in the multi-dimensional feature set of the constructed load transfer scheme includes the index after the transfer (expected transferred load), the line current value before and after the transfer (including the change in the line current value on this side and the change in the line current value on the opposite side), the number of important users on the line before and after the transfer (including the change in the number of important users on the line on this side and the change amount of the number of important users on the line on the opposite side), the number of dual-power users on the line before and after the transfer (including the change in the number of dual-power users on the line on this side and the change in the number of dual-power users on the line on the opposite side), the line load rate before and after the transfer (including the change in the current line load rate and the change amount of the line load rate on the opposite side), the main transformer load rate of the line before and after the transfer (including the change in the current line main transformer load rate and the change in the main transformer load rate of the line on the opposite side), and other data. The other data includes the maximum current value of the current line in the most recent week, the maximum current value of the opposite line in the most recent week, the number of distribution automation switches on the current line, the number of interconnection points on the current line, etc., which can construct a comprehensive and representative feature set.

[0064] Table 1: Feature set of load transfer scheme

[0065]

[0066] In a specific application embodiment, a feature set containing 22 can be constructed for each candidate scheme according to Table 1, and the model uses the above features to perform a comprehensive scoring on each candidate scheme, reflecting the comprehensive performance of the scheme in terms of safety, reliability, economy, and business impact.

[0067] As an alternative implementation, the latest transfer supply plan data can be obtained from a local database or a remote interface. The original data is in JSON format. To facilitate subsequent analysis and processing, the original JSON data can be converted into a data structure available within the system. The conversion process includes parsing each content in the JSON data and then mapping it to a data model predefined by the system to unify the data format and ensure the accuracy and efficiency of subsequent processing.

[0068] As an alternative implementation, when obtaining the feature data, it also includes performing preprocessing to obtain high-quality feature data, and after completing the data conversion, performing a security check on each transfer supply plan to ensure that the plan will not cause grid security problems in actual applications. For example, during the security check process, the following key parameters can be checked:

[0069] Line load rate: Evaluate whether the line load rate exceeds the safety threshold after the implementation of the transfer supply plan. An excessively high line load rate may cause line overload, increase the risk of faults, and affect the safe operation of the power grid.

[0070] Main transformer load rate: Check the load change of the main transformer after the implementation of the plan to ensure that the load rate of the main transformer remains within the safe operation range. An excessively high load rate may accelerate equipment aging and even cause equipment failures.

[0071] Equipment operation years: Consider the operation years of equipment such as lines and main transformers, and evaluate their reliability and potential aging risks. Equipment with a longer operation years may require additional maintenance or early replacement to ensure safety.

[0072] Metering point integrity: Verify the information of the metering point to confirm whether there is a situation of "missing metering point". The data accuracy and integrity of the metering point are crucial for power metering, settlement, and user charging. Missing or inaccurate metering points may lead to economic losses and data analysis deviations.

[0073] By comprehensively checking the above key parameters, transfer supply plans that may have potential safety hazards can be identified. For example, if a certain plan causes the load rate of the line or the main transformer to exceed the set safety threshold, it can be determined that the plan does not meet the safety requirements and needs to be adjusted or discarded. Similarly, if it is identified that the equipment operation years are too long and there is an obvious aging risk, a prompt message can be output to prompt that the equipment needs to be repaired or replaced before implementing the plan.

[0074] Step S02. Calculate the performance score of the load transfer supply plan to be evaluated according to the characteristic parameters of each load transfer supply plan in the characteristic set of the load transfer supply plan. The load transfer supply plan includes a single load transfer supply plan and a combined plan, and the combined plan is formed by combining two or more single load transfer supply plans.

[0075] In this embodiment, the performance scores of each load transfer scheme are calculated by using characteristic parameters to obtain model labels, thereby effectively quantifying and measuring the effectiveness of each transfer scheme.

[0076] For the performance evaluation of individual load transfer schemes, this embodiment considers three dimensions: safety, reliability, and cost. Characteristic parameters are used to calculate the safety, reliability, and cost indicators of different load transfer schemes to form a comprehensive load transfer score. As an alternative implementation, calculating the performance score of each load transfer scheme includes: calculating the safety performance score by using the post-transfer line load ratios of the local line and the opposite-side line, the main transformer load ratio, and the main transformer load ratios of the local line and the opposite-side line, that is, using the line load ratio, main transformer load ratio, etc. as safety indicators, and evaluating the safety performance of the transfer scheme by using the line load ratios and main transformer load ratios before and after the transfer of the local and opposite-side lines; using reliability indicators such as the number of important users, the number of dual-power users, the historical power outage duration, the number of power outages, and the number of affected dual-power users for scoring, that is, using the number of important users, the number of dual-power users, the historical power outage duration, the number of power outages, and the number of affected dual-power users as reliability indicators, and evaluating the reliability of the transfer scheme according to the reliability indicators; calculating the cost indicator score by using the line loss rate, main transformer load balance, and line load balance, that is, using the line loss rate, operating cost, etc. as cost indicators to evaluate the economy of the transfer scheme. Finally, the safety performance score, reliability indicator score, and cost indicator score are combined to obtain the performance score of the corresponding load transfer scheme.

