Optimization method and system for electricity transaction retail package

Abstract

The invention relates to the technical field of electricity transaction management, in particular to an electricity transaction retail package optimization method and system, and the method comprises the steps: constructing an electricity selling company cost measurement and calculation model to calculate a monthly cost measurement and calculation price, and setting the monthly cost measurement and calculation price as a package pricing cost base line; establishing an objective function of package optimization by taking the revenue maximization of the electricity selling company as an objective, and forming an initial optimization model; a compliance package optimization model is formed by loading preset package constraint conditions; solving the compliance package optimization model by adopting an intelligent optimization algorithm, ensuring that the solved package price is higher than the cost base line, and outputting an optimal package solution; the optimal package solution comprises a scheme for dividing peak, flat and valley time periods of all moments generated every month, and electricity prices which correspond to all the time periods and conform to the package price multiple rule. The problems that an existing package mode is low in efficiency, low in customer satisfaction and unstable in income, and market contradictions are likely to be caused are solved.

Description

TECHNICAL FIELD

[0001] The application relates to the technical field of power transaction management, in particular to an optimization method and system for a power transaction retail package. BACKGROUND

[0002] With the deepening of the power market reform, power selling companies are facing the challenge of transforming from traditional distribution to market-oriented services. Although power retail packages can improve customer satisfaction and power selling company revenue in theory, they still have technical bottlenecks in actual design and operation.

[0003] Currently, the package design process highly depends on experience and has the following problems: First, the automation level is low, and the decision-making efficiency and accuracy are insufficient. Package design needs to consider multiple dynamic factors such as price fluctuations, time-of-use electricity demand, and user electricity habits, resulting in long development cycles, high costs, and an inability to quickly respond to market changes.

[0004] Second, the strategy is single and rigid, and it is difficult to meet individual needs. Existing estimation models lack flexibility and cannot generate customized solutions based on customer industry characteristics, electricity consumption curves, and price sensitivity. This results in low matching between packages and actual user needs, poor customer satisfaction and signing willingness, and limits the market competitiveness and revenue growth of power selling companies.

[0005] Third, the compliance cost is high, and it is difficult to find an optimal solution under complex constraints. Package design must strictly follow a series of mandatory technical regulations from the power grid company, such as peak, flat, and valley period division, time length ratio, and price multiple relationship. Under such complex multi-constraint conditions, it is almost impossible to find the optimal solution through experience in a large number of permutations and combinations, which severely restricts the scale development of the business.

[0006] Therefore, the industry urgently needs a solution that can automatically and intelligently handle complex constraints and quickly generate individualized, high-yield package solutions. SUMMARY

[0007] To solve the above problems, the present application provides an optimization method and system for a power transaction retail package, which realizes fast package development, adjustment, and push through digital means, accelerates the intelligent transformation of the market, and improves operational efficiency and service quality.

[0008] In a first aspect, the present application provides an optimization method for a power transaction retail package, comprising the following steps: Obtain the customer's electricity fee basic data and the pre-signed electricity purchase contract data; Based on the electricity cost data and the pre-signed electricity purchase contract data, a power selling company cost estimation model is constructed to calculate a monthly cost estimation price, and the monthly cost estimation price is set as a cost bottom line of the package pricing. With the goal of maximizing the revenue of the power selling company, an initial optimization model is established. The preset package constraint conditions are loaded into the initial optimization model to form a final compliance package optimization model. An intelligent optimization algorithm is used to solve the compliance package optimization model, and it is ensured that the solved package price is higher than the cost bottom line, and an optimal package solution is output; the optimal package solution includes a peak, flat, valley period division scheme generated for each time of each month, and a corresponding electricity price that meets the package price multiple rule.

[0009] As a further limitation of the technical scheme of the application, the power selling company cost estimation model is constructed to calculate the monthly cost estimation price, and the specific formula is:

[0010] Wherein, is the actual decomposed electricity quantity of the month day hour, is the actual electricity price of the month day hour, is the medium and long-term electricity quantity of the month day hour, is the medium and long-term electricity price of the month day hour, is the total number of days of the month is the floating price of the enterprise's acquisition cost; ; ; Wherein, represents the electricity time weight of the month hour, and the corresponding electricity price type is the weight of five electricity price types: peak, peak, flat, valley, and deep valley, is the 5-type time period proportion of the month hour, and the total number of time periods is 24, corresponding to the peak, peak, flat, valley, and deep valley price types, This represents the total electricity consumption for each of the five electricity price categories. This refers to the officially released breakdown curve percentages for each month, day, and hour. and The ratio represents the proportion of two types of transactions, namely the officially authorized proportion and the proportion of medium- and long-term transactions. and These respectively represent official authorization and medium-to-long-term... moon sky The existing electricity bill at that time was broken down into electricity prices.

[0011] By constructing a complete mathematical model and employing intelligent optimization algorithms for solving it, the traditional model relying on human experience has been completely transformed. The system can automatically process massive amounts of electricity bill and contract data, quickly and accurately complete cost calculations and package optimization, shortening the original package design cycle of several days or even weeks to hours or even minutes. This enables electricity sales companies to respond quickly to market changes and significantly reduces labor and time costs.

