Buryer medium and long term contract electric quantity decomposition optimization method based on information gap decision theory

Abstract

The invention is suitable for the technical field of power transaction, and provides a buyer medium-and-long-term contract power decomposition optimization method based on an information gap decision theory, and the method comprises the steps: obtaining a predicted spot time-of-use electricity price and a predicted load baseline based on the obtained predicted spot time-of-use electricity price; defining a time-of-use electricity price disturbance item and a load demand disturbance item as uncertain variable reference values of an information gap decision theory; adopting an entropy weight method to calculate weights of a time-of-use electricity price disturbance item and a load demand disturbance item for the obtained historical typical daily spot time-of-use electricity price, the historical load baseline and the historical meteorological data, and distributing a maximum prediction deviation threshold value set by a decision maker to each disturbance item according to the weights to establish a fluctuation range; taking the minimization of the total electricity purchase cost as a target function, introducing a risk preference coefficient to generate a robust constraint condition, and establishing a load response constraint and an electric quantity supply and demand balance constraint; solving the objective function by adopting a multivariate universe optimization algorithm and outputting an optimal contract electric quantity decomposition value of each time period; and the power utilization cost of the user is effectively reduced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power transaction, in particular to a buyer medium and long-term contract power decomposition optimization method based on information gap decision theory. BACKGROUND

[0002] At present, in the process of power market reform, a relatively sound multi-level power spot market has been formed. In order to ensure stable supply, the market adopts a combination mode of medium and long-term contracts and spot market; among them, the medium and long-term contract is the ballast of power supply and price, has the characteristics of physical futures, and is used to confirm the future power supply and demand rights and responsibilities; the spot market is a supplementary way to fulfill the power supply and demand rights and responsibilities in real time within a day. Although the two are a unified whole, they have different operation cycles, and the medium and long-term contract power needs to be decomposed to the same time cycle as the intraday market in the day-ahead stage to reduce transaction costs and improve social welfare, so power decomposition has become a necessary pre-process for market participants.

[0003] In the prior art, long-term contract power decomposition includes typical day power checking, load characteristic or capacity proportion allocation, power balance checking, etc., and has the problems of huge workload or lack of fairness. At the same time, for the optimization of time-of-use electricity price and load demand uncertainty, if the path of improving prediction accuracy is adopted, a large amount of technical development investment and advanced prediction tools are needed, and the true probability distribution of load and time-of-use electricity price needs to be mastered, which is difficult to implement.

[0004] In view of this, a buyer medium and long-term contract power decomposition optimization method based on information gap decision theory is proposed. SUMMARY

[0005] The present application provides a buyer medium and long-term contract power decomposition optimization method based on information gap decision theory, which is used to solve the problems of large workload or lack of fairness in existing power decomposition.

[0006] The first aspect of the present application provides a buyer medium and long-term contract power decomposition optimization method based on information gap decision theory, comprising: obtaining historical typical day spot time-of-use electricity price, historical load baseline and historical meteorological data, and predicted spot time-of-use electricity price, predicted load baseline and predicted meteorological data, and receiving user input of total medium and long-term contract power and contract price; based on the predicted spot time-of-use electricity price and the predicted load baseline, defining the time-of-use electricity price disturbance term and the load demand disturbance term as the uncertainty variable reference value of the information gap decision theory; using entropy weight method to calculate the weight of the time-of-use electricity price disturbance term and the load demand disturbance term based on the historical typical day spot time-of-use electricity price, the historical load baseline and the historical meteorological data, and according to the weight, distributing the maximum prediction deviation threshold set by the decision maker to each disturbance term to establish a fluctuation range; A risk preference coefficient is introduced to generate a robust constraint to limit the cost upper bound in the worst uncertainty scenario, and a load response constraint containing demand cross elasticity coefficient and an electricity supply and demand balance constraint are established. A multiverse optimization algorithm is used to solve the objective function and output the optimal contract electricity decomposition value of each period.

[0007] Further, based on the predicted spot time-of-use electricity price and the predicted load baseline, the time-of-use electricity price disturbance term and the load demand disturbance term are defined as the benchmark values of the uncertain variables of the information gap decision theory, including: The predicted spot time-of-use electricity price is taken as the benchmark value of the time-of-use electricity price disturbance term; The load demand value of the predicted load baseline after meteorological data correction is taken as the benchmark value of the load demand disturbance term; Based on the benchmark values, an uncertainty set of the information gap decision theory is constructed.

[0008] Further, the entropy weight method is used to calculate the weights of the time-of-use electricity price disturbance term and the load demand disturbance term based on the historical typical day spot time-of-use electricity price, the historical load baseline and the historical meteorological data, including: The historical typical day spot time-of-use electricity price, the historical load baseline and the historical meteorological data are respectively standardized, and the sample frequencies of the standardized data are calculated; Based on the sample frequencies, the information entropy of each uncertain variable is calculated; According to the information entropy, the weights of the time-of-use electricity price disturbance term and the load demand disturbance term are determined.

[0009] Further, the maximum prediction deviation threshold set by the decision maker is allocated to each disturbance term according to the weights to establish the fluctuation range, including: The maximum prediction deviation threshold is allocated to the time-of-use electricity price disturbance term and the load demand disturbance term according to the weight proportion, to obtain the time-of-use electricity price fluctuation threshold and the load demand fluctuation threshold; The fluctuation range of the time-of-use electricity price disturbance term is established as the benchmark value plus or minus the time-of-use electricity price fluctuation threshold; The fluctuation range of the load demand disturbance term is established as the benchmark value plus or minus the load demand fluctuation threshold.

[0010] Further, the expression of the uncertainty set is: Wherein: is the uncertainty set based on the information gap decision theory, is the benchmark value set of the uncertain variables, is the acceptable prediction error threshold according to the decision purpose, is the th uncertain scenario, and Time-of-use electricity price disturbance variables during different time periods For the first In the uncertain scenario, the first Time-of-use electricity price forecast benchmark value for the period The entropy weight is the time-of-use electricity price. For the first Load demand disturbance variables during different time periods For the first Baseline value for load demand forecasting for a given period The load demand entropy weight is used.

[0011] Furthermore, the formula for calculating the objective function is as follows: in: Indicates the contract fulfillment period; This represents a basic time period in the spot market, typically one hour. This represents the buyer's total electricity purchase cost over the contract period; This represents the winning bid price for medium- to long-term contracts. Assuming segmented pricing is not considered, then... It is a constant; This indicates that the total amount of medium- and long-term winning bids will be broken down into the amount of electricity purchased in each time period; Indicating that in the previous stage... Spot price forecast for the specified period; Indicates that the buyer is Electricity purchased from the spot market during a given period.

[0012] Furthermore, the robust constraints include: The maximum electricity purchase cost under uncertain scenarios is limited to no more than the minimum cost under risk-free scenarios. times; Robust constraints are expressed as follows: in: This represents the cost function form after the decision-maker considers decision variables and uncertain variables. Represents decision variables, Represents an uncertain variable. This represents the risk preference coefficient of decision-makers. To minimize costs in risk-free scenarios, , This represents the maximum positive deviation of the spot electricity price from the benchmark value under uncertain scenarios.

[0013] Furthermore, the load-related constraints are expressed as follows: wherein: is the user real-time load demand considering the price cross-period effect, represents the reference load demand level given by the user load baseline; represents the demand cross-elasticity coefficient between the time period and the time period, i.e. the change in the demand of the time period caused by the change in the price difference between the time period and the time period per unit the demand change of the time period caused by the user postponing the demand of the time period to the time period caused by the change in the price difference between the time period and the time period; the demand change of the time period caused by the user postponing the demand of the time period to the time period caused by the change in the price difference between the time period and the time period; the demand change of the time period caused by the user postponing the demand of the time period to the time period caused by the change in the price difference between the time period and the time period; the demand change of the time period caused by the user postponing the demand of the time period to the time period caused by the change in the price difference between the time period and the time period; the demand change of the time period caused by the user postponing the demand of the time period to the time period caused by the change in the price difference between the time period and the time period; is the time-of-use price disturbance variable of the time period under the th uncertainty scenario, is the time-of-use price disturbance variable of the time period under the th uncertainty scenario, is the time-of-use price disturbance variable of the time period under the th uncertainty scenario, , is the maximum deviation of the load demand caused by other factors except the price.