[0077] Specifically, if the load transfer scheme to be evaluated is an individual load transfer scheme, in the safety indicator score, assume that the post-transfer line load ratios of the local side and the opposite side are l 1, l 2, respectively, and the main transformer load ratios of the local side and the opposite side are l 3, l 4. The increase in the local load ratio and main transformer load ratio indicates an improvement in equipment utilization, so the score should be increased accordingly; relatively, the increase in the opposite-side load ratio and main transformer load ratio may exacerbate the burden on the opposite-side system, so the score should be decreased accordingly. The safety performance score can be specifically obtained by using the following calculation expression:

[0078] (1)

[0079] Wherein, represents the safety performance score, and are the normalized scores of the local line load ratio and the main transformer load ratio respectively, indicating an improvement in equipment utilization, so they are directly included in the score, and They are the normalized scores of the opposite line load rate and the main transformer load rate respectively. Since the higher the value, the heavier the burden on the opposite system, both need to be inverted.

[0080] In the reliability index scoring, since power outage events have a negative impact on power supply reliability, the number of important users k 1. The number of dual-power users k 2. The historical power outage duration k 3. The number of power outages k 4 and the number of dual-power users affected k The increase in the values of the 5 indicators will lead to a decrease in the scoring result. The reliability index scoring can be obtained according to the following calculation expression:

[0081] (2)

[0082] Among them, represents the reliability index scoring, , , , , respectively represent the normalized scores of the number of important users, the number of dual-power users, the historical power outage duration, the number of power outages, and the number of dual-power users affected. All reliability indicators need to be inverted to ensure that the higher the value, the better the performance.

[0083] In the cost index scoring, the increase in the line loss rate means the decrease in energy efficiency. Therefore, the worse the economy of the scheme, the lower the corresponding score. The line loss rate can be used e as a cost index for economic evaluation. The calculation expression of the cost index scoring can be expressed as:

[0084] The calculation expression of the cost index scoring is:

[0085] (3)

[0086] Among them, represents the cost index scoring, is the normalized score of the line loss rate, the normalized score of the main transformer load balance, and the normalized score of the line load balance. The higher the line loss rate, the worse the economy, so it also needs to be inverted.

[0087] Furthermore, the above indicators are normalized using the maximum and minimum values, and the positive and negative indicators are processed separately to convert different indicators into a unified scoring system to ensure that all indicators are on the same evaluation scale. Among them, for the indicators that are positively correlated with the scoring, the formula for calculating the normalized score is:

[0088] (4)

[0089] Among them, S i represents the index in a certain score i of the normalized score, X i is the original value of the index corresponding to the current plan, X min and X max are respectively the minimum value and the maximum value of the index in the plan.

[0090] For negative indicators, that is, the higher the indicator value, the worse the performance of the plan. The normalized score needs to be reversed to achieve the goal that the higher the indicator value, the lower the score. The reverse score calculation formula is 1 - S i .

[0091] The score of the final power transfer plan is the sum of the scores of safety, reliability and economy, that is:

[0092] (5)

[0093] Furthermore, other adjustment measures can also be considered during the scoring process. For example, if the power transfer plan causes important users to transfer out from this side, then 100 points (configurable) will be subtracted from the scores of safety, reliability and economy respectively.

[0094] As an optional implementation method, for the performance scoring of a combined plan formed by two or more load transfer plans, that is, the load transfer plan to be evaluated is a combined plan, the following steps can be adopted:

[0095] Step S201. Check whether the load transfer plan to be evaluated contains a sub-plan that only involves isolation operations. If it exists, then form a processed combined plan after removing the sub-plan involving isolation operations;

[0096] Step S202. If it is determined that there is a plan that is only an isolation operation in the load transfer plan to be evaluated, then form a processed combined plan after deleting the sub-plan involving isolation operations;

[0097] Step S203. Calculate the safety performance score, reliability index score and the cost index score of each sub-plan in the processed combined plan respectively, and finally determine the safety performance score, reliability index score and cost index score of the load transfer plan to be evaluated according to the scoring results of each sub-plan;

[0098] A combined solution usually consists of multiple sub-solutions, and some of the sub-solutions may only involve isolation operations (for example, the switch is only disconnected without a closing operation). In the process of evaluating the performance of the combined solution in this embodiment, by excluding the sub-solutions that only involve isolation operations, it is possible to avoid the interference of the sub-solutions that only involve isolation operations on the overall score. For the evaluation of the combined solution, first screen the sub-solutions that only involve isolation operations. After excluding the isolation operation sub-solutions, integrate the indicators of the remaining sub-solutions to construct the scoring object of the combined solution. By comparing the scoring indicators of all sub-solutions, the final score of the combined solution is determined. For example, the maximum impact principle, entropy weight dynamic adjustment, and non-linear normalization are used to calculate the final score of the combined solution.

[0099] Specifically, after excluding the isolation operation sub-solutions, according to the safety performance score, reliability index score, and cost index of each load transfer scheme in the combined solution, the safety performance score of the combined solution can be determined by using the maximum impact principle, the reliability index score can be determined by entropy weight dynamic adjustment, and the cost index score can be determined by non-linear normalization; finally, the performance score of the combined solution is obtained by comprehensively determining the safety performance score, reliability index score, and cost index score of the combined solution.

[0100] As an alternative implementation, the calculation expressions for the corresponding index scores of the combined solution are respectively:

[0101] The calculation expression for the safety performance score can be:

[0102] (6)

[0103] (7)

[0104] Among them, represents the safety performance score of the combined solution, is the information entropy weight of sub-solution S i in terms of safety, represents the safety performance score of sub-solution S i , represents the information entropy of sub-solution S i , n represents the number of sub-solutions in the combined solution.

[0105] The calculation expression for the reliability index score is:

[0106] (8)

[0107] (9)

[0108] Among them, represents the reliability index score of the combined solution, represents the sub-solution S i 's reliability index score, is the fuzzy adjustment coefficient used to control the balance between the maximum value and the mean value, is the sub-solution S i 's reliability membership degree, is the minimum value of the reliability index score, is the maximum value of the reliability index score.

[0109] The calculation expression of the cost index score is:

[0110] (10)

[0111] where, represents the cost index score of the combined solution, represents the cost index score of the sub-solution S i γ represents a preset coefficient. If γ is greater than 1, it means more attention is paid to economic balance, and vice versa, more attention is paid to the lowest economic solution.

[0112] Finally, add the S 安全 , S 可靠 , S 经济 of the combined solution to get the total score.

[0113] For example, consider a combined solution consisting of three sub-solutions A, B, and C. Among them, sub-solutions A and B are complete transfer operations, while sub-solution C only involves isolation operations (no load transfer). During the evaluation process of the combined solution, first screen all sub-solutions and eliminate sub-solution C that only involves isolation operations to avoid its interference with the overall score. Subsequently, integrate the scores of the remaining sub-solutions A and B. For each scoring index of the combined solution, use the method of combining the maximum value principle and the information entropy weighted mean to calculate, ensuring that the final score of the combined solution can not only reflect the contribution of the best sub-solution but also comprehensively consider the overall impact of all effective sub-solutions. For example, if the safety performance score of sub-solution A is 0.7 and the safety performance score of sub-solution B is 0.8, then the safety performance score of the combined solution is calculated as follows:

[0114] (11)

[0115] where, P A , P Bis the weight calculated according to the information entropy, which is used to balance the contributions of different sub - schemes in terms of security and ensure the rationality of scoring. Similarly, the reliability index score and economic score of the combined scheme are calculated, and finally the security, reliability, and economic scores of the combined scheme are added together to obtain the total score.

[0116] Step S03. Construct the feature vectors of each load transfer scheme according to the multi - dimensional feature set of the load transfer scheme, combine every two load transfer schemes to form a combined sample, splice the feature vectors of the combined sample to form a combined feature vector, and construct a training database from all the combined feature vectors.

[0117] In this embodiment, by combining every two load transfer schemes to form a combined sample, and then splicing the feature vectors of the combined sample to form a combined feature vector, combined samples for model training are generated to further evaluate and compare the advantages and disadvantages of different power transfer schemes. Using the pairwise combination method to generate training samples can enhance the model's learning ability of the relationship between the advantages and disadvantages of the schemes.

[0118] Specifically, assume that the feature set of the transfer scheme obtained from step S1 is B ={ b 1, b 2,…, b n}, n where is the number of transfer schemes; to meet the requirements of model training, each transfer scheme feature b i ={ z 1, z 2,…, z 22} is reshaped into a feature vector with a length of 22; during the model construction process, to measure the advantages and disadvantages between schemes, a sample set b i , b j is generated by pairwise combination, where i and j represent scheme 1 and scheme 2 respectively. Through this permutation and combination method, all possible pairs of schemes can be generated. When processing each pair of schemes (denoted as scheme 1 and scheme 2), a new combined feature vector is constructed, which is formed by splicing the feature vectors of scheme 1 and scheme 2. Assume that the feature vector of scheme 1 is b i ={ z i1 , z i2 ,…, z i22}, and the feature vector of scheme 2 is Bj ={ z j1 , z j2 ,…, z j22}, by concatenating these two feature vectors, the final combined feature vector b ij will have a length of 44. Through this construction method, the differences between the solutions can be effectively captured, providing rich feature information for the model.

[0119] Step S04. Determine the classification labels of all features in the feature database according to the magnitude relationship of the performance scores of every two load transfer solutions, forming a label database.

[0120] In this embodiment, during the process of constructing the training database, the score difference between the two load transfer solutions in the combined solution is used as the classification label, that is, the score difference is converted into a classification label, enabling the model to effectively learn the relationship between the features and the advantages and disadvantages of the solutions. After training, the model can directly distinguish the performance advantages and disadvantages of different solutions.