[0012] As a further limitation of the technical solution of this invention, the objective function for package optimization is:

[0013] , ; In the formula, For the first Estimated monthly package price For the first moon The package price is the same for the same time slot every day. For the estimated number of packages moon sky Real-time decomposition of electricity, For time-sharing percentages, To superimpose a fixed-price constant, The proportion of spot goods, Indicates the first moon Time-based weighting of package power consumption For the first Month Total number of electricity price types Indicates the first The total number of the three electricity price types included in the month, corresponding to peak, flat, and valley respectively.

[0014] Based on customers' historical electricity consumption data, the optimization model can tailor the best package plan for customers with different electricity consumption habits and needs. By outputting personalized 24-hour peak, flat, and off-peak time periods and corresponding electricity prices, it greatly improves the fit between the package and the customer's actual electricity consumption curve, thereby effectively improving customer satisfaction and willingness to sign up, and solving the problem of poor adaptability of the traditional one-size-fits-all package model.

[0015] As a further limitation of the technical solution of the present invention, the package constraints include: Package time period mapping rules: Map the five time periods of the basic electricity price type (peak, peak, flat, valley, and deep valley) to the package electricity price time periods that only allow three types of peak, flat, and valley; Package pricing multiplier rules: Set a first fixed multiplier relationship between the monthly package's flat-peak electricity price and off-peak electricity price, and a second fixed multiplier relationship between the flat-peak electricity price and peak-peak electricity price; As a further limitation of the technical solution of the present invention, the preset package constraints are loaded into the initial optimization model, specifically including the following steps: Regarding the time-slot mapping rules for package deals: Define a basic electricity price type vector for each time t out of the 24 times of the day. and a vector of allowed package electricity price types ; The mapping rule is transformed into a constraint on the value of each time t, which restricts the decision variable, namely the package electricity price type at time t. Must meet ; Regarding the rules for multipliers in package prices: The rules are transformed into price relationship constraints, that is, for any month m, the average price of the package segment. Valley segment price Peak price The following conditions must be met:

[0016]

[0017] in, , It is a preset fixed constant; By introducing value constraints and price relationship constraints, the initial optimization model is transformed into the compliant package optimization model.

[0018] The prescribed mandatory package constraints (such as time period mapping rules, price multiple relationships, etc.) are integrated into the optimization model as a hard condition set. The system can automatically find and output the optimal package solution that satisfies all constraint conditions and maximizes the retail electricity company's revenue within the compliance framework. The business compliance is ensured, and the company's revenue is maximized at the same time.

[0019] The retail electricity company can provide better quality and more diversified retail package services at lower cost and higher efficiency, which not only enhances its own profitability and market competitiveness, but also improves market transparency and resource allocation efficiency through digital means.

[0020] As a further limitation of the technical solution of the application, the package time period mapping rule is specifically: The basic electricity price peak period is mapped to the package peak period; The basic electricity price peak period is mapped to the package peak period or the flat period; The basic electricity price flat period is mapped to the package peak period, flat period or valley period; The basic electricity price valley period is mapped to the package flat period or valley period; The basic electricity price deep valley period is mapped to the package valley period.

[0021] As a further limitation of the technical solution of the application, the decision variable of the initial optimization model is the peak, flat, and valley period division of the package and the electricity price of each period; the smart optimization algorithm is used to solve the compliance package optimization model, which is to encode the decision variable into a chromosome and then iteratively optimize it to find the combination of decision variables that maximizes the objective function value.

[0022] As a further limitation of the technical solution of the application, the smart optimization algorithm is used to solve the compliance package optimization model and ensure that the solved package price is higher than the cost bottom line, and the optimal package solution is output, specifically including the following steps: S61: Select the genetic algorithm as the smart optimization algorithm and configure the algorithm parameters, including population size, crossover rate, mutation rate and maximum iteration number; S62: Encode the decision variable into the chromosome of the algorithm; wherein the chromosome gene includes discrete coding representing the peak, flat, and valley type division of 24 time points per month, and continuous coding representing the peak, flat, and valley electricity price; based on the coding rule, an initial population is randomly generated; S63: Perform the following operations on each generation of population: S631: Calculate the revenue of each chromosome corresponding scheme as the original fitness; check whether all the value constraints, price relationship constraints and cost bottom line constraints are satisfied; for the chromosome that violates the constraints, apply fitness penalty according to the violation degree; S632: Select, crossover, and mutation operations are performed based on fitness to generate a new generation of population; S64: When the maximum number of iterations is reached, select the chromosome that satisfies all constraints and has the highest original fitness from the population of all generations, decode it into the optimal solution and output it.

[0023] As a further limitation of the technical solution of the present invention, S62 includes: A chromosome is defined as consisting of time-division segments and valence segments cascaded together; a time-division segment is a sequence of length 24. Each gene locus in the sequence The value of is a discrete integer. =1 indicates that time t is divided into the peak period. =2 indicates the normal segment. =3 indicates a valley period; the electricity price segment is a sequence of length 3. The gene positions in the sequence are continuous real values, representing the peak, flat and off-peak electricity prices of the package, respectively; Based on the defined chromosome structure, perform the following operations to generate an initial population containing a preset number of chromosomes: For time-segmented segments, each gene locus The values ​​are randomly selected uniformly from the set {1,2,3}; For the electricity price segment, the value of each gene locus is uniformly and randomly generated within the preset range of corresponding electricity price values.

[0024] Secondly, the present invention also provides an optimization system for electricity trading retail packages, comprising: The data acquisition module is configured to acquire customers' basic electricity bill data and pre-signed electricity purchase contract data; The cost calculation module is configured to build a cost calculation model for the electricity sales company based on basic electricity cost data and pre-signed electricity purchase contract data to calculate the monthly cost calculation price, and set the monthly cost calculation price as the cost floor for package pricing; The model building module is configured to establish an objective function for optimizing the electricity sales company's revenue, thereby forming an initial optimization model. The constraint loading module is configured to load preset package constraints into the initial optimization model to form the final compliant package optimization model. The intelligent solution module is configured to use an intelligent optimization algorithm to solve the compliant package optimization model and ensure that the solved package price is higher than the cost floor, and output the optimal package solution; the optimal package solution includes a peak, flat and valley time period division scheme for 24 time periods generated each month, and the electricity price corresponding to each time period that conforms to the package price multiple rule.

[0025] As a further limitation of the technical solution of the application, the cost estimation module constructs a cost estimation model of the power selling company to calculate the monthly cost estimation price, and the specific formula is:

[0026] Wherein, is the actual decomposed power of the month day hour, is the actual electricity price of the month day hour, is the medium and long term power of the month day hour, is the medium and long term electricity price of the month day hour, is the total number of days in the month is the floating price of the cost obtained by the enterprise; ; ; Wherein, represents the power time weight of the month hour, and the corresponding electricity price type is the weight of five electricity price types of peak, peak, flat, valley and deep valley, is the proportion of 5 kinds of electricity price types in the month hour, and the total number of time periods is 24, which respectively correspond to the peak, peak, flat, valley and deep valley price types, is the total of the power of the five kinds of electricity price types, is the official public decomposition curve proportion value of each month, each day and each hour, and are the proportions of the two kinds of transactions, respectively representing the official authorized proportion and the medium and long term transaction proportion; and respectively represent the existing electricity bill decomposition price of the month day hour of the official authorization and the medium and long term.

[0027] As a further limitation of the technical solution of the application, the model construction module is specifically configured to establish the following objective function:

[0028] , ; In the formula, is the estimated price of the package in the first month, is the package price in the first month at the same time period every day, is the real-time decomposition power of the package in the first month day at the same time period, is the time-sharing proportion, is the superimposed fixed price constant, is the proportion of spot, represents the time-sharing weight of the package power in the first month at the same time period, is the total number of power types of the first month type, represents the total number of three power types in the first month, which corresponds to peak, flat, and valley, respectively.

[0029] As a further limitation of the technical scheme of the application, the constraint loading module loads the following constraints: Package time period mapping rule: mapping the five time periods of the basic power type, namely peak, peak, flat, valley, and deep valley, to the package power time period which only allows three types of peak, flat, and valley; Package price multiple rule: setting a first fixed multiple relationship between the flat segment price and the valley segment price of the package in each month, and a second fixed multiple relationship between the flat segment price and the peak segment price; As a further limitation of the technical scheme of the application, the constraint loading module is specifically configured for: For the package time period mapping rule: For each time t in the 24 hours of each day, define a basic power type vector and an allowed package power type vector ; Convert the mapping rule into the value constraint of each time t, limit the decision variable, i.e. the package power type at time t, which must satisfy ; For the package price multiple rule: Convert the rule into a price relationship constraint, i.e. for any month m, the flat segment price , the valley segment price , and the peak segment price ​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​Satisfy between:

[0030]

[0031] Wherein, , Is a preset fixed constant; By introducing the value constraint and the price relationship constraint, the initial optimization model is converted into the compliance package optimization model.

[0032] As a further limitation of the technical scheme of the application, the package period mapping rule is specifically: The basic electricity price peak period is mapped to the package peak period; The basic electricity price peak period is mapped to the package peak period or flat period; The basic electricity price flat period is mapped to the package peak period, flat period or valley period; The basic electricity price valley period is mapped to the package flat period or valley period; The basic electricity price deep valley period is mapped to the package valley period.

[0033] As a further limitation of the technical scheme of the application, the intelligent solving module specifically includes: The algorithm configuration unit is configured to select the genetic algorithm as the intelligent optimization algorithm, and configure the population size, crossover rate, mutation rate and maximum iteration number parameters; The coding and initialization unit is configured to encode the decision variables into chromosomes and generate an initial population; Wherein, the chromosome gene includes discrete coding representing 24 time point peak, flat, valley type division per month, and continuous coding representing peak, flat, valley price; The iterative optimization unit is configured to iteratively optimize the population, and the iterative optimization unit further includes: The fitness evaluation subunit is configured to calculate the original fitness of each chromosome, check the violation degree of each constraint, and apply fitness penalty to obtain the final fitness; The evolution operation subunit is configured to perform selection, crossover and mutation operations according to the final fitness to generate a new generation of population; The result output unit is configured to output the optimal package solution decoded from the chromosome with the highest original fitness and satisfying all constraints when the maximum iteration number is reached.

[0034] From the above technical solution can be seen, the present application has the following advantages: constructing a scientific package calculation model, combining customer multi-dimensional data to customize a package that meets the needs, improving customer recognition and satisfaction, and thus improving the signing intention and signing rate. With the help of digital means, the package is quickly formulated, adjusted and pushed, accelerating the market transformation to digitalization and intelligentization, and improving the operation efficiency and service quality.

