CN-122022861-A - Wholesale retail market price conduction optimization method considering risk allocation

CN122022861ACN 122022861 ACN122022861 ACN 122022861ACN-122022861-A

Abstract

The invention discloses a wholesale retail market price conduction optimization method considering risk allocation, which is used for solving the defects of a price conduction mechanism in the existing electric power market in the aspects of risk collaborative quantification and allocation. According to the method, a double-layer optimization model aiming at minimizing electricity purchasing cost and maximizing retail benefit of an electricity selling company is constructed, cost deviation caused by market uncertainty is quantified by introducing condition risk value, and randomness of spot electricity price and user load is processed by adopting Monte Carlo simulation and K-means clustering. And in the model, various conditions such as electricity purchasing balance, real-time market deviation, retail package electricity quantity and electricity price constraint, user electricity consumption behavior, peak-valley difference limitation and the like are simultaneously considered, and a CPLEX solver is utilized for carrying out efficient solving. The invention realizes reasonable allocation of market risk between the electricity selling company and the user, improves the price signal transmission efficiency and the user side response capability, and provides scientific decision support for electricity purchasing strategies and retail pricing of the electricity selling company in the electricity market environment.

Inventors

JIA XING
CHEN MENGYAO
ZHUANG XIAODAN
DENG HUI
HE YANG
CHEN CHENG

Assignees

浙江电力交易中心有限公司

Dates

Publication Date: 20260512
Application Date: 20251203

Claims (10)