[0014] Further, the multi-universe optimization algorithm is used to solve the objective function and output the optimal contract power decomposition value of each time period, comprising: defining a multi-universe search space matrix, wherein the rows represent the solution space, the columns represent the types of variables to be optimized, and the matrix elements represent the candidate solutions of each variable; when the random number is not less than the standard inflation rate of the current solution space, the original solution is retained, otherwise it is replaced by the corresponding variable solution of other solution space through roulette selection operation; based on the current optimal solution space, the variable values of the optimal solution space are updated to the target solution space after adding random disturbance items; dynamically adjusting the wormhole transmission parameters, so that the wormhole generation probability decreases with the increase of the iteration step number, and the transmission rate increases with the increase of the iteration step number; iteratively performing the screening update and wormhole transmission operation until convergence, and outputting the optimal contract power decomposition value meeting the constraint condition.

[0015] The second aspect of the application provides a buyer medium and long-term contract power decomposition optimization system based on information gap decision theory, comprising: a data acquisition unit for acquiring historical typical day spot time-of-use electricity price, historical load baseline and historical weather data, and predicting spot time-of-use electricity price, predicting load baseline and predicting weather data, and receiving user input medium and long-term contract total power and agreement price; an uncertainty variable reference value determination unit for defining the time-of-use electricity price disturbance item and the load demand disturbance item as the uncertainty variable reference value of the information gap decision theory based on the predicted spot time-of-use electricity price and the predicted load baseline; The disturbance item fluctuation range determination unit is configured to calculate the weights of the time-of-use electricity price disturbance item and the load demand disturbance item by using an entropy weight method based on historical typical day spot time-of-use electricity prices, historical load baselines and historical weather data, and to distribute a maximum prediction deviation threshold set by a decision maker to each disturbance item to establish a fluctuation range according to the weights. The objective function and constraint condition determination unit is configured to take minimizing the total electricity purchase cost as an objective function, introduce a risk preference coefficient to generate a robust constraint condition to limit the cost upper bound in a worst uncertainty scenario, establish a load response constraint containing a demand cross-elasticity coefficient and an electricity supply and demand balance constraint. The optimal contract electricity quantity decomposition value output unit is configured to solve the objective function by using a multiverse optimization algorithm and output optimal contract electricity quantity decomposition values in each time period.

[0016] As can be seen from the above technical solutions, the present application has the following advantages: Based on the historical typical day data and prediction data and the medium and long-term contract electricity quantity and contract price input by a user, the present application uses information gap decision theory to depict the uncertainty of time-of-use electricity prices and load demand, uses an entropy weight method to calculate the weights of time-of-use electricity price disturbance items and load demand disturbance items, and distributes a maximum prediction deviation threshold set by a decision maker to each disturbance item to establish a fluctuation range according to the weights. The present application takes robust optimization of minimizing the electricity purchase cost of a buyer and controlling the risk of a user under a given maximum error as an objective, constructs a medium and long-term contract electricity quantity decomposition model of the buyer, and gives an optimal electricity quantity decomposition method of the buyer by solving the model. The present application can make the cost control decision clear and operable, avoid huge investment and workload caused by pursuing prediction accuracy, reduce the electricity cost of a user, and thus improve the overall welfare of society. BRIEF DESCRIPTION OF DRAWINGS

[0017] Figure 1 FIG. 1 is a flowchart of an embodiment of a buyer medium and long-term contract electricity quantity decomposition optimization method based on information gap decision theory in the present application; Figure 2 FIG. 2 is a flowchart of another embodiment of a buyer medium and long-term contract electricity quantity decomposition optimization method based on information gap decision theory in the present application; Figure 3 FIG. 3 is a flowchart of another embodiment of a buyer medium and long-term contract electricity quantity decomposition optimization method based on information gap decision theory in the present application; Figure 4 FIG. 4 is a flowchart of another embodiment of a buyer medium and long-term contract electricity quantity decomposition optimization method based on information gap decision theory in the present application; Figure 5 FIG. 5 is a flowchart of another embodiment of a buyer medium and long-term contract electricity quantity decomposition optimization method based on information gap decision theory in the present application. DETAILED DESCRIPTION

[0018] The terms "first", "second", "third", "fourth" and the like in the description of this application and in the claims, if any, are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the use of the terms so termed herein is solely for descriptive purposes and not for pronouncing the limitations of the application described except as set forth in the appended claims. Moreover, the terms "comprise", "comprising", "corresponding" and "corresponds" and any variations thereof are intended to cover a non-exclusive inclusion, for example, a process, method, system, product or apparatus that comprises a list of steps or units can not necessarily be limited to those steps or units which are expressly listed or to method combinations consisting only of the steps or units but can include additional steps or units, neither of which are expressly listed, or additional method combinations. In addition, elements herein described as comprising one feature can comprise more than one of such feature.