[0121] As an alternative implementation, determining the classification labels of all features in the feature database according to the magnitude relationship of the performance scores of every two load transfer solutions can be carried out in the following steps:

[0122] Step S401. Calculate the comprehensive score difference between every two load transfer solutions, where the comprehensive score difference is the comprehensive score of the first load transfer solution minus the comprehensive score of the second load transfer solution among the two load transfer solutions, and the comprehensive score is calculated based on the safety performance score, reliability index score, and cost index score;

[0123] Step S402. If the comprehensive score difference is greater than zero, the classification label is set to the first category, which is used to mark that the first load transfer solution is superior to the second load transfer solution; otherwise, the classification label is set to the second category, which is used to mark that the second load transfer solution is superior to or equivalent to the first load transfer solution.

[0124] Specifically, during the training process, when the score difference is converted into a classification label, if the difference is greater than zero, it is set to the priority class to train the model to distinguish the relative advantages and disadvantages between the solutions. For example, if the comprehensive score of solution 1 minus the comprehensive score of solution 2 is greater than 0, it is marked as category 1 (indicating that solution 1 is superior to solution 2); otherwise, it is marked as category 0 (indicating that solution 2 is superior to solution 1). Subsequently, the classifier is trained with the goal of maximizing the information gain or classification accuracy. After training, the model can directly distinguish the advantages and disadvantages of the solutions.

[0125] Step S05. Construct a training database from the feature database and the label database, and use the training database to train a classification model. After the training is completed, a load transfer scheme classification model is obtained.

[0126] Preferably, in this embodiment, a random forest (RF) classification model based on the ascidian swarm optimization algorithm is adopted to train and form a load transfer scheme classification model, so as to realize the efficient optimization of the load transfer scheme classification task. This model uses the swarm intelligence optimization strategy of the ascidian swarm optimization algorithm to dynamically adjust the hyperparameters of the random forest model, thereby significantly improving the classification performance and accuracy of the model.

[0127] As an alternative implementation, the steps of training the random forest classification model based on the ascidian swarm optimization algorithm include:

[0128] Step S501. Initialize the random forest classification model, including setting basic parameters such as the number of decision trees n estimator , the maximum depth max_depth, and the minimum number of samples min_samples_split at the leaf nodes; initialize the ascidian population, and set the population size, the maximum number of iterations, and the fitness function.

[0129] Specifically, the default parameters can be set as follows: the number of decision trees (n_estimators) = 100, the maximum depth (max_depth) = None, and the minimum number of samples at the leaf nodes (min_samples_split) = 2. Initialize the ascidian population and set the population size (such as 30 ascidians) and the maximum number of iterations (such as 100 times). At this stage, the random forest (RF) algorithm is used as the basic model, and parameters such as the initial number of decision trees, the maximum depth, and the minimum number of samples at the nodes are set to construct the initial structural framework of the model.

[0130] Step S502. Construct an ascidian population. Each individual in the ascidian population corresponds to a structural initialization scheme of a random forest. In this structural initialization scheme, the training sample subset and the feature subset used by each sub-decision tree during the training process are defined, that is, by clarifying the "training sub-samples" and "feature screening" methods, the construction structure of each decision tree in the forest is initialized.

[0131] In this embodiment, a new "structural initialization scheme" is introduced on the basis of step S501. The core idea is to encode the sample and feature selection process of each tree in the random forest as a vector, so as to have a clear and controllable scheme in the optimization stage, so that the construction process of the entire random forest model is no longer completely random, but can be optimized purposefully with the help of swarm intelligence algorithms, and finally improve the classification performance of the model.

[0132] Specifically, represent the structure initialization scheme of each random forest as a vector X i , which is composed of several sub-vectors spliced together, where each sub-vector corresponds to the construction details of a tree in the forest: use D (j) to represent the j training samples selected for the (j) th tree. In the traditional random forest, each tree obtains a part of the samples through Bootstrap sampling. In this embodiment, the sampling result is represented as a vector D , where each element represents the index of a selected sample. The form of the feature mask vector is d , that is, it represents the feature mask of the tree. The length of this vector is d , representing the total number of features. In the vector, when the value at each position is 1, it means that the corresponding feature is used in the construction process of the tree, and when it is 0, it means that it is not used. Through the above method, the feature subset considered in the construction of each tree can be controlled, so that the tree construction process is more targeted. The form of the whole vector is:

[0133]

[0134] where is the number of trees in the forest, represents vector splicing.

[0135] Initially, generate a group of the above structure initialization schemes randomly to form a sea squirt population. In this population, each X i individual represents a candidate forest construction strategy. By explicitly encoding the construction process of each tree in the random forest, these structure initialization schemes can be searched during the optimization phase. Use the sea squirt swarm algorithm to traverse different X i values, so as to find the random forest construction strategy that can achieve the optimal classification performance on the validation set. Using this method can change the traditional random construction method, making the model construction process change from "blind randomness" to "purposeful search", thereby improving the generalization ability and classification accuracy of the entire random forest.