[0035] The package is automatically and intelligently formulated, reducing manual intervention and reducing labor costs, solving the problems of low efficiency, easy errors and high cost in manual operation under the existing mode. The package is dynamically adjusted by comprehensively considering the market and customer demand, helping the electricity sales company to maximize the profit, enhance the competitiveness and profitability, and overcome the disadvantages of single calculation scheme that is difficult to flexibly adjust according to market and customer demand to affect the profit.

[0036] A power transaction retail package optimization model based on multiple constraint conditions is proposed, which comprehensively considers constraints, customer demand and market dynamics, and solves the optimal package by using an intelligent algorithm. Break the single limitation of traditional package calculation scheme, provide a variety of targeted packages, improve the adaptability of the package and market competitiveness, and meet the diversified electricity demand of customers.

[0037] Optimize the package strategy, balance the interests of customers and electricity sales companies, and improve the market transparency and fairness in a digital and intelligent mode. Promote the rational allocation of resources, create a fair, orderly and sustainable development environment, solve the problems of low efficiency, low customer satisfaction and unstable income of the existing package mode, and ensure the long-term benign operation of the market. BRIEF DESCRIPTION OF DRAWINGS

[0038] In order to more clearly illustrate the technical solutions of the present application, the drawings needed in the description will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0039] Figure 1 The flowchart of the method provided by the embodiment of the present application is shown.

[0040] Figure 2 The block diagram of the system provided by the embodiment of the present application is shown. DETAILED DESCRIPTION

[0041] In order to make the application purposes, features, advantages of the present application more obvious and easy to understand, the following will use specific examples and drawings to clearly and completely describe the technical solutions protected by the present application. Obviously, the following described examples are only a part of the examples of the present application, but not all the examples. Based on the examples in the present application, all other examples obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0042] Unless otherwise defined, all technical and scientific terms used in the present application have the same meanings as commonly understood by those skilled in the art to which the present application belongs. The terms used in the specification of the present application are only for the purpose of describing specific examples, and are not intended to limit the present application.

[0043] As shown in Figure 1 The embodiment of the present application provides an optimization method of power transaction retail package, which comprises the following steps: S1, obtaining the electricity fee basic data of the customer and the pre-signed electricity purchase contract data; specifically, collecting data from the electricity metering system and the electricity transaction platform system through a pre-set standardized data interface; wherein the electricity fee basic data at least includes the historical electricity load curve of the customer, the actual executed electricity quantity and electricity price data; and the pre-signed electricity purchase contract data at least includes the electricity quantity, electricity price of the official authorized transaction contract and the bilateral electricity transaction contract, and the official published decomposition curve proportion value.

[0044] In the electricity fee basic data collection link, efficient and stable TCP network protocol and standardized data import interface are adopted to accurately capture and store key basic data such as electricity fee bill. Through this technical means, the integrity and accuracy of the data are ensured, and solid and reliable data support is provided for subsequent package calculation, so that the whole calculation process is established on the basis of real and effective data.

[0045] S2, based on the electricity fee basic data and the pre-signed electricity purchase contract data, constructing a power selling company cost calculation model to calculate a monthly cost calculation price, and setting the monthly cost calculation price as the cost bottom line of package pricing; In the optimization design of the calculation model, firstly, a power selling company cost calculation price model and a customer package cost price model are constructed respectively, which lay a foundation for accurate cost evaluation from different angles. Then, combined with rich information such as customer historical electricity fee bills, a complete annual package model of the power selling company for customers is carefully constructed. The model can comprehensively consider various factors, accurately give the optimal package solution on the basis of the power selling company cost calculation price, and provide customers with more cost-effective choices.

[0046] In view of the current power trading market, there are many problems, such as the artificial package calculation process is complicated and inefficient, the package calculation scheme is single and difficult to meet the diversified needs of customers, the customer signing willingness is low, and the electricity sales company's revenue needs to be improved. A retail package optimization model based on multiple constraint conditions is proposed. This model fully considers various actual constraint factors, and establishes a retail package maximum revenue objective function, aiming to maximize the revenue of the electricity sales company through scientific and reasonable model design, while improving customer satisfaction and market competitiveness. Through the precise operation of the algorithm, the specific division of the 24-hour peak, flat and valley periods of the package per month can be solved, as well as the corresponding peak, flat and valley electricity price values. The final optimal feasible solution that meets all conditions is the optimal package solution that meets the preset format requirements.

[0047] The cost calculation price of the electricity sales company plays a key role in the development of retail packages. To ensure the accuracy and practicality of cost calculation, monthly is used as the basic time scale for calculation, and the monthly calculation results are summarized to obtain the annual total price. This price accurately reflects the minimum cost required by the electricity sales company when cooperating with power customers and issuing retail packages. From the perspective of business operation, if the actual package price is lower than the cost calculation price, the possibility of the electricity sales company making a profit in cooperation with customers will be minimal. Therefore, when developing retail packages, a basic requirement must be strictly followed, that is, the package price should always be higher than the cost calculation price of the electricity sales company, in order to protect the basic economic interests of the electricity sales company in cooperation and ensure the sustainability and stability of the business.

[0048] Monthly cost calculation price, specific formula:

[0049] Among them, is the actual decomposed electricity of the month day hour, is the actual electricity price of the month day hour, is the medium and long-term electricity of the month day hour, is the medium and long-term electricity price of the month day hour, is the total number of days in the month is the floating price of the enterprise's acquisition cost; ; ; wherein, represents the electricity time-sharing weight at the month , and the corresponding electricity price type is the weight of five price types of peak, peak, flat, valley, and deep valley, is the proportion of the 5 price types in the month , and the total number of time periods is 24, corresponding to the price types of peak, peak, flat, valley, and deep valley, is the total of the electricity of the five price types, is the official published decomposition curve ratio value, and are the proportions of the two types of transactions, respectively representing the official authorized proportion and the medium and long-term transaction proportion; and represent the existing electricity bill decomposition price of the month day at the time.

[0050] S3, maximize the revenue of the electricity sales company as the goal, establish the target function of the package optimization, form the initial optimization model; The target function of the package optimization is:

[0051] , ; In the formula, is the estimated price of the month , and is the package price at the time of the month , which has the same price in the same time period every day, is the real-time decomposition electricity of the package estimated month day at the time, is the time-sharing proportion, is the superimposed fixed price constant, is the proportion of spot, represents the package electricity time-sharing weight at the month time, is the total number of price types of the month time.The total number of the three types of electricity prices corresponds to the peak, flat, and valley, respectively.

[0052] S4, loading the preset package constraint condition into the initial optimization model to form a final compliance package optimization model; the package constraint condition comprises: The package period mapping rule: mapping the five periods of the basic electricity price type, i.e., the peak, the peak, the flat, the valley, and the deep valley, to the package electricity price period of the three types of peak, flat, and valley; The package price multiple rule: setting a first fixed multiple relationship between the flat period price and the valley period price of the package in each month, and a second fixed multiple relationship between the flat period price and the peak period price; In this step, the preset package constraint condition is loaded into the initial optimization model, specifically comprising the following steps: For the package period mapping rule: For each time t in the 24 hours of a day, a basic electricity price type vector and an allowed package electricity price type vector are defined. The mapping rule is converted into a value constraint for each time t, limiting the decision variable, i.e., the package electricity price type of time t, which must satisfy ; For the package price multiple rule: The rule is converted into a price relationship constraint, i.e., for any month m, the flat period price , the valley period price , and the peak period price satisfy:

[0053]

[0054] wherein , are preset fixed constants; The initial optimization model is converted into the compliance package optimization model by introducing the value constraint and the price relationship constraint.

[0055] The package period mapping rule specifically is: The basic electricity price peak period is mapped to the package peak period; The basic electricity price peak period is mapped to the package peak period or the flat period; The basic electricity price flat period is mapped to the package peak period, the flat period, or the valley period; The basic electricity price valley period is mapped to the package flat period or the valley period; The basic electricity price deep valley period is mapped to the package valley period.

[0056] Specifically, the following constraints must be met for the electricity sales company to issue a package plan to a customer:

[0057] wherein, the electricity price type for the month is peak, the package electricity price type for the month can only be peak.

[0058]

[0059] wherein, the electricity price type for the month is peak, the package electricity price type for the month can be peak or flat.

[0060]

[0061] wherein, the electricity price type for the month is flat, the package electricity price type for the month can be peak, flat, or valley.

[0062]

[0063] wherein, the electricity price type for the month is valley, the package electricity price type for the month can be flat or valley.

[0064]

[0065] wherein, the electricity price type for the month is deep valley, the package electricity price type for the month can be valley.

[0066]

[0067] wherein, represents the price value of the peak period of the month represents the price value of the flat period of the month represents the price value of the peak period of the month represents the price value of the flat period of the month represents the price value of the peak period of the month represents the price value of the flat period of the month represents the price value of the peak period of the month represents the price value of the flat period of the month

[0068]

[0069] represents the price value of the flat period of the month represents the price value of the flat period of the month

[0070] represents the price value of the flat period of the month represents the price value of the flat period of the month represents the price value of the flat period of the month

[0071] represents the price value of the flat period of the month represents the price value of the flat period of the month

[0072] represents the price value of the flat period of the month represents the price value of the flat period of the month represents the price value of the flat period of the month represents the price value of the flat period of the month

[0073] represents the price value of the flat period of the month represents the price value of the flat period of the month represents the price value of the flat period of the month represents the price value of the flat period of the month

[0074] S5, the smart optimization algorithm is used to solve the compliance contract optimization model, and it is ensured that the contract price obtained by solving is higher than the cost bottom line, and an optimal contract solution is output; the optimal contract solution includes a peak, flat, and valley period division scheme generated for 24 time points per month, and a price corresponding to each period that meets the contract price multiple rule.

[0075] The decision variable of the initial optimization model is the peak, flat, and valley period division of the contract and the price of each period; the smart optimization algorithm is used to solve the compliance contract optimization model, which is to code the decision variable as a chromosome and then iteratively optimize it to find the combination of decision variables that maximizes the objective function value.

[0076] Specifically, the smart optimization algorithm is used to solve the compliance contract optimization model, and it is ensured that the contract price obtained by solving is higher than the cost bottom line, and an optimal contract solution is output, which specifically includes the following steps: S61: select a genetic algorithm as the smart optimization algorithm, and configure the algorithm parameters, including population size, crossover rate, mutation rate, and maximum iteration number; S62: Encode the decision variables into the algorithm's chromosome; wherein, the chromosome genes include discrete codes representing the peak, flat, and valley types at 24 time points each month, and continuous codes representing peak, flat, and valley electricity prices; based on the encoding rules, randomly generate the initial population; this step specifically includes: A chromosome is defined as consisting of time-division segments and valence segments cascaded together; a time-division segment is a sequence of length 24. Each gene locus in the sequence The value of is a discrete integer. =1 indicates that time t is divided into the peak period. =2 indicates the normal segment. =3 indicates a valley period; the electricity price segment is a sequence of length 3. The gene positions in the sequence are continuous real values, representing the peak, flat and off-peak electricity prices of the package, respectively; Based on the defined chromosome structure, perform the following operations to generate an initial population containing a preset number of chromosomes: For time-segmented segments, each gene locus The values ​​are randomly selected uniformly from the set {1,2,3}; For the electricity price segment, the value of each gene locus is uniformly and randomly generated within the preset range of corresponding electricity price values.

[0077] S63: Perform the following operations on each generation of the population: S631: Calculate the profit of each chromosome's corresponding scheme as the initial fitness; verify whether all the value constraints, price relationship constraints, and cost bottom line constraints are satisfied; for chromosomes that violate the constraints, apply fitness penalties according to the degree of violation; In S631, constraint verification and fitness evaluation are implemented through the following steps: S6311: Decode the chromosome into a specific package plan, and calculate the total revenue of the electricity sales company for that plan based on the objective function, using it as the initial fitness. ; S6312: Calculate the degree of violation of each constraint by this chromosome: Time-of-day mapping violation : Traverse the 24 gene loci of the time segment and count their values ​​to see if they meet the value constraints. The number of gene loci; Price multiple violation

[0078] in , , The value decoded from the electricity price segment; Cost bottom line violation degree : if , then , otherwise = 0; S6313: calculate the fitness after punishment according to the formula , wherein is a preset punishment coefficient; the algorithm takes as the final fitness of the chromosome for subsequent evolution operations.

[0079] S632: perform selection, crossover and mutation operations according to the fitness to generate a new generation population; specifically: Selection operation: adopt roulette selection method, calculate the probability of being selected according to the final fitness of each chromosome in the population , and select chromosomes from the current population to form a parent population according to the probability; Crossover operation: randomly pair the chromosomes in the parent population, and for each pair of chromosomes, determine whether to perform crossover according to a preset crossover rate; if so, adopt single-point crossover method, randomly select a crossover point and exchange the chromosome fragments after the point; Mutation operation: for each gene position in the population after the crossover operation, determine whether to perform mutation according to a preset mutation rate; if so, for a gene position in the time period division segment, its value is mutated to another random value in the set ; for a gene position in the electricity price segment, its value is randomly disturbed to a new real value within its preset value range.

[0080] S64: when the maximum number of iterations is reached, select a chromosome that meets all constraints and has the highest original fitness from the historical populations, decode it into an optimal package solution and output. Specifically: Maintain a global optimal solution recorder during the algorithm running process; After each generation iteration is completed, scan the current population, and if a chromosome meets = 0 and = 0 and = 0, it is a feasible solution; compare the original fitness of the feasible solution with the value in the global optimal solution recorder, and if it is better, update the global optimal solution recorder to the chromosome; When the algorithm is terminated due to reaching the maximum number of iterations, decode the chromosome in the global optimal solution recorder into the optimal package solution and output.

[0081] As shown in Figure 2 , the embodiment of the application also provides an optimization system for a power transaction retail package, comprising: The data acquisition module is configured to acquire customers' basic electricity bill data and pre-signed electricity purchase contract data; The cost calculation module is configured to build a cost calculation model for the electricity sales company based on basic electricity cost data and pre-signed electricity purchase contract data to calculate the monthly cost calculation price, and set the monthly cost calculation price as the cost floor for package pricing; The model building module is configured to establish an objective function for optimizing the electricity sales company's revenue, thereby forming an initial optimization model. The constraint loading module is configured to load preset package constraints into the initial optimization model to form the final compliant package optimization model. The intelligent solution module is configured to use an intelligent optimization algorithm to solve the compliant package optimization model and ensure that the solved package price is higher than the cost floor, and output the optimal package solution; the optimal package solution includes a peak, flat and valley time period division scheme for 24 time periods generated each month, and the electricity price corresponding to each time period that conforms to the package price multiple rule.

[0082] In some embodiments, the cost calculation module constructs a cost calculation model for the electricity sales company to calculate the monthly cost calculation price, specifically using the following formula:

[0083] in, For the first moon sky The actual amount of electricity decomposed at that time, For the first moon sky The actual electricity price at that time For the first moon sky Medium and long-term power consumption For the first moon sky Medium- and long-term electricity prices at that time For the first Total number of days in a month Floating prices that acquire costs for businesses; ; ; in, Indicates the first moon The time-of-use weighting of electricity consumption corresponds to the weighting of five electricity price types: peak, peak, flat, valley, and deep valley. For the first Month The time period proportion of the five electricity price types, the total number of time periods is 24, corresponding to peak, peak, flat, valley, and deep valley price types, The total sum of electricity for each of the five electricity price types, The official published decomposition curve proportion value per month per day per hour, And The proportion of the two types of transactions, respectively representing the official authorization proportion and the medium and long term transaction proportion; And Respectively represent the first Month Day Time of the official authorization and medium and long term decomposition electricity price.

[0084] In some embodiments, the model building module is specifically configured to establish the following objective function:

[0085] , ; In the formula, The estimated price of the package in the first Month, The package price in the first Month Time, the price of the same time period each day is the same, The real-time decomposition electricity of the package estimated in the first Month Day Time, The time-sharing proportion, The superimposed fixed price constant, The proportion of spot, Indicates the time-sharing weight of the package electricity in the first Month Time, The total number of electricity price types in the first Month Type, Indicates the total number of three electricity price types in the first Month, corresponding to peak, flat, and valley.

[0086] In some embodiments, the package constraint conditions loaded by the constraint loading module include: Package time period mapping rule: map the five time periods of basic electricity price types, peak, peak, flat, valley, and deep valley, to the package electricity price time period which only allows peak, flat, and valley types; The package price multiple rule is: a first fixed multiple relationship between the flat segment price and the valley segment price and a second fixed multiple relationship between the flat segment price and the peak segment price are set each month; In some embodiments, the constraint loading module is specifically configured to: For the package time period mapping rule: For each time t of the 24 times of a day, a basic electricity price type vector And an allowed package electricity price type vector are defined. The mapping rule is converted into a value constraint for each time t, limiting the decision variable, i.e., the package electricity price type of time t must satisfy ; For the package price multiple rule: The rule is converted into a price relationship constraint, i.e., for any month m, the package flat segment price , the valley segment price , and the peak segment price satisfy:

[0087]

[0088] Wherein, , is a preset fixed constant. By introducing the value constraint and the price relationship constraint, the initial optimization model is converted into the compliance package optimization model.

[0089] In some embodiments, the package time period mapping rule is specifically: The basic electricity price peak time period is mapped to the package peak time period; The basic electricity price peak time period is mapped to the package peak time period or the package flat time period; The basic electricity price flat time period is mapped to the package peak time period, the package flat time period or the package valley time period; The basic electricity price valley time period is mapped to the package flat time period or the package valley time period; The basic electricity price deep valley time period is mapped to the package valley time period.

[0090] In some embodiments, the intelligent solving module specifically includes: An algorithm configuration unit is configured to select a genetic algorithm as the intelligent optimization algorithm and configure the population size, the crossover rate, the mutation rate and the maximum iteration number parameters. The encoding and initializing unit is configured to encode the decision variables into chromosomes and generate an initial population; wherein the chromosome genes include discrete encoding representing the division of 24 time points of each month into peak, flat and valley types, and continuous encoding representing the peak, flat and valley electricity prices; The iterative optimization unit is configured to iteratively optimize the population, and further includes: The fitness evaluation sub-unit is configured to calculate the original fitness of each chromosome, check the degree of violation of each constraint, and impose a fitness penalty to obtain the final fitness; The evolutionary operation sub-unit is configured to perform selection, crossover and mutation operations according to the final fitness to generate a new generation of population; The result output unit is configured to output the optimal package solution decoded from the chromosome with the highest original fitness that satisfies all constraints when the maximum number of iterations is reached.

[0091] The system is equipped with a core processor and corresponding memory, which can quickly and accurately process massive basic data. For example, when collecting electricity bills, the system can quickly classify and calculate electricity consumption and electricity prices. The large-capacity, high-response-speed memory is similar to a data warehouse, which can safely store customer electricity bill information and quickly call data to ensure stable operation and accurate decision-making of the system.

[0092] The system parameter setting is flexible and intelligent, with two modes of intelligent parameters and manually input parameters. The intelligent parameters are automatically generated based on historical data relying on the algorithm model, reflecting market dynamics; the manually input parameters meet special needs, such as inputting specific values during business negotiations, and the system recalculates accordingly to provide personalized services.

[0093] The above description of disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined in this application can be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown in this application, but will conform to the widest scope consistent with the principles and novel features disclosed in this application.

Claims

1. A method for optimizing a retail package of electricity transactions, characterized in that, The method comprises the following steps: obtaining the electricity fee basic data of a customer and the pre-signed electricity purchase contract data; based on the electricity fee basic data and the pre-signed electricity purchase contract data, constructing a cost calculation model of a power selling company to calculate a monthly cost calculation price, and setting the monthly cost calculation price as a cost bottom line of package pricing; establishing a target function of package optimization with the goal of maximizing the revenue of the power selling company, and forming an initial optimization model; loading preset package constraints into the initial optimization model to form a final compliant package optimization model; solving the compliant package optimization model by using an intelligent optimization algorithm, ensuring that the package price obtained by solving is higher than the cost bottom line, and outputting an optimal package solution; the optimal package solution comprises a division scheme of peak, flat and valley time periods for each time of each month generated, and electricity prices corresponding to each time period that meet the package price multiple rule.

2. The method of claim 1, wherein, The cost calculation model of the power selling company is constructed to calculate the monthly cost calculation price, and the specific formula is: Wherein, is the actual decomposition of the electric quantity of the first month day hour, is the actual price of the first month day hour, is the medium and long term electric quantity of the first month day hour, is the medium and long term price of the first month day hour, is the total number of days of the first month, is the floating price of the enterprise acquisition cost; ； ； Wherein, represents the first month time electricity time-sharing weight, corresponding to the weight of five types of electricity price type, peak, peak, flat, valley, deep valley, is the first month time 5 types of electricity price type time period proportion, the total number of time periods is divided into 24 time periods, respectively corresponding to the peak, peak, flat, valley, deep valley price type, is the total of 5 types of electricity price type, is the official announcement of the decomposition of each month each day each hour decomposition curve proportion value, and is the proportion of two kinds of transaction type, respectively represents the official authorized proportion and the proportion of medium and long term transaction; and respectively represent the official authorization and the first month day time of the existing electricity bill decomposition price.

3. The method of claim 2, wherein, The target function of package optimization is: ， ； In the formula, is the estimated price of the package for the month, is the package price for the month at the same time period each day, is the real-time decomposition power of the package for the month day at the time, is the time-sharing ratio, is the superimposed fixed price constant, is the spot ratio, represents the time-sharing weight of the package power for the month at the time, is the real-time decomposition power of the package for the month type of price type total, represents the total number of three types of price types contained in the month, respectively corresponding to peak, flat, and valley.

4. The method of claim 3, wherein, The package constraints include: The package time period mapping rule: mapping the five time periods of basic electricity price type, namely, peak, peak, flat, valley and deep valley, to the three types of package electricity price time periods, namely, peak, flat and valley; The package price multiple rule: setting a first fixed multiple relationship between the flat period price and the valley period price of each month, and a second fixed multiple relationship between the flat period price and the peak period price.

5. The method of claim 4, wherein, The preset package constraints are loaded into the initial optimization model, specifically including the following steps: For the package time period mapping rule: For each of the 24 time instants of the day t, a base tariff type vector and an allowed package tariff type vector are defined The mapping rule is converted into a value constraint of each time t, and a decision variable, i.e. a package electricity price type of the time t , must satisfy ; For the package price multiple rule: transforming the rules into price relationship constraints, i.e. for any one month m, the flat segment price , the valley segment price , the peak segment price satisfy: wherein, , is a preset fixed constant; The initial optimization model is converted into the compliant package optimization model by introducing value constraints and price relationship constraints.

6. The method of claim 5, wherein, The package time period mapping rule is specifically: The basic electricity price peak time period is mapped to the package peak time period; The basic electricity price peak time period is mapped to the package peak time period or the flat time period; The basic electricity price flat time period is mapped to the package peak time period, the flat time period or the valley time period; The basic electricity price valley time period is mapped to the package flat time period or the valley time period; The basic electricity price deep valley time period is mapped to the package valley time period.

7. The method of claim 6, wherein, The decision variables of the initial optimization model are the peak, flat and valley time period division of the package and the electricity price of each time period; the intelligent optimization algorithm is used to solve the compliant package optimization model, which is to encode the decision variables into chromosomes and then iteratively optimize them to find the combination of decision variables that maximizes the value of the target function.

8. The method of claim 7, wherein, The intelligent optimization algorithm is used to solve the compliant package optimization model, and it is ensured that the package price obtained by solving is higher than the cost bottom line, and the optimal package solution is output, specifically including the following steps: S61: select the genetic algorithm as the intelligent optimization algorithm, and configure the algorithm parameters, including population size, crossover rate, mutation rate and maximum iteration number; S62: encode the decision variables into chromosomes of the algorithm; wherein the chromosome genes include discrete codes representing the peak, flat and valley type division of each time of each month, and continuous codes representing the peak, flat and valley electricity price; based on the coding rule, an initial population is randomly generated; S63: perform the following operations on each generation of population: S631: Calculate the yield of each chromosome corresponding scheme as the original fitness; check whether to meet all the value constraints, price relationship constraints and cost bottom line constraints; for the chromosome that violates the constraints, apply fitness penalty according to the degree of violation; S632: According to the fitness, selection, crossover and mutation operations are carried out to generate a new generation of population; S64: When the maximum number of iterations is reached, select the chromosome with the highest original fitness from the population of all generations that meets all the constraints, and decode it into the optimal package solution and output.

9. The method of claim 8, wherein, S62 includes: A chromosome is defined as consisting of time-division segments and valence segments concatenated together; a time-division segment is a sequence of length 24. Each gene locus in the sequence The value of is a discrete integer. =1 indicates that time t is divided into the peak period. =2 indicates the normal segment. =3 indicates a valley period; the electricity price segment is a sequence of length 3. The gene positions in the sequence are continuous real values, representing the peak, flat and off-peak electricity prices of the package, respectively; Based on the defined structure of the chromosome, the following operations are performed to generate an initial population containing a preset number of chromosomes: For the time period division segment, the value of each gene position is randomly selected from the set {1, 2, 3} uniformly; For each gene site, the value is randomly generated within the preset corresponding electricity price value range.

10. An optimization system for retail packages of electricity transactions, characterized by, It includes: A data acquisition module configured to acquire the customer's electricity fee basic data and the pre-signed electricity purchase contract data; A cost estimation module configured to construct a cost estimation model of the power selling company based on the electricity fee basic data and the pre-signed electricity purchase contract data to calculate the monthly cost estimation price, and set the monthly cost estimation price as the cost bottom line of the package pricing; A model construction module configured to establish a target function of package optimization with the goal of maximizing the revenue of the power selling company, forming an initial optimization model; A constraint loading module configured to load the preset package constraint conditions into the initial optimization model to form a final compliant package optimization model; An intelligent solving module configured to solve the compliant package optimization model using an intelligent optimization algorithm and ensure that the solved package price is higher than the cost bottom line, and output the optimal package solution; the optimal package solution includes the division scheme of 24 time points of peak, flat and valley periods generated for each month, and the electricity price corresponding to each period that meets the package price multiple rule.

Images