[0019] Embodiment one The method implemented in this embodiment can be implemented in a system, which can be implemented in a server or in a terminal, and the specific implementation is not limited. From the perspective of system implementation, the method for buying long-term contract electricity decomposition optimization based on information gap decision theory in this application will be introduced. Please refer to Figures 1 to 5 The method provided by the embodiment of the application comprises the following steps: S1. Obtain historical typical day spot hourly price, historical load baseline and historical weather data, and predicted spot hourly price, predicted load baseline and predicted weather data, and receive the total electricity quantity of the long-term contract and the contract price input by the user; This step provides a data basis for subsequent quantification of uncertainty by obtaining historical typical day spot hourly price, historical load baseline and historical weather data. Among them, the historical spot price is used to calculate the information entropy of the spot price fluctuation, the historical load baseline is combined with the historical weather data, and the information entropy and weight of the uncertain variable are solved through standardization, frequency statistics and other steps; the predicted spot hourly price, the predicted load baseline and the predicted weather data are used to define the central range of uncertainty and construct the benchmark decision scenario; the total electricity quantity of the long-term contract and the contract price input by the user are received, wherein the total electricity quantity is used as the total quantity constraint of the decomposition electricity, and the contract price is used as the calculation coefficient of the long-term electricity purchase cost, to ensure that the model meets the actual contract terms.

[0020] S2. Based on the predicted spot hourly price and the predicted load baseline, define the hourly price disturbance term and the load demand disturbance term as the benchmark value of the uncertain variable of the information gap decision theory; This step aims to define the benchmark and fluctuation range framework for the uncertain variables in the introduced information gap decision theory model. By extracting the core components from the forecast data and mapping them to the nominal values ​​of uncertain variables, it provides a premise for subsequent calculation of information entropy weights based on historical data, quantifying the risk boundaries of time-of-use electricity price fluctuations and load demand fluctuations. This is specifically achieved through the following steps: S211. Use the predicted spot time-of-use electricity price as the benchmark value for the time-of-use electricity price disturbance term; Specifically, taking the previous stage for Spot price forecast for the period As a central reference for the uncertainty of electricity prices during this period, the information entropy weight is subsequently calculated using historical data. With decision threshold Then, its fluctuation range was derived.

[0021] S212. Use the load demand value after the forecast load baseline is corrected by meteorological data as the benchmark value for the load demand disturbance term; By correcting the load baseline using forecasted meteorological data, a load deviation benchmark is obtained when non-electricity price factors are at the forecast level. Subsequently, information entropy weights were used. and , and construct its fluctuation range.

[0022] S213. Constructing the uncertainty set of information gap decision theory based on benchmark values.

[0023] The expression for an indeterminate set is: in: For the set of uncertainties based on the information gap decision theory, For the set of baseline values ​​for uncertain variables, To determine the acceptable prediction error threshold based on the decision-making objective, For the first In the uncertain scenario, the first Time-of-use electricity price disturbance variables during different time periods For the first In the uncertain scenario, the first Time-of-use electricity price forecast benchmark value for the period The entropy weight is the time-of-use electricity price. For the first Load demand disturbance variables during different time periods For the first Baseline value for load demand forecasting for a given period The load demand entropy weight is used.

[0024] S3. The entropy weight method is used to calculate the weights of the time-of-use electricity price disturbance item and the load demand disturbance item based on the historical typical day spot time-of-use electricity price, historical load baseline and historical meteorological data. According to the weights, the maximum forecast deviation threshold set by the decision-maker is allocated to each disturbance item to establish the fluctuation range. The weights are calculated through the following steps: S311. Standardize the historical typical day spot time-of-use electricity price, historical load baseline and historical meteorological data respectively, and calculate the sample frequency of each data after standardization; S312. Calculate the information entropy of each uncertain variable based on the sample frequency; S313. Determine the weights of the time-of-use electricity price disturbance term and the load demand disturbance term based on information entropy.

[0025] The set of uncertainties as shown above For random variables, take a total length of Typical daily forecast data, denoted as ,in Date markers indicating typical days; Indicates a typical day The time period is defined; the typical daily forecast sequence is standardized as follows: Calculate the frequency of a single typical daily data point in the overall sample: Uncertain variables The information entropy is: The weight of a single uncertain variable can be calculated using the following formula: Finally, the maximum allowable prediction deviation of the uncertain variable can be obtained as follows: .