[0136] Step S503. Execute the sea squirt swarm structure optimization strategy, iteratively update the position of each individual, where in each iteration, update the individual position according to the classification performance of the random forest model constructed by the individual's corresponding structure scheme on the validation set.

[0137] Specifically, in each iteration, according to the foraging and cooperation behavior mechanism of the ascidian swarm algorithm, a corresponding random forest model is constructed based on the structural scheme described by each individual, and its classification performance is tested on the validation set. Then, the individual positions are updated based on the classification performance of the random forest model constructed according to the corresponding structural scheme of the individual. After multiple iterations, a better structural initialization scheme is continuously evolved. To ensure that the model can effectively distinguish between positive and negative classes during the training process, the cross-entropy loss function can be used to measure the deviation between the predicted probability and the actual label. Its mathematical expression is:

[0138] (12)

[0139] where N is the number of samples, y i is the true label (0 or 1), p i is the predicted probability of the model for sample i . By minimizing this loss function, the model can effectively adjust its parameters to improve the classification accuracy of the samples.

[0140] In this embodiment, a multi-index weighted fitness function can be used to reflect the overall performance of the model on the validation set. For example:

[0141] (13)

[0142] where ω1, ω2, and λ are preset weights, represents the accuracy index, represents the F1 score calculated using precision and recall, and Overfitting Penalty represents the overfitting penalty index.

[0143] Based on the above fitness evaluation, using the core update mechanism of the ascidian swarm algorithm, each structural initialization scheme is continuously improved through global search. Specifically, the update formula for updating the individual positions by selecting the three individuals X α 、X β and X δ as guidance is set as:

[0144] (14)

[0145]

[0146] where represents the position of the individual in the i-th iteration, represents the updated position of the individual after the i-th iteration,r 1 and r 2 are uniformly distributed random numbers, is a control factor that decreases with the number of iterations.

[0147] Through the above update mechanism, each ascidian individual continuously moves towards the global optimum direction in the population, so that the corresponding structure initialization scheme gradually evolves to be more suitable for the current data set, thereby improving the classification accuracy of the entire forest model on the validation set.

[0148] Step S504. When the set maximum number of iterations is reached, or when the change in the fitness function value is less than the convergence threshold for several consecutive rounds, terminate the iteration and output the structure initialization scheme with the best current fitness.

[0149] Specifically, when the number of iterations reaches the preset maximum value or the change in the optimal fitness is lower than the threshold for several consecutive rounds the iteration terminates. At this time, select the individual with the highest fitness The corresponding structure initialization scheme is used as the final optimization result.

[0150] Step S505. Apply the optimal structure initialization scheme obtained in step S504 to construct a random forest classification model, and use the complete training set to complete the model training, and finally obtain a classifier that meets the accuracy requirements.

[0151] Specifically, use the optimal structure scheme in step S504 S ∗ Construct a complete random forest model and perform final training on the model over the entire training set until the classification accuracy meets the preset requirements. Through this series of structure optimization and training processes, the obtained random forest classifier can not only overcome the uncertainty brought by random initialization in the initial construction stage, but also show better robustness and classification performance in various complex and high-dimensional feature data environments.

[0152] To verify the performance of the model, this embodiment further selects a test set to compare the advantages and disadvantages of the schemes. Assume that the test scheme set is , including the feature vectors of all test schemes, where g is a test scheme, and n t represents the total number of test schemes; randomly select two schemes from the test set G ( g i , g j ), where i ≠ j, the feature vectors of the two selected test schemes are concatenated to form an input vector with a dimension of (1, 44) and passed to the trained random forest model; the output of the model is the judgment of the superiority and inferiority between the two schemes, with a dimension of (1, 1), and the superiority and inferiority relationship between the two test schemes is evaluated by the positive and negative magnitudes of the output difference. g i , g j superiority and inferiority relationship.

[0153] In this embodiment, the diversity and richness of the combined samples can provide a sufficient learning basis for the model, enabling it to better capture the subtle differences between different schemes. Furthermore, by generating a large number of combined samples, the generalization ability of the model can be greatly enhanced, thereby effectively improving the accuracy of the evaluation of new schemes. At the same time, using the random forest model for training can not only systematically evaluate and compare different power transfer schemes, but also provide more scientific and accurate support for the decision-making process, which helps to generate the required power transfer schemes more efficiently and accurately in the planning and operation of the power system.

[0154] Step S06. Obtain multiple load transfer schemes to be recommended, input them into the trained classification model, and re-arrange and compare all the schemes pairwise based on the superiority and inferiority comparison results (classification results of the load transfer scheme classification model) output by the model, so as to generate the recommended results of the load transfer schemes.

[0155] In this embodiment, multiple load transfer schemes to be recommended are input into the trained classification model. The classification model can respectively distinguish the superiority and inferiority between each scheme, sort each load transfer scheme according to the judgment results, and finally obtain the sorted list of all the schemes to be used to determine the optimal load transfer scheme, thereby assisting the final scheme selection and decision-making. The scheme with the highest score can be preferentially selected for implementation to optimize the operation efficiency and reliability of the power grid. The recommended results determined in the above manner can not only ensure that the selected scheme is the best choice under the current circumstances, but also improve the transparency and interpretability of the decision-making process, thereby improving the operation efficiency and service quality of the power grid.

[0156] As an optional implementation method, according to the scoring results, the quicksort method can be used to sort each load transfer scheme. The steps include:

[0157] Step S601. Take the schemes to be recommended as the current scheme set, and randomly select a scheme from the current scheme set as the reference scheme;

[0158] Step S602. Select a scheme other than the reference scheme from the current scheme set as the current comparison scheme;

[0159] Step S603. Input the current comparison scheme and the benchmark scheme into the trained classification model to obtain the performance classification result between the current comparison scheme and the benchmark scheme. Select the next scheme other than the benchmark scheme from the current scheme set as the current comparison scheme, and return to Step S502 until all the schemes in the current scheme set are compared with the benchmark scheme;

[0160] Step S604. Divide the schemes with performance superior to the benchmark scheme in the current classification result into one group, and the remaining schemes into another group. Take the two divided groups of schemes as the current scheme set respectively, and return to Step S502 until all the schemes in the schemes to be recommended are compared to obtain the sorting result of the schemes to be recommended;

[0161] Step S605. Output the intelligent recommendation result according to the sorting result of the schemes to be recommended;

[0162] Step S606. Determine whether the current recommendation result is consistent with the actual situation. If not, construct a feedback sample according to the actual superiority and inferiority relationship between the schemes to be recommended and add it to the training sample set to update the training sample set.

[0163] Specifically, when there is a difference between the actual scheduling result and the model recommendation result, the system will construct a feedback sample according to the input order of the scheme pair and the actual superiority and inferiority relationship, and append it to the training sample library to participate in the subsequent model update. This feedback update process will continue until the model output is consistent with the actual scheduling result. Taking Scheme A and Scheme B as an example, if the model classifies the concatenated input [A, B] as label 1 (i.e., Scheme A is superior to B), but it is found in actual operation that B is better, at this time, the constructed sample input is [A, B], and the label is 0. This sample pair will be used as a group of feedback samples and appended to the training sample library to participate in the subsequent model update. By continuously iteratively adding such feedback samples, the model will gradually learn to conform to the actual situation and output a judgment result that meets the actual operation requirements in subsequent classifications. Finally, through continuous sample appending and incremental training, the model will gradually correct the original sorting error until when inputting the same concatenated feature [A, B], it can accurately output a label of 0, correctly judge that Scheme B is superior to Scheme A, so as to realize the dynamic correction and continuous optimization of the recommendation logic.

[0164] In summary, in this embodiment, the data of historical load transfer schemes (including single and combined schemes) is first subjected to data cleaning, data verification, and data conversion. Through feature engineering, a typical feature set of load transfer is selected, including 22 types of index features before and after transfer. Safety, reliability, cost, and other scoring indicators are formed for different historical transfer schemes as data labels, and a comprehensive score label is formed by adding the three scores. Among them, the combined scheme is formed by combining multiple single schemes. After excluding the isolation scheme, the maximum value of the scoring indicators of all single schemes is selected to calculate the combined score. Subsequently, based on the existing load transfer scheme dataset, a random forest model is constructed and trained with the goal of maximizing information gain. The input of the model is the concatenation of all features of paired schemes, totaling 44 features, and the output of the model is the comparison result of the advantages and disadvantages of the two schemes. Finally, for multiple load transfer schemes to be recommended, the schemes are combined in pairs and input into the trained random forest model to evaluate the pros and cons relationship. Finally, all candidate schemes are re-ranked to select the optimal scheme, realizing the intelligent recommendation of suitable transfer schemes. Through the above steps, the rapid evaluation and sorting of a large number of candidate schemes can be achieved, meeting the requirements of power grid real-time scheduling. The evaluation and comparison of multiple feasible load transfer scheme sets in different scenarios can be carried out, and the machine learning algorithm is used to optimize and rank the feasible schemes to form different strategy recommendation sequences. Compared with the traditional weighted scoring algorithm, the decision-making process can be made more objective and reliable, reducing the interference of human factors. Thus, by reducing power outage events and improving power supply stability, in summary, compared with the traditional formulation of load transfer schemes based on expert rules and mechanism models, the present invention can quickly and accurately formulate load transfer schemes, reduce the human error rate, and at the same time has stronger scalability and can better adapt to the increasingly complex load changes of the power system.

[0165] This embodiment further provides an intelligent recommendation device for load transfer schemes based on machine learning, including a processor and a memory. The memory is used to store a computer program, and the processor is used to execute the computer program to execute the method as described above.

[0166] It can be understood that the above method of this embodiment can be executed by a single device, such as a computer or a server, etc., or can also be applied to a distributed scenario where multiple devices cooperate with each other to complete. In the case of a distributed scenario, one of the multiple devices can only execute one or more of the above steps of this embodiment, and the multiple devices interact with each other to complete the above method. The processor can be implemented in the form of a general-purpose CPU, a microprocessor, an application-specific integrated circuit, or one or more integrated circuits, etc., and is used to execute relevant programs to implement the above method of this embodiment. The memory can be implemented in the form of a read-only memory ROM, a random access memory RAM, a static storage device, and a dynamic storage device, etc. The memory can store an operating system and other application programs. When implementing the above method of this embodiment through software or firmware, the relevant program codes are stored in the memory and are called and executed by the processor.

[0167] This embodiment further provides a computer-readable storage medium storing a computer program, and when the computer program is executed by a processor, the above method is implemented.

[0168] Those skilled in the art should understand that the above embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can be in the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can be in the form of a computer program product implemented on one or more computer-readable storage media (including but not limited to disk memories, CD-ROMs, optical memories, etc.) containing computer-usable program codes. The present invention is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, and the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for implementing the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks. These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured article including an instruction device, and the instruction device implements the functions in Figure 1 one process or multiple processes and / or blocks Figure 1The functions specified in one or more boxes. These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process. Thus, the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one or more processes and / or boxes. Figure 1 One process or more processes and / or boxes Figure 1 The steps of the functions specified in one box or more boxes.

[0169] The above are only the preferred embodiments of the present invention and do not impose any form of limitation on the present invention. Although the present invention has been disclosed above with the preferred embodiments, it is not intended to limit the present invention. Therefore, any simple modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention shall fall within the scope of the protection of the technical solution of the present invention.

Claims

1. An intelligent recommendation method for load transfer schemes based on machine learning, characterized by the steps Including: Step S01. Obtain the characteristic parameters of multiple different load transfer schemes, and construct a multi-dimensional feature set of the load transfer schemes. The characteristic parameters include line transfer index parameters and the change amounts of the line transfer index parameters before and after transfer. Step S02. Calculate the performance scores of the load transfer schemes to be evaluated according to the characteristic parameters of each load transfer scheme in the multi-dimensional feature set of the load transfer schemes. The load transfer schemes include individual load transfer schemes and combined schemes. The combined scheme is formed by combining two or more individual load transfer schemes. Step S03. Construct the feature vectors of each load transfer scheme according to the multi-dimensional feature set of the load transfer schemes, and combine every two load transfer schemes to be evaluated and compared to form a combined sample. Concatenate the feature vectors of the combined sample to form a combined feature vector, and form a feature database from all the combined feature vectors. Step S04. Determine the classification labels of all the features in the feature database according to the magnitude relationship of the performance scores of every two load transfer schemes, and form a label database. Step S05. Construct a training database from the feature database and the label database, and use the training database to train a classification model. After the training is completed, a transfer scheme classification model is obtained. Step S06. Obtain multiple load transfer schemes to be recommended, input them into the trained transfer scheme classification model, and compare and rearrange all the schemes pairwise according to the classification results of the transfer scheme classification model to obtain the load transfer scheme recommendation result.

2. The intelligent recommendation method for load transfer scheme based on machine learning according to claim 1, wherein The line transfer index parameters include any one or more of line current value, number of important users on the line, number of dual-power users on the line, line load rate, and main transformer load rate of the line.

3. The intelligent recommendation method for load transfer scheme based on machine learning according to claim 1, characterized in that In step S02, if the load transfer scheme to be evaluated is an individual load transfer scheme, calculating the performance score of the load transfer scheme to be evaluated includes: calculating the safety performance score using the post-transfer line load rate and main transformer load rate of the local line and the opposite line, and the main transformer load rate of the local line and the opposite line; calculating the reliability index score using any one or more of the number of important users, number of dual-power users, historical power outage duration, number of power outages, and number of affected dual-power users; calculating the cost index score using line loss rate, main transformer load balance, and line load balance; and comprehensively obtaining the performance score of the corresponding load transfer scheme by combining the safety performance score, reliability index score, and cost index score.

4. The intelligent recommendation method for load transfer scheme based on machine learning according to claim 3, characterized in that, In step S02, if the load transfer scheme to be evaluated is a combined scheme, calculating the performance score of the load transfer scheme to be evaluated includes: Checking whether the load transfer scheme to be evaluated contains a sub-scheme that only involves isolation operations. If it exists, then form a processed combined scheme after removing the sub-scheme involving isolation operations. Respectively calculate the safety performance score, reliability index score, and cost index score of each sub-scheme in the processed combined scheme, and finally determine the safety performance score, reliability index score, and cost index score of the load transfer scheme to be evaluated according to the scoring results of each sub-scheme.

5. The intelligent recommendation method for load transfer scheme based on machine learning according to claim 4, characterized in that, If the load transfer scheme to be evaluated is a single load transfer scheme, the calculation expression of the safety performance score is as follows: Among them, represents the safety performance score, and are the normalized scores of the line load rate and main transformer load rate on this side respectively, and are the normalized scores of the line load rate and main transformer load rate on the opposite side respectively; The calculation expression of the reliability index score is as follows: Among them, represents the reliability index score, , , , , respectively represent the normalized scores of the number of important users, the number of dual-power users, the historical power outage duration, the number of power outages, and the number of affected dual-power users; The calculation expression of the cost index score is as follows: Among them, is the cost index score, is the normalized score of line loss rate, is the normalized score of main transformer load balance, is the normalized score of line load balance; If the load transfer scheme to be evaluated is a combined scheme, the calculation expression of the safety performance score is as follows: Among them, represents the safety performance score of the combination scheme, is the sub-scheme S i the information entropy weight of [[sub-scheme 1]] in terms of security, represents the sub-scheme S i 's safety performance score, represents the sub-scheme S i 's information entropy, n represents the number of sub-schemes in the combination scheme; The calculation expression of the reliability index score is as follows: Among them, represents the reliability index score of the combination scheme, represents the sub-scheme S i 's reliability index score, is the fuzzy adjustment coefficient used to control the balance between the maximum value and the mean value, is the sub-scheme S i 's reliability membership degree, is the minimum value of the reliability index score, is the maximum value of the reliability index score; The calculation expression of the cost index score is as follows: Among them, represents the cost index score of the combination plan, represents the sub-plan S i 's cost index score, and γ represents a preset coefficient.

6. The intelligent recommendation method for load transfer scheme based on machine learning according to any one of claims 1 to 5, characterized in that In step S04, according to the magnitude relationship of the performance scores of every two load transfer schemes, the classification labels of all features in the feature database are determined, including: Calculate the comprehensive score difference between every two load transfer schemes. The comprehensive score difference is the comprehensive score of the first load transfer scheme in the two load transfer schemes minus the comprehensive score of the second load transfer scheme. The comprehensive score is calculated based on the safety performance score, reliability index score, and cost index score; If the comprehensive score difference is greater than zero, the classification label is set to the first category, which is used to mark that the first load transfer scheme is superior to the second load transfer scheme; otherwise, the classification label is set to the second category, which is used to mark that the second load transfer scheme is superior to or equivalent to the first load transfer scheme.

7. The intelligent recommendation method for load transfer scheme based on machine learning according to any one of claims 1 to 5, characterized in that, In step S05, a random forest classification model optimized by a sea squirt swarm algorithm is used to construct the structure, and the load transfer scheme classification model is trained. The training process of the load transfer scheme classification model includes the following steps: Step S501. Initialize the random forest classification model to construct the basic framework of the load transfer scheme classification model; Step S502. Construct a sea squirt population. Each individual in the sea squirt population corresponds to a structure initialization scheme of a random forest. The structure initialization scheme defines the training sample subset and feature subset used by each sub-decision tree during training to initialize the construction structure of each decision tree in the forest; Step S503. Execute the sea squirt swarm structure optimization strategy to iteratively update the position of each individual. In each iteration, the position of the individual is updated according to the classification performance of the random forest model constructed by the corresponding structure scheme of the individual on the validation set; Step S504. When the set maximum number of iterations is reached, or the change in the fitness function value is less than the convergence threshold for several consecutive rounds, terminate the iteration and output the currently fitness-optimal structure initialization scheme; Step S505. Apply the optimal structure initialization scheme obtained in step S504 to construct a random forest classification model, and use the complete training set to complete the model training, and finally obtain a load transfer scheme classification model that meets the preset accuracy requirements.

8. The intelligent recommendation method for load transfer scheme based on machine learning according to any one of claims 1 to 5, characterized in that, Step S06 includes: Step S601. Use the scheme to be recommended as the current scheme set, and randomly select a scheme from the current scheme set as the reference scheme; Step S602. Select a scheme other than the reference scheme from the current scheme set as the current comparison scheme; Step S603. Input the current comparison scheme and the baseline scheme into the trained classification model to obtain the performance classification result between the current comparison scheme and the baseline scheme. Select the next scheme other than the baseline scheme from the current scheme set as the current comparison scheme, and return to step S602 until all schemes in the current scheme set are compared with the baseline scheme; Step S604. Divide the schemes with performance superior to the baseline scheme in the current classification result into one group, and the remaining schemes into another group. Use the two divided groups of schemes as the current scheme set respectively, and return to step S502 until all schemes in the schemes to be recommended are compared to obtain the sorting result of the schemes to be recommended; Step S605. Output the intelligent recommendation result according to the sorting result of the schemes to be recommended; Step S606. Determine whether the current recommendation result is consistent with the actual working condition. If not, construct a feedback sample according to the actual superiority and inferiority relationship between the schemes to be recommended and add it to the training sample set to update the training sample set.

9. An intelligent recommendation device for load transfer schemes based on machine learning, comprising a processor and a memory, wherein the memory is used for storing computer programs, characterized in that, The processor is configured to execute the computer program to perform the method according to any one of claims 1 to 8.

10. A computer-readable storage medium storing a computer program, characterized in that, The computer program, when executed by the processor, implements the method according to any one of claims 1 to 8.

Images