1. A wholesale retail market price conduction optimization method considering risk allocation, comprising the steps of: S1, constructing a double-layer optimization model framework for wholesale retail market price conduction: The framework comprises an upper model and a lower model, wherein the upper model takes the total electricity purchasing cost of an electricity selling company as a decision target, and the lower model takes the retail benefit of the electricity selling company as a decision target; S2, processing uncertainty of markets and loads: selecting a typical daily curve from the scene curves by adopting a K-means clustering method, decomposing the middle-long term contract electric quantity based on the typical daily load curve, and obtaining the middle-long term decomposition electric quantity under the typical day; S3, constructing a wholesale retail market price conduction optimization model considering risk allocation: Based on the frame of the step S1 and the data processed in the step S2, a specific double-layer optimized mathematical model is established; The objective function of the upper model is to minimize the total electricity purchasing cost of an electricity selling company, wherein the total electricity purchasing cost comprises middle-long-term contract electricity purchasing cost, daily market electricity purchasing cost and real-time market electricity purchasing cost, and condition risk value CVaR is introduced to carry out risk quantification and adjustment on cost deviation; The objective function of the lower model is to maximize retail benefit of an electricity selling company, and the decision variable is retail package pricing and structure facing to users, wherein constraint conditions of the lower model comprise retail package electric quantity constraint, package electric price constraint, user electricity consumption behavior constraint, user package selection uniqueness constraint and electricity consumption peak-valley difference constraint; S4, solving the double-layer optimization model: And (3) solving the model established in the step (S3) by adopting a mathematical programming solver to obtain an optimal electricity purchasing strategy of an electricity selling company in a wholesale market and an optimal retail package pricing mechanism in a retail market, wherein the electricity purchasing strategy and the pricing mechanism together form a price conduction scheme considering risk allocation.
2. The method according to claim 1, wherein in step S2, the spot market price fluctuation is subjected to a lognormal distribution, the user load randomness is subjected to a normal distribution, and after a random scene is generated through Monte Carlo simulation, scene reduction is performed by using K-means clustering, and uncertainty is represented by a typical daily curve.
3. The method according to claim 1, wherein in the step S3, the medium-long term contract electricity purchase cost of the upper model is calculated based on contract electricity amounts and corresponding electricity prices in peak, flat and valley periods, and the day-ahead market electricity purchase cost and the real-time market electricity purchase cost are calculated based on day-ahead electricity prices, real-time electricity prices and corresponding electricity purchase amounts in typical day scenes, respectively.
4. The method of claim 1, wherein in step S3, the retail packages of the lower model are multi-package combinations comprising basic electricity prices and green electricity prices, and the user cost model aims at minimizing the total electricity expenditure of the user and is linked with the electricity company benefit model.
5. The method according to claim 1, characterized in that in step S3, the introduced conditional risk value CVaR is used to measure the expected loss of the electricity purchase cost of the electricity selling company exceeding the risk value VaR at a given confidence level, and the risk cost is integrated into the upper objective function to achieve a risk-adjusted cost minimization.
6. A wholesale retail market price conduction optimization system that considers risk sharing, comprising: The model frame construction module is used for constructing a double-layer optimization frame comprising an upper-layer cost minimization model and a lower-layer benefit maximization model; The uncertainty processing module is used for generating a random scene through Monte Carlo simulation and extracting typical scene data through K-means clustering; the risk optimization model construction module is used for constructing a specific electricity purchasing cost model and a retail package optimization model in the framework according to the integrated condition risk value CVaR, and configuring corresponding constraint conditions; and the model solving module is used for calling a mathematical programming solver to solve the risk optimization model and outputting an optimal electricity purchasing and pricing strategy.
7. The system of claim 6, wherein the uncertainty processing module is configured to simulate and cluster off-the-shelf electricity prices subject to a lognormal distribution and loads subject to a normal distribution, and output a typical daily curve and corresponding medium-to-long term decomposition power.
8. The system of claim 6, wherein an upper model in the risk optimization model building module comprises a medium-long term contract cost calculation unit, a spot market cost calculation unit, a CVaR risk quantization unit and a purchase electricity balance and deviation constraint unit, a lower model in the risk optimization model building module comprises a retail package benefit calculation unit, a user electricity charge cost calculation unit, a package electric quantity and electricity price constraint unit and a user behavior and peak-valley difference constraint unit, and the model solving module adopts a CPLEX solver to solve the double-layer optimization model and obtain a price conduction mechanism for realizing risk allocation.
9. A non-volatile storage medium, characterized in that the non-volatile storage medium comprises a stored program, wherein the program, when run, controls a device in which the non-volatile storage medium is located to perform the method of any one of claims 1 to 5.
10. A terminal device, characterized in that the terminal device comprises a processor, a memory, a communication interface and a bus, the processor, the memory and the communication interface being connected via the bus and performing communication with each other, the memory storing executable program code, the processor running a program corresponding to the executable program code by reading the executable program code stored in the memory for performing the method according to any of the preceding claims 1-5.

Description

Wholesale retail market price conduction optimization method considering risk allocation Technical Field The invention relates to an optimization method, in particular to a wholesale retail market price conduction optimization method considering risk allocation. Background Under the background of continuous deepening of the power market reform, the price of the power wholesale market is influenced by factors such as supply-demand relation, renewable energy output fluctuation, market structure, main risk preference and the like, and the remarkable uncertainty is presented. The electricity company acts as a hub connecting wholesale and retail markets, and the conduction efficiency between electricity purchase costs and retail pricing directly affects the electricity costs of the end users. With the perfection of the electric power market trading mechanism, electricity selling companies face the realistic challenges of how to design a reasonable price conduction mechanism in a complex market environment, and to realize effective risk allocation and resource allocation optimization. At present, the existing research has been studied to a certain extent on price conduction mechanisms of the electric wholesale-retail market, partial results focus on analysis of modes such as delay conduction, real-time conduction and proportional conduction, and the like, and the characteristics of different mechanisms in the aspects of risk allocation and price signal effectiveness are revealed. However, most of the existing researches start from a single conduction mode, and lack of system integration of collaborative optimization and risk quantification management of different mechanisms in actual operation, so that when an electricity selling company handles severe fluctuation of wholesale market price, a conduction strategy which takes efficiency and robