Fuzzy boundary inference method and device for revenue rule information
By obtaining and encoding reference parameters and data of income rule information, and using evolutionary algorithms to infer the boundaries of income rule information, the problem of inaccurate income calculation in the prior art is solved, and accurate calculation and boundary inference of non-ideal working conditions are realized. Improves the completeness and accuracy of inference.
Patent Information
- Application Number
- CN202411056240.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-02
- Publication Date
- 2024-11-08
AI Technical Summary
In complex actual operating environments, existing income rule information cannot accurately calculate and evaluate returns under non-ideal operating conditions, making it difficult to perform targeted performance adjustments and optimizations.
By obtaining income rule information, reference income parameters and data at historical moments, scene identification and rule assumptions are performed, gene sequences are encoded and calculations are used to infer the boundaries of income rule information to improve the completeness, universality and accuracy of the boundaries sex.
It realizes accurate calculation and automated solution of returns, adapts to the fine-tuning requirements of rules in different regions, and data-driven counter-solving and inferring fuzzy and uncertain market rule boundaries, improving the completeness and accuracy of inference boundaries.
Smart Images

Figure CN118917419A_ABST
Abstract
Description
Technical Field
[0001] The present disclosure relates to the technical field of data processing, and in particular to a fuzzy boundary inference method and device for revenue rule information. Background Art
[0002] In the relevant technology, the revenue rule information stipulates the revenue calculation method and assessment standards under the ideal operating state. However, the actual operating environment of the grid-connected entity is extremely complex, and non-ideal operating conditions often occur. Taking the secondary frequency regulation in the power market scenario as an example, the revenue rule information stipulates the performance and revenue calculation method of the power generation unit (the grid-connected entity on the power generation side, such as thermal power plants, hydropower plants, and new energy storage, such as flywheel energy storage, etc.) when it responds normally. However, when the power generation unit receives a new frequency regulation instruction but does not respond normally, the secondary frequency regulation revenue cannot be calculated. These rule boundaries that are not disclosed in the revenue rule information may make it difficult for the subject object to evaluate the actual benefits and assessments, and even more difficult to make targeted performance adjustments and optimizations.
[0003] Therefore, how to construct the boundaries of profit rule information, improve the completeness, universality and accuracy of the inferred boundaries, automatically solve the problem of precise calculation of profits, adapt to the needs of fine-tuning rules in different regions, and data-drivenly reverse and infer the boundaries of fuzzy and uncertain market rules has become one of the important research directions. Summary of the invention
[0004] The present disclosure aims to solve one of the technical problems in the related art at least to a certain extent. To this end, one purpose of the present disclosure is to propose a fuzzy boundary inference method for revenue rule information.
[0005] The second objective of the present disclosure is to propose a fuzzy boundary inference device for profit rule information.
[0006] A third objective of the present disclosure is to provide an electronic device.
[0007] A fourth object of the present disclosure is to provide a non-transitory computer-readable storage medium.
[0008] A fifth object of the present disclosure is to provide a computer program product.
[0009] To achieve the above-mentioned purpose, the first embodiment of the present disclosure proposes a fuzzy boundary inference method for revenue rule information, including:
[0010] Obtain the profit rule information, the reference profit parameters at historical moments, and the reference profit data corresponding to the reference profit parameters;
[0011] Based on the revenue rule information, reference revenue parameters and reference revenue data, scenario identification and rule assumptions are performed to obtain the rule assumption set and rule boundary parameter set under the specified target scenario;
[0012] Perform rule encoding on the rule hypothesis set and the rule boundary parameter set to obtain the gene sequence;
[0013] Based on the preset fitness function, the evolutionary algorithm is used to calculate and decode the gene sequence to obtain the boundary of the profit rule information.
[0014] To achieve the above-mentioned purpose, the second aspect of the present disclosure proposes a fuzzy boundary inference device for profit rule information, comprising:
[0015] The first acquisition module is used to obtain the income rule information, the reference income parameters at the historical moment, and the reference income data corresponding to the reference income parameters;
[0016] The second acquisition module is used to perform scenario recognition and rule assumptions based on the revenue rule information, reference revenue parameters and reference revenue data, and acquire a rule assumption set and a rule boundary parameter set under a specified target scenario;
[0017] The third acquisition module is used to perform rule encoding on the rule hypothesis set and the rule boundary parameter set to obtain the gene sequence;
[0018] The boundary inference module is used to calculate and decode the gene sequence using an evolutionary algorithm based on a preset fitness function to obtain the boundary of the profit rule information.
[0019] To achieve the above-mentioned purpose, a third aspect of the present disclosure provides an electronic device, including:
[0020] at least one processor; and
[0021] a memory communicatively connected to at least one processor; wherein,
[0022] The memory stores instructions that can be executed by at least one processor, and the instructions are executed by at least one processor so that the at least one processor can execute the fuzzy boundary inference method of profit rule information provided in the embodiment of the first aspect of the present disclosure.
[0023] To achieve the above-mentioned purpose, the fourth aspect embodiment of the present disclosure proposes a computer-readable storage medium having computer instructions stored thereon, wherein the computer instructions are used to enable a computer to execute a fuzzy boundary inference method of the profit rule information provided in the first aspect embodiment of the present disclosure.
[0024] To achieve the above-mentioned purpose, the fifth aspect of the present disclosure proposes a computer program product, including a computer program, which, when executed by a processor, implements the fuzzy boundary inference method of profit rule information provided in the first aspect of the present disclosure.
[0025] The disclosed embodiments can construct the boundaries of revenue rule information, improve the completeness, universality and accuracy of the inferred boundaries, automatically solve the problem of accurate calculation of revenue, adapt to the needs of fine-tuning rules in different regions, and perform data-driven inversion and inference of fuzzy and uncertain market rule boundaries. BRIEF DESCRIPTION OF THE DRAWINGS
[0026] Figure 1 is a flow chart of a fuzzy boundary inference method of revenue rule information according to an embodiment of the present disclosure;
[0027] Figure 2 is a flow chart of a fuzzy boundary inference method of revenue rule information according to an embodiment of the present disclosure;
[0028] Figure 3 is a flow chart of a fuzzy boundary inference method of revenue rule information according to an embodiment of the present disclosure;
[0029] Figure 4 is a schematic diagram of evolutionary computing according to an embodiment of the present disclosure;
[0030] Figure 5 It is a structural block diagram of a fuzzy boundary inference device for revenue rule information according to an embodiment of the present disclosure;
[0031] Figure 6 It is a schematic diagram of the structure of an electronic device according to an embodiment of the present disclosure. DETAILED DESCRIPTION
[0032] Embodiments of the present disclosure are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present disclosure, and should not be construed as limiting the present disclosure.
[0033] The following describes the fuzzy boundary inference method and device of the revenue rule information according to the embodiment of the present disclosure in conjunction with the accompanying drawings.
[0034] Figure 1 is a flowchart of a fuzzy boundary inference method for revenue rule information according to an embodiment of the present disclosure. Figure 1 As shown, the method comprises the following steps:
[0035] S101, obtaining profit rule information, reference profit parameters at historical moments, and reference profit data corresponding to the reference profit parameters.
[0036] Taking the electricity market scenario as an example, the revenue rule information stipulates the revenue calculation method and assessment criteria under the ideal operating state.
[0037] In some implementations, the actual operating environment of the grid-connected entity is extremely complex, and non-ideal operating conditions often occur. Taking secondary frequency regulation as an example, the revenue rule information specifies the performance and revenue calculation method of the power generation unit (the grid-connected entity on the power generation side, such as a thermal power plant, a hydropower plant, and new energy storage, such as a flywheel energy storage, etc.) when it responds normally.
[0038] By way of example, in some implementations, the revenue rule information may be as follows.
[0039] Assuming that the primary frequency regulation service revenue of a certain power generation unit is R, the calculation formula of R is as follows:
[0040] R=K×D×C R
[0041] Among them, the primary frequency regulation performance index K = K1 × K2 × K3, including response time K1, action rate K2, response accuracy K3, the corresponding explanation is shown in Table 1:
[0042] Table 1
[0043]
[0044] The calculation formula is as follows:
[0045]
[0046] where t d is the lag time when market players participate in a frequency modulation, t N,d is the standard lag time, set to 1s. If K1 is less than 0.1, set to 0.1.
[0047]
[0048] Where V N is the standard response rate, which is 4s; V is the actual response rate of the market entity to reach the target value, V h It is the actual response rate of the market entity to restore the starting value. If K2 is less than 0.1, it is taken as 0.1.
[0049]
[0050] Where P is the actual output value when the market entity makes adjustments, P EIt is the theoretical target value of output when the market players make adjustments. If K3 is less than 0.1, then 0.1 is taken.
[0051] Based on this, we can get the average daily adjustment performance K of the market entity. a :
[0052]
[0053] Where n represents the number of adjustments on the day.
[0054] At the same time, the calculation formula for the primary frequency regulation mileage D is as follows:
[0055] D=N×ΔP s·max
[0056] Where N is the equivalent number of frequency modulation actions. When the frequency exceeds the dead zone of frequency modulation, the market entity performs a frequency modulation until the frequency returns to the dead zone of frequency modulation. The duration of the action is t.
[0057] When t≤30s, N is 1;
[0058] When t>30s, every 30s is counted as one equivalent action. If the final time is less than 30s, it is counted as one equivalent action.
[0059] ΔP s·max is the maximum contribution of primary frequency modulation, which is:
[0060]
[0061] Where, f(t) represents the frequency of the unit at time t, f n Indicates the rated frequency of the unit, ΔP sn is the active value of the primary frequency regulation service that the market entity should provide, which is determined based on prior knowledge. P0 is the active value of the market entity when the frequency exceeds the primary frequency regulation dead zone, P s (t) is the active power value of the market entity during the primary frequency modulation action, and δ is the primary frequency modulation speed inequality rate.
[0062] C R It is the declared price, which is equal to the frequency regulation service settlement price of the market entity on that day. The declared price can be specified during training and is expressed as the charge per megawatt of electricity.
[0063] The market entity’s daily frequency modulation service income R d The calculation formula is:
[0064]
[0065] R j It represents the revenue of the j-th frequency regulation service. The monthly revenue of a frequency regulation of a market entity is the sum of the daily revenues of that month.
[0066] The above is the profit rule information used as an example in this application, the reference profit parameters at historical moments, such as the reference profit parameters of the previous year, and the reference profit data corresponding to the reference profit parameters.
[0067] In some embodiments, the reference benefit parameters may include, for example, response time, action rate, response accuracy, primary frequency modulation adjustment mileage, and declared price. In some embodiments, the reference benefit parameters may include, for example, the lag time when the market subject participates in primary frequency modulation, the standard lag time, the standard response rate, the actual response rate when the market subject reaches the target value, the actual response rate when the market subject recovers the starting value, the actual output value when the market subject adjusts, the theoretical target value of output when the market subject adjusts, the equivalent number of primary frequency modulation actions, the maximum contribution of primary frequency modulation, and the declared price. The reference benefit data may be, for example, the actual benefit data of the previous preset time period.
[0068] S102, performing scenario recognition and rule assumption based on the revenue rule information, reference revenue parameters and reference revenue data, and obtaining a rule assumption set and a rule boundary parameter set under a specified target scenario.
[0069] In actual data processing, the existence of various non-ideal working conditions may affect the profit calculation under the above-mentioned profit rule information. The calculation results of the code written according to the above-mentioned profit rule information may not be able to accurately fit the actual value. For example: the absolute value of the difference between the initial output and the target output of the power generation unit is within the primary frequency regulation dead zone. At this time, it may be possible that the target output value is reached without exceeding 10% of the total primary frequency regulation. In this case, the calculation formula of the primary frequency regulation performance index and the adjustment mileage is unclear. For example, scene recognition can be performed based on profit rule information, reference profit parameters and reference profit data. The scene that meets the preset conditions is an ideal scene, and the scene that does not meet the preset conditions is a non-ideal scene. The preset condition here is that the absolute value of the difference between the initial output and the target output of the power generation unit is not within the primary frequency regulation dead zone. In other implementations, the preset conditions can also be other conditions determined based on the profit rule information, which will not be repeated here.
[0070] In some embodiments, by using data analysis techniques such as clustering, different categories of non-ideal operating conditions can be obtained, and then rule assumptions are made for different operating conditions to obtain possible calculation rules, which are finally summarized into a rule assumption set; at the same time, parameters not specified in the rules are extracted (the value of the parameter is unclear, but the approximate value range of the parameter can be obtained), such as the size of the frequency modulation dead zone, etc., to obtain a rule boundary parameter set.
[0071] For example, some elements in the rule assumption set may be as shown in Table 2, and some elements in the rule boundary parameter set may be as shown in Table 3.
[0072] Table 2
[0073]
[0074] Table 3
[0075] Parameter name Parameter Description p Primary frequency modulation dead zone size <![CDATA[V n ]]> Standard response rate
[0076] S103, performing rule encoding on the rule hypothesis set and the rule boundary parameter set to obtain a gene sequence.
[0077] According to the rule hypothesis set and the rule boundary parameter set, the rules suitable for evolutionary computing are encoded. For the rule boundary parameters, they are directly converted into genes and the upper and lower limits are determined. For example, the size of the frequency modulation dead zone is set to p1, and the search space is [0,3]. For rule assumptions based on non-ideal scenarios, each rule assumption can be regarded as a gene, and the upper and lower limits of the gene search range are set according to the number of rule assumptions. For example, the rule assumption set for the above non-ideal working conditions corresponds to two rule assumptions, so the precision of this parameter is set to an integer, and the search space is {0,1}. Thus, the rule assumption set and the rule boundary parameter set are rule encoded and integrated to obtain the gene sequence.
[0078] S104, performing evolutionary algorithm calculation and decoding on the gene sequence based on a preset fitness function to obtain the boundary of the profit rule information.
[0079] For example, using the preset output data as the data set, the reference benefit data in each frequency modulation instruction cycle is (k1, k2, …, k p ), p∈N * , p is the number of frequency regulation instructions; in this scenario, the reference benefit parameter includes the output data of the power generation unit in each frequency regulation instruction And frequency modulation command value data is the number of data. After obtaining the data input and gene list, input them into the evolutionary algorithm module and set the fitness evaluation function as:
[0080]
[0081] in, represents the predicted revenue data of the ith frequency modulation instruction in the data set, k i represents the reference return data of the ith frequency modulation instruction, and the gene list is {x1,x2,...,x n}. Then, each individual in the population is evaluated according to the fitness function. The smaller the value of the fitness evaluation function, the better the current individual effect. After obtaining the function value, the algorithm will select each individual in the population. A preset selection strategy, such as roulette, can be adopted here. After that, the algorithm will perform genetic operations, including a hybridization strategy using a single-point crossover and a mutation strategy using a basic bit mutation. The completion of the above operations means that a round of iteration is completed. When the number of iterations reaches the maximum number of iterations, the search will stop. At this point, the evolutionary algorithm ends, and the gene corresponding to the individual with the smallest fitness function value can be determined and decoded, thereby determining the boundary of the primary frequency regulation calculation rule in the power market.
[0082] It should be noted that in the embodiments of the present disclosure, each gene corresponds to a rule hypothesis in the rule hypothesis set. Once the gene value is determined, the rule is also determined. Fuzzy boundaries refer to multiple possible calculation rules. The present disclosure can make fuzzy boundaries clear, that is, infer uncertain boundaries. Gene sequences are also gene combinations, and the parameter value space indicates which values a gene can take.
[0083] In the disclosed embodiments, scenario recognition and rule assumptions are performed based on revenue rule information, reference revenue parameters, and reference revenue data to obtain a set of rule assumptions and a set of rule boundary parameters in a specified target scenario; rule encoding is performed on the rule assumption set and the set of rule boundary parameters to obtain a gene sequence; evolutionary algorithm calculation and decoding are performed on the gene sequence based on a preset fitness function to obtain the boundary of the revenue rule information. The boundary of the revenue rule information can be constructed to improve the completeness, universality, and accuracy of the inferred boundary, and an automated solution can be provided for the problem of precise calculation of revenue. The fine-tuning needs of rules in different regions can be adaptively met, and the fuzzy and uncertain market rule boundaries can be inversely solved and inferred in a data-driven manner.
[0084] Figure 2 is a flowchart of a fuzzy boundary inference method for revenue rule information according to an embodiment of the present disclosure. Figure 2 As shown, the method comprises the following steps:
[0085] S201, obtaining profit rule information, reference profit parameters at historical moments, and reference profit data corresponding to the reference profit parameters.
[0086] For the introduction of step S201, please refer to the relevant content in the above embodiment, which will not be repeated here.
[0087] S202: Perform data analysis on the revenue rule information to determine the revenue rule logic under the candidate scenario.
[0088] Conduct data analysis on the revenue rule information to determine the revenue rule logic under candidate scenarios, that is, to build a revenue rule framework applicable to the ideal state.
[0089] S203, based on the profit rule logic, reference profit parameters, reference profit data, and profit rule information, obtain target rule information in the target scenario.
[0090] In the disclosed embodiment, the reference profit parameters are calculated based on the profit rule logic to obtain the predicted profit data, the reference profit data and the predicted profit data are compared, and the inaccurate data segments in the predicted profit data are determined. For example, the difference between each data segment of the reference profit data and the predicted profit data can be obtained. If the difference is within the error range corresponding to the data segment, the data segment is an inaccurate data segment. If the difference is not within the error range corresponding to the data segment, the data segment is an inaccurate data segment.
[0091] Based on the inaccurate data segments, cluster analysis is performed on the profit rule information and profit rule logic. In other words, data processing techniques such as cluster analysis are used to screen and summarize non-ideal scenarios that actually exist in the profit calculation but are not covered by the rules, thereby realizing data-driven non-ideal scenario identification. Optionally, the profit rule information and profit rule logic involved in the inaccurate data segments can be clustered to obtain target rule information under the target scenario.
[0092] It should be noted that in the embodiments of the present disclosure, the candidate scene is an ideal scene, and the target scene is a non-ideal scene. The calculation rules of the ideal scene are fixed. The present disclosure identifies the boundaries under the ideal scene, identifies the non-ideal fuzzy boundaries, and makes the fuzzy boundaries clear, that is, infers the uncertain boundaries. Only non-ideal scenes need to be verified through rule assumptions.
[0093] S204: Make rule assumptions based on the target rule information, and obtain a rule assumption set and a rule boundary parameter set.
[0094] Based on the target rule information, rule assumptions are made, and rule assumption data and rule boundary parameter data are obtained. In some embodiments, the rule assumption text is obtained from a preset database, and the rule assumption text is retrieved and classified based on the target rule information to obtain a rule assumption set and a rule boundary parameter set. In the embodiment of the present application, the elements in the rule assumption set are possible calculation rules for different working conditions under the target scenario, and the elements in the rule boundary parameter set are parameters not specified in the rule and their value ranges.
[0095] In some implementations, cluster analysis is performed on the rule hypothesis data and the rule boundary parameter data to obtain a rule hypothesis set and a rule boundary parameter set.
[0096] S205, performing rule encoding on the rule hypothesis set and the rule boundary parameter set to obtain a gene sequence.
[0097] S206, performing evolutionary algorithm calculation and decoding on the gene sequence based on a preset fitness function to obtain the boundary of the profit rule information.
[0098] For the introduction of step S205 to step S206, reference may be made to the relevant contents in the above embodiment, which will not be repeated here.
[0099] The present disclosure performs data analysis on the revenue rule information, determines the revenue rule logic under the candidate scenarios, obtains the target rule information under the target scenario based on the revenue rule logic, reference revenue parameters, reference revenue data, and revenue rule information, makes rule assumptions based on the target rule information, and obtains a rule assumption set and a rule boundary parameter set. The present disclosure can perform performance adjustment and optimization on the boundaries of the revenue rule information in a targeted manner, and the participating entities can clarify the applicable boundaries of the market rule framework and the detailed rules under special conditions, and then design targeted operation control strategy optimization, and accurately analyze and infer the fuzzy boundaries of the revenue rule information.
[0100] Figure 3 is a flowchart of a fuzzy boundary inference method for revenue rule information according to an embodiment of the present disclosure. Figure 3 As shown, the method comprises the following steps:
[0101] S301, obtaining profit rule information, reference profit parameters at historical moments, and reference profit data corresponding to the reference profit parameters.
[0102] S302, performing scenario identification and rule assumption based on the revenue rule information, reference revenue parameters and reference revenue data, and obtaining a rule assumption set and a rule boundary parameter set under a specified target scenario.
[0103] For the introduction of step S301 to step S302, reference may be made to the relevant contents in the above embodiment, which will not be repeated here.
[0104] S303, real number encoding is performed on the rule hypothesis set to obtain a target rule hypothesis sequence.
[0105] Considering the input restrictions of evolutionary computation, it is necessary to perform rule encoding operations. The rule assumption set P: {p1, p2, …, p s}, s∈N * , perform real number encoding on the rule hypothesis set to obtain a candidate rule hypothesis sequence. Real number encoding is a data encoding method represented by decimal numbers, which can be used to represent integers, decimals, negative numbers, etc.
[0106] That is, the rule hypothesis sequence is mapped to a preset integer set to obtain the target rule hypothesis sequence.
[0107] For example, for any non-ideal operating condition, the rule assumption p i,i∈{1,2,...,s}, and maps it to the integer set {1,2,...,m}, where m represents the total number of candidate rule hypotheses, p i =1 means taking the first candidate rule hypothesis, p i =2 means taking the second candidate rule hypothesis, and so on.
[0108] S304: Constrain the rule boundary parameters based on a preset value range to obtain target boundary parameters.
[0109] For example, the rule boundary parameter set Q:{q1,q2,…,q r}, r∈N * , for Q:{q1,q2,…,q r Any parameter q in i , by setting the upper and lower limits of its value (q max ,q min ) to constrain the value range.
[0110] S305, integrating the target rule hypothesis sequence and the target boundary parameters to obtain a gene sequence suitable for the evolutionary algorithm.
[0111] Before the evolutionary algorithm is used to calculate and decode the gene sequence based on the preset fitness function, it also includes: obtaining the target profit data based on the reference profit parameter, the gene sequence, and the profit rule logic; training the hyperparameters of the evolutionary algorithm based on the target profit data and the reference profit data so that the absolute value of the difference between the target profit data and the reference profit data is minimized.
[0112] Under a certain electricity market revenue rule, the required reference revenue parameters {y1,y2,…,y t}, the reference benefit parameter represents the required parameter data in the actual scenario, which can correspond to different types of parameters in different scenarios; and the corresponding reference benefit data {k1, k2, …, k t}, t∈N * , represents the number of data. At the same time, the gene list formed by combining the rule assumptions and rule boundary parameters is {x1,x2,...,x n}, n∈N * , represents the number of genes. The task of fuzzy rule boundary inference is to use this information to minimize the absolute value of the difference between the predicted value and the true value of the power generation unit revenue, that is:
[0113]
[0114] The system needs to construct the objective function F(·), and set the algorithm hyperparameters (such as population size, crossover rate, mutation rate, and generation number) during the algorithm training process. Finally, the training is completed by using the preset selection strategy and termination conditions.
[0115] S306, performing evolutionary calculation on the gene sequence based on a preset fitness function, and determining the target gene corresponding to the individual with the smallest fitness function value.
[0116] like Figure 4 As shown in the figure, an example is given. After obtaining the rule hypothesis set and the rule boundary parameter set (actual problem parameters), in the actual implementation, in order to deal with consistency considerations, the parameter value space is determined in the form of determining upper and lower limits; discrete and continuous parameters are distinguished by setting precision. Then the parameters are encoded into genes using the real number coding method, and the parameter value space is converted into a gene search space. The parameter value space is determined by the P and Q sets. For example, the p1 value space in P is [1,2,3], and the q1 value space in Q is (1,3). Then the population is initialized, and each individual has only one gene. The "individual" here is used to carry gene values in evolutionary calculations. For example: in individual m, p1 takes the value of 1, q1 takes the value of 1.2, and so on.
[0117] On this basis, the individual's genes are mutated and crossover operated. It is worth noting that this part is used to imitate the reproduction and mutation process of organisms in nature, and is one of the most influential parts of the overall algorithm. Common crossover operations include: single-point crossover, multi-point crossover, and uniform crossover; common mutation operations include: point mutation, insertion mutation, deletion mutation, and inversion mutation. Furthermore, the evolutionary algorithm will perform fitness calculations, which is the part that best reflects the relevance to the rule fuzzy boundary inference task. In the disclosed embodiment, the fitness function can use the following formula:
[0118]
[0119] where k i represents the i-th reference income data, It indicates the predicted revenue data calculated based on the parameter data of the current power generation unit. Next, the stopping condition is judged. For example, the maximum number of iterations can be set as the stopping condition. When the number of iterations exceeds the preset iteration threshold, the iteration is stopped and the gene of the individual with the smallest fitness function value is output; otherwise, the individual is updated and the algorithm steps are repeated.
[0120] S307, decoding the target gene to determine the boundary of the profit rule information.
[0121] like Figure 4 As shown in the figure, after obtaining the optimal individual gene, decoding is performed to obtain the optimal parameters, and the power market profit rule F(·) in the current problem is obtained to complete the fuzzy boundary inference task.
[0122] Based on this, the evolutionary computing processing method adapted to the current problem can accurately inverse and infer the fuzzy boundaries of the electricity market revenue rules, providing an optimized solution for the accurate calculation of the revenue of power generation units participating in the electricity market.
[0123] In the disclosed embodiment, the rule hypothesis set is real-number encoded to obtain a target rule hypothesis sequence, the rule boundary parameters are constrained based on a preset numerical range to obtain the target boundary parameters, the target rule hypothesis sequence and the target boundary parameters are integrated to obtain a gene sequence suitable for the evolutionary algorithm, which can improve the completeness, universality and accuracy of the inferred boundary, perform evolutionary calculations on the gene sequence based on a preset fitness function, determine the target gene corresponding to the individual with the smallest fitness function value, decode the target gene, determine the boundary of the profit rule information, automatically solve the problem of accurate profit calculation, adapt to the needs of fine-tuning rules in different regions, and data-drivenly reverse and infer the boundaries of fuzzy and uncertain market rules.
[0124] Figure 5 is a structural block diagram of a fuzzy boundary inference device for profit rule information according to an embodiment of the present disclosure, such as Figure 5 As shown, the fuzzy boundary inference device 500 for benefit rule information includes:
[0125] The first acquisition module 510 is used to acquire the revenue rule information, the reference revenue parameters at the historical moment, and the reference revenue data corresponding to the reference revenue parameters;
[0126] The second acquisition module 520 is used to perform scenario recognition and rule assumption based on the revenue rule information, reference revenue parameters and reference revenue data, and acquire a rule assumption set and a rule boundary parameter set under a specified target scenario;
[0127] The third acquisition module 530 is used to perform rule encoding on the rule hypothesis set and the rule boundary parameter set to obtain the gene sequence;
[0128] The boundary inference module 540 is used to perform evolutionary algorithm calculation and decoding on the gene sequence based on a preset fitness function to obtain the boundary of the profit rule information.
[0129] In some implementations, the second acquisition module 520 is further configured to:
[0130] Conduct data analysis on the revenue rule information to determine the revenue rule logic under the candidate scenarios;
[0131] Based on the profit rule logic, reference profit parameters, reference profit data, and profit rule information, obtain the target rule information under the target scenario;
[0132] Make rule assumptions based on the target rule information and obtain a rule assumption set and a rule boundary parameter set.
[0133] In some implementations, the second acquisition module 520 is further configured to:
[0134] Calculate the reference profit parameters based on the profit rule logic to obtain the predicted profit data;
[0135] Compare the reference income data with the forecast income data to determine the inaccurate data segments in the forecast income data;
[0136] Based on the inaccurate data segments, cluster analysis is performed on the revenue rule information and revenue rule logic, and the target rule information under the target scenario is determined from the revenue rule information and revenue rule logic.
[0137] In some implementations, the boundary inference module 540 is further configured to:
[0138] Obtain target income data based on reference income parameters, gene sequences, and income rule logic;
[0139] The hyperparameters of the evolutionary algorithm are trained based on the target profit data and the reference profit data so that the absolute value of the difference between the target profit data and the reference profit data is minimized.
[0140] In some implementations, the second acquisition module 520 is further configured to:
[0141] Make rule assumptions based on target rule information and obtain rule assumption data and rule boundary parameter data;
[0142] Cluster analysis is performed on the rule hypothesis data and rule boundary parameter data respectively to obtain the rule hypothesis set and the rule boundary parameter set.
[0143] In some implementations, the third acquisition module 530 is further configured to:
[0144] Perform real number encoding on the rule hypothesis set to obtain the target rule hypothesis sequence;
[0145] Constrain the rule boundary parameters based on the preset value range to obtain the target boundary parameters;
[0146] The target rule hypothesis sequence and target boundary parameters are integrated to obtain the gene sequence suitable for evolutionary algorithm.
[0147] In some implementations, the boundary inference module 540 is further configured to:
[0148] Perform evolutionary calculation on the gene sequence based on the preset fitness function to determine the target gene corresponding to the individual with the smallest fitness function value;
[0149] Decode the target gene and determine the boundaries of the profit rule information.
[0150] The disclosed embodiments can construct the boundaries of revenue rule information, improve the completeness, universality and accuracy of the inferred boundaries, automatically solve the problem of accurate calculation of revenue, adapt to the needs of fine-tuning rules in different regions, and perform data-driven inversion and inference of fuzzy and uncertain market rule boundaries.
[0151] The acquisition, transmission, storage, use, and processing of data in the technical solution disclosed in this disclosure are in compliance with the relevant provisions of laws and regulations.
[0152] It should be noted that in the embodiments of the present disclosure, certain software, components, models and other existing solutions in the industry may be mentioned, which should be regarded as exemplary. Their purpose is only to illustrate the feasibility of implementing the technical solution of the present application, but it does not mean that the applicant has or will necessarily use the solution.
[0153] Figure 6 It is a schematic diagram of the structure of an electronic device according to an embodiment of the present disclosure.
[0154] like Figure 6 As shown, the electronic device 800 includes:
[0155] The memory 801 and the processor 802, a bus 803 connecting different components (including the memory 801 and the processor 802), the memory 801 stores a computer program, and when the processor 802 executes the program, the fuzzy boundary inference method of the profit rule information of the embodiment of the present disclosure is implemented.
[0156] Bus 803 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processor or a local bus using any of a variety of bus structures. For example, these architectures include but are not limited to Industry Standard Architecture (ISA) bus, Micro Channel Architecture (MAC) bus, Enhanced ISA bus, Video Electronics Standards Association (VESA) local bus and Peripheral Component Interconnect (PCI) bus.
[0157] The electronic device 800 typically includes a variety of electronic device readable media. These media can be any available media that can be accessed by the electronic device 800, including volatile and non-volatile media, removable and non-removable media.
[0158] The memory 801 may also include computer system readable media in the form of volatile memory, such as random access memory (RAM) 804 and / or cache memory 805. The electronic device 800 may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, the storage system 806 may be used to read and write non-removable, non-volatile magnetic media ( Figure 6 not shown, usually called a "hard drive"). Although Figure 6 Not shown in the figure, a disk drive for reading and writing to a removable non-volatile disk (e.g., a "floppy disk"), and an optical disk drive for reading and writing to a removable non-volatile optical disk (e.g., a CD-ROM, a DVD-ROM, or other optical media) may be provided. In these cases, each drive may be connected to the bus 803 via one or more data medium interfaces. The memory 801 may include at least one program product having a set (e.g., at least one) of program modules that are configured to perform the functions of the various embodiments of the present disclosure.
[0159] A program / utility 808 having a set (at least one) of program modules 807 may be stored, for example, in the memory 801, such program modules 807 including but not limited to an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment. The program modules 807 generally perform the functions and / or methods of the embodiments described in the present disclosure.
[0160] The electronic device 800 may also communicate with one or more external devices 809 (e.g., keyboard, pointing device, display 811, etc.), one or more devices that enable a user to interact with the electronic device 800, and / or any device that enables the electronic device 800 to communicate with one or more other computing devices (e.g., network card, modem, etc.). Such communication may be performed through an input / output (I / O) interface 812. Furthermore, the electronic device 800 may also communicate with one or more networks (e.g., local area network (LAN), wide area network (WAN) and / or public network, such as the Internet) through a network adapter 813. Figure 6 As shown, the network adapter 813 communicates with other modules of the electronic device 800 via the bus 803. It should be understood that, although not shown in the figure, other hardware and / or software modules can be used in conjunction with the electronic device 800, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.
[0161] The processor 802 executes various functional applications and data processing by running the programs stored in the memory 801 .
[0162] It should be noted that the implementation process and technical principles of the electronic device of this embodiment can be found in the aforementioned explanation of the fuzzy boundary inference method of the revenue rule information of the embodiment of the present disclosure, and will not be repeated here.
[0163] In order to implement the above embodiments, the present disclosure also proposes a computer-readable storage medium.
[0164] When the instructions in the computer-readable storage medium are executed by the processor of the electronic device, the electronic device can perform the fuzzy boundary inference method of the profit rule information as described above. Optionally, the computer-readable storage medium can be a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, etc.
[0165] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the specification and practicing the invention disclosed herein. The present disclosure is intended to cover any variations, uses or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary techniques in the art that are not disclosed in the present disclosure. The description and examples are to be considered exemplary only, and the true scope and spirit of the present disclosure are indicated by the following claims.
[0166] It should be understood that the present disclosure is not limited to the exact structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.
Claims
1. A fuzzy boundary inference method for profit rule information, characterized in that: include: Obtaining profit rule information, reference profit parameters at historical moments, and reference profit data corresponding to the reference profit parameters; Based on the revenue rule information, the reference revenue parameters and the reference revenue data, scenario identification and rule assumptions are performed to obtain a rule assumption set and a rule boundary parameter set under a specified target scenario; Performing rule encoding on the rule hypothesis set and the rule boundary parameter set to obtain a gene sequence; Based on a preset fitness function, the gene sequence is calculated and decoded using an evolutionary algorithm to obtain the boundary of the profit rule information.
2. The method according to claim 1, characterized in that The performing of scenario identification and rule assumption based on the revenue rule information, the reference revenue parameter and the reference revenue data to obtain a rule assumption set and a rule boundary parameter set under a specified target scenario includes: Performing data analysis on the revenue rule information to determine the revenue rule logic under the candidate scenarios; Based on the profit rule logic, the reference profit parameter, the reference profit data, and the profit rule information, obtaining target rule information under a target scenario; A rule hypothesis is made based on the target rule information to obtain the rule hypothesis set and the rule boundary parameter set.
3. The method according to claim 2, characterized in that The acquiring target rule information in a target scenario based on the revenue rule logic, the reference revenue parameter, the reference revenue data, and the revenue rule information includes: Calculate the reference profit parameter based on the profit rule logic to obtain predicted profit data; Comparing the reference revenue data with the predicted revenue data, and determining inaccurate data segments in the predicted revenue data; A cluster analysis is performed on the revenue rule information and the revenue rule logic based on the inaccurate data segment, and the target rule information under the target scenario is determined from the revenue rule information and the revenue rule logic.
4. The method according to claim 3, characterized in that Before performing evolutionary algorithm calculation and decoding on the gene sequence based on the preset fitness function, the method further includes: Based on the reference profit parameter, the gene sequence, and the profit rule logic, obtaining target profit data; The hyperparameters of the evolutionary algorithm are trained based on the target profit data and the reference profit data so that the absolute value of the difference between the target profit data and the reference profit data is minimized.
5. The method according to claim 2, characterized in that: The performing rule assumptions based on the target rule information to obtain the rule assumption set and the rule boundary parameter set includes: Make a rule hypothesis based on the target rule information, and obtain rule hypothesis data and rule boundary parameter data; Cluster analysis is performed on the rule hypothesis data and the rule boundary parameter data respectively to obtain the rule hypothesis set and the rule boundary parameter set.
6. The method according to claim 1, characterized in that The step of performing rule encoding on the rule hypothesis set and the rule boundary parameter set to obtain a gene sequence includes: Performing real number encoding on the rule hypothesis set to obtain a target rule hypothesis sequence; Constraining the rule boundary parameters based on a preset value range to obtain target boundary parameters; The target rule hypothesis sequence and the target boundary parameter are integrated to obtain the gene sequence suitable for the evolutionary algorithm.
7. The method according to claim 1, characterized in that The step of performing evolutionary calculation and decoding on the gene sequence based on a preset fitness function to obtain the boundary of the profit rule information includes: Performing evolutionary calculation on the gene sequence based on a preset fitness function to determine the target gene corresponding to the individual with the smallest fitness function value; The target gene is decoded to determine the boundary of the revenue rule information.
8. A fuzzy boundary inference device for profit rule information, characterized in that: include: A first acquisition module is used to acquire revenue rule information, reference revenue parameters at historical moments, and reference revenue data corresponding to the reference revenue parameters; A second acquisition module is used to perform scenario recognition and rule assumption based on the revenue rule information, the reference revenue parameter and the reference revenue data, and acquire a rule assumption set and a rule boundary parameter set under a specified target scenario; A third acquisition module is used to perform rule encoding on the rule hypothesis set and the rule boundary parameter set to obtain a gene sequence; The boundary inference module is used to perform evolutionary algorithm calculation and decoding on the gene sequence based on a preset fitness function to obtain the boundary of the profit rule information.
9. An electronic device, comprising: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 7.
10. A non-transitory computer-readable storage medium storing computer instructions, wherein: The computer instructions are used to cause the computer to execute the steps of the method according to any one of claims 1-7.