[0026] The maximum prediction deviation threshold set by the decision-maker is allocated to each disturbance term according to the weights to establish the fluctuation range, which is achieved through the following steps: S321. The maximum prediction deviation threshold is allocated to the time-of-use electricity price disturbance item and the load demand disturbance item according to the weight ratio to obtain the time-of-use electricity price fluctuation threshold and the load demand fluctuation threshold; S322. Establish the fluctuation range of the time-of-use electricity price disturbance term as the benchmark value plus or minus the time-of-use electricity price fluctuation threshold; S323. Establish the fluctuation range of the load demand disturbance term as the benchmark value plus or minus the load demand fluctuation threshold.

[0027] Specifically, S321 is based on the time-of-use electricity price entropy weight calculated by S313. and load demand entropy weight The maximum prediction deviation threshold set by the decision-maker will be used. Allocated according to weighted proportions; among which, the fluctuation threshold of the time-of-use electricity price disturbance term is... The fluctuation threshold of the load demand disturbance term is ; S322 the time-of-use price reference value defined in S211 centered, establish its fluctuation range , quantify the boundary of price uncertainty; S323 the load demand reference value defined in S212 centered, establish its fluctuation range , thus get the expression of the uncertainty set corresponding to S213, explicitly the limit scenario of load disturbance.

[0028] S4. Take the minimum total electricity purchase cost as the objective function, introduce the risk preference coefficient to generate robust constraint conditions to limit the cost upper bound under the worst uncertainty scenario, establish the load response constraint containing the demand cross elasticity coefficient and the power supply and demand balance constraint; Without considering uncertainty for the time being, the buyer determines the winning electricity and winning price in the medium and long-term market, then in the time scale consistent with the spot market, the decomposition electricity of each period per day during the contract performance period needs to be determined to make decisions, so as to minimize the electricity cost in the whole period, therefore the objective function can be expressed as: Among them: denotes the contract performance period; denotes a basic time period of the spot market, taking one hour; denotes the total electricity purchase cost of the buyer in the contract period; denotes the winning price of the medium and long-term contract, assuming that the segmented price is not considered, then is a constant; denotes the decomposition of the winning total amount of the medium and long-term contract to the electricity purchase amount of each period; denotes the spot price forecast of the period in the day-ahead stage; denotes the electricity purchase amount of the buyer from the spot market in the period.

[0029] Since the winning total amount of the medium and long-term contract has been determined, the decomposition electricity and the total amount have the following constraint relationship: Among them, denotes the total electricity purchase amount of the buyer in the performance period of the medium and long-term contract which has been determined.

[0030] As mentioned earlier, in order to increase the scope of application, it is assumed that the purpose of the buyer's electricity purchase includes both self-use and power supply to individual users, therefore the load demand needs to be included in the model. Since the electricity price will have an impact on the user demand, this impact mainly reflects in the cross-period planning of user load level, therefore the relationship between user load demand and time-of-use price can be expressed in the following form: For the user real-time load demand after considering the price cross-period impact, represents the reference load demand level given by the user load baseline; represents the demand cross elasticity coefficient between the period and the period, i.e., the demand change of the user caused by the price difference change of the period and the period; the demand change of the period caused by the period demand postponing to the period.

[0031] In the contract performance period, the electricity purchase amount of the buyer in the medium and long-term market is decomposed to the intra-day period, and the sum of the electricity purchase amount in the spot market should meet the user or own load demand, so there are the following constraint conditions: In the above model, the uncertainty of the variable is not considered, and in the case where there is no uncertainty, the objective function is defined as: The solution of the above formula is , which is the minimum cost item obtained without introducing uncertainty.

[0032] The variables with uncertainty in the model include the time-of-use electricity price in the spot market and the load demand of the user, wherein the time-of-use electricity price itself can be a random variable, and for the load demand of the user, since contains the time-of-use electricity price, a new random variable can be added here, and it is updated as: Based on the above analysis, after introducing uncertainty, the decision becomes a risk decision, and the risk decision is divided into risk-averse type, risk-neutral type and risk-preference type according to the attitude of the decision maker to the risk premium. As described before, since the electricity resources are difficult to store and the electricity market is difficult to speculate, generally, market participants are risk-averse type, therefore, risk aversion is adopted as the decision style of the decision maker.

[0033] Then, under the condition of risk aversion, the decision maker does not pursue excess expected return, but pursues a minimum return level under the condition of maximum controllable risk exposure, at this time, its utility function can be expressed as: In the above formula, represents the robust function based on utility; represents the risk preference utility of the decision maker; represents the utility after considering uncertainty; the meaning of the above formula is that, under the condition of risk aversion, the goal of the decision maker is to pursue the expected utility not less than the risk preference utility under the risk threshold (at this time the prediction error reaches the maximum), and the risk preference utility The expression of the risk preference utility is: wherein, represents the risk preference coefficient of the decision maker, which can be obtained by questionnaire survey or other risk assessment tools; represents the expected utility of the decision maker under the condition of no risk.

[0034] The maximum electricity purchase cost under the uncertainty scenario is limited to not more than times of the minimum cost under the no-risk scenario; The robust constraint condition is represented as: wherein, represents the cost function form of the decision maker after considering the decision variable and the uncertain variable; represents the decision variable, represents the uncertain variable, represents the risk preference coefficient of the decision maker, is the minimum cost under the no-risk scenario, , is the maximum positive deviation of the spot electricity price from the benchmark value under the uncertainty scenario.

[0035] In the above formula, is the constraint condition of the robust function, i.e. the maximum prediction deviation acceptable by the decision maker, which depends on the risk preference of the decision maker and is a prior parameter. When the uncertain parameters in the above formula exceed one, the maximum prediction deviation of each parameter is not a prior parameter and is obtained by the entropy weight method, which is not described herein.

[0036] S5. The multiverse optimization algorithm is used to solve the objective function and output the optimal contract electricity decomposition value of each period.

[0037] This step is realized through the following sub-steps: S511. Define a multiverse search space matrix, wherein the rows represent the solution space, the columns represent the types of variables to be optimized, and the matrix elements represent the candidate solutions of each variable; S512. When the random number is not less than the standard inflation rate of the current solution space, the original solution is retained, otherwise the corresponding variable solution of other solution space is selected by roulette operation to replace the current solution; S513. Based on the current optimal solution space, the variable value of the optimal solution space is updated to the target solution space after adding a random disturbance term according to the wormhole generation probability and transmission rate parameters; S514. Dynamically adjust the wormhole transmission parameters, so that the wormhole generation probability decreases with the increase of the iteration step number, and the transmission rate increases with the increase of the iteration step number; S515. Iteratively perform the screening update and wormhole transmission operation until convergence, and output the optimal contract power decomposition value that meets the constraint condition.

[0038] Specifically, the multi-element search space is located as: Wherein, the element represents the jth solution of the ith variable; the row in the matrix represents the space where the solution is located, and the column represents the type of variable to be solved; represents the total number of variables to be solved; represents the total number of solution spaces; the above matrix is actually a set of alternative solutions. For these alternative solutions, the following properties are also possessed:

[0039] Wherein, represents a random number between [0, 1] generated by the program; represents the standard inflation rate of the ith space, which is a parameter set when the algorithm is deployed; represents the jth solution in the ith solution space selected by roulette;

[0040] Different spaces will transmit matter through a wormhole-like mechanism, and according to their respective standard inflation rates, the space with lower inflation rate is more likely to transmit matter to other spaces. This mechanism is actually a roulette solution search process. Assuming that there is currently an optimal solution space, the wormhole is established between the optimal space and other spaces, and this iterative process can be represented as: In the above formula, represents the jth variable in the current optimal space; represents the upper bound value of the jth variable; represents the lower bound value of the jth variable; , , are also random numbers in the interval [0, 1] generated by the program;​​​​​ and are system parameters, respectively, wormhole generation probability and transmission rate, and their expressions are: wherein, denotes the current iteration step number; denotes the maximum iteration step number; denotes the search approximation speed changing with the iteration step number; denotes the initial transmission rate of each space; denotes the maximum value in the initial transmission rate sequence of each space; denotes the minimum value in the initial transmission rate sequence of each space.

[0041] Through the above steps, the decision variables , , the control variables , , the objective function and the constraint condition are deployed into the above algorithm, and when the algorithm converges, the optimal power decomposition of each period in the given medium and long-term contract power and protocol period can be obtained.

[0042] The above embodiment sets the decision as a robust condition of risk aversion by constructing a robust distribution model of the decomposition of the market participant medium and long-term contract power, and quantitatively converts the decision of the decision maker to pursue cost minimization under uncertainty into a controllable cost target under a given risk preference. The decision of cost control can be clear and operable, avoiding the huge investment and workload brought by pursuing prediction accuracy, and can reduce the electricity cost of users, thereby improving the overall welfare of the society.

[0043] Embodiment two An embodiment of a buyer medium and long-term contract power decomposition optimization system based on information gap decision theory in the application includes the following steps: A data acquisition unit is configured to acquire historical typical day spot time-of-use electricity price, historical load baseline and historical weather data, and predicted spot time-of-use electricity price, predicted load baseline and predicted weather data, and receive user inputted total medium and long-term contract power and contract price; An uncertain variable reference value determination unit is configured to define time-of-use electricity price disturbance items and load demand disturbance items as uncertain variable reference values of the information gap decision theory based on the predicted spot time-of-use electricity price and the predicted load baseline; A disturbance item fluctuation range determination unit is configured to calculate weights of the time-of-use electricity price disturbance items and the load demand disturbance items by using an entropy weight method on the historical typical day spot time-of-use electricity price, the historical load baseline and the historical weather data, and distribute a maximum prediction deviation threshold set by a decision maker to each disturbance item to establish a fluctuation range according to the weights; The target function and constraint condition determining unit takes minimizing total power purchase cost as a target function, introduces a risk preference coefficient to generate a robust constraint condition to limit the cost upper bound in the worst uncertainty scenario, and establishes load response constraints containing demand cross-elasticity coefficients and power supply and demand balance constraints. The optimal contract power decomposition value output unit is configured to solve the target function by using a multi-universe optimization algorithm and output optimal contract power decomposition values in each time period.

[0044] The specific limitations of the system can be referred to the limitations of the method in the above, which will not be repeated here. Each module in the above system can be realized by software, hardware and their combinations in whole or in part. The above modules can be embedded in or independent of the processor in the computer device in hardware form, or stored in the memory in the computer device in software form, so as to be called and executed by the processor to perform the operations corresponding to each module. In addition, each model and function mentioned in the above embodiment one will be integrated in the intelligent power contract decomposition optimization system proposed in the present application.

[0045] It can be understood that those skilled in the art can combine various embodiments in the above embodiments under the guidance of the above embodiments to obtain various embodiments of the technical solutions.

[0046] The above only describes the preferred embodiments of the present application and is not intended to limit the present application. Any modification, equivalent replacement and improvement made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for long-term contract power decomposition optimization based on information gap decision theory, characterized in that, The method comprises the following steps: Obtain historical typical day spot time-of-use electricity price, historical load baseline and historical meteorological data, and predict spot time-of-use electricity price, predicted load baseline and predicted meteorological data, and receive user input of total medium and long-term contract electricity and contract price; Based on the predicted spot time-of-use electricity price and the predicted load baseline, define the time-of-use electricity price perturbation term and the load demand perturbation term as the reference value of the uncertain variable of the information gap decision theory; Use the entropy weight method to calculate the weight of the time-of-use electricity price perturbation term and the load demand perturbation term based on the historical typical day spot time-of-use electricity price, the historical load baseline and the historical meteorological data, and distribute the maximum prediction deviation threshold set by the decision maker to each perturbation term according to the weight to establish the fluctuation range; Take the minimization of the total purchase cost as the objective function, introduce the risk preference coefficient to generate the robust constraint condition to limit the cost upper limit under the worst uncertainty scenario, establish the load response constraint containing the demand cross elasticity coefficient and the power supply and demand balance constraint; Solve the objective function by using the multiverse optimization algorithm and output the optimal contract electricity decomposition value of each period.

2. The buyer medium and long term contract power decomposition optimization method based on information gap decision theory of claim 1, wherein, The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps:

3. The buyer medium to long term contract power decomposition optimization method based on information gap decision theory of claim 1, wherein, The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps:

4. The buyer medium and long term contract power decomposition optimization method based on information gap decision theory of claim 3, wherein, The expression of the uncertain set is: The calculation formula of the objective function is: The robust constraint condition comprises: The robust constraint condition is expressed as:

5. The buyer medium and long term contract electricity decomposition optimization method based on information gap decision theory according to any one of claims 1-4, characterized in that, The load response constraint is expressed as: in: For the set of uncertainties based on the information gap decision theory, For the set of baseline values ​​for uncertain variables, To determine the acceptable prediction error threshold based on the decision-making objective, For the first In the uncertain scenario, the first Time-of-use electricity price disturbance variables during different time periods For the first In the uncertain scenario, the first Time-of-use electricity price forecast benchmark value for the period The entropy weight is the time-of-use electricity price. For the first Load demand disturbance variables during different time periods For the first Baseline value for load demand forecasting for a given period The load demand entropy weight is used.

6. The information gap decision theory based buyer medium and long term contract electric power decomposition optimization method according to claim 1, characterized in that, The method comprises the following steps: in: Indicates the contract fulfillment period; This represents a basic time period in the spot market, typically one hour. This represents the buyer's total electricity purchase cost over the contract period; This represents the winning bid price for medium- to long-term contracts. Assuming segmented pricing is not considered, then... It is a constant; This indicates that the total amount of medium- and long-term winning bids will be broken down into the amount of electricity purchased in each time period; Indicating that in the previous stage... Spot price forecast for the specified period; Indicates that the buyer is Electricity purchased from the spot market during a given period.

7. The buyer medium and long term contract power decomposition optimization method based on information gap decision theory of claim 6, wherein, Define a multiverse search space matrix, wherein the rows represent the solution space, the columns represent the types of variables to be optimized, and the matrix elements represent the candidate solutions of each variable; limiting the maximum electricity purchase cost in an uncertain scenario to not more than the minimum cost in a risk-free scenario times; When the random number is not less than the standard inflation rate of the current solution space, the original solution is retained, otherwise it is replaced by the corresponding variable solution of other solution space through roulette selection operation; wherein: represents the form of the cost function after the decision maker considers the decision variables and the uncertain variables; represents the decision variables, represents the uncertain variables, represents the risk preference coefficient of the decision maker, is the minimum cost for the risk-free scenario, , is the maximum positive deviation of the spot electricity price from the benchmark value under the uncertainty scenario.

8. The buyer medium and long term contract power decomposition optimization method based on information gap decision theory of claim 7, wherein, ​ wherein: is the user real-time load demand considering the price inter-temporal effect, represents the reference load demand level given by the user load baseline; represents the demand cross elasticity coefficient between the and the period, i.e. the change in the period's price difference leads to the user's period demand postponing to the period's demand change; is the time-of-use price perturbation variable of the period under the uncertainty scenario, , is the maximum load demand deviation caused by other factors except for the price.

9. The information gap decision theory based buyer medium to long term contract electric power decomposition optimization method according to claim 1, characterized in that, ​ ​ ​ Based on the current optimal solution space, the variable values of the optimal solution space are updated to the target solution space by superimposing random disturbance terms on the wormhole generation probability and transmission rate parameters; The wormhole transmission parameters are dynamically adjusted so that the wormhole generation probability decreases and the transmission rate increases with the increase of the iteration step number; The screening update and wormhole transmission operations are iteratively performed until convergence, and the optimal contract power decomposition value that meets the constraint condition is output.

10. A buyer medium and long term contract power decomposition optimization system based on information gap decision theory, characterized in that, The method of any one of claims 1-9 is used, comprising: a data acquisition unit for acquiring historical typical day spot time-of-use electricity price, historical load baseline and historical weather data, and predicting spot time-of-use electricity price, predicting load baseline and predicting weather data, and receiving user input of total medium and long-term contract power and contract price; an uncertain variable reference value determination unit for defining time-of-use electricity price disturbance terms and load demand disturbance terms as uncertain variable reference values of information gap decision theory based on the predicted spot time-of-use electricity price and predicted load baseline; a disturbance term fluctuation range determination unit for calculating the weights of time-of-use electricity price disturbance terms and load demand disturbance terms using an entropy weight method on historical typical day spot time-of-use electricity price, historical load baseline and historical weather data, and distributing the maximum prediction deviation threshold set by the decision maker to each disturbance term to establish the fluctuation range; a target function and constraint condition determination unit for taking minimizing total electricity purchase cost as the objective function, introducing a risk preference coefficient to generate a robust constraint condition to limit the cost upper bound in the worst uncertainty scenario, establishing load response constraints containing demand cross-elasticity coefficients and power supply and demand balance constraints; an optimal contract power decomposition value output unit for solving the objective function using a multiverse optimization algorithm and outputting the optimal contract power decomposition value of each period.

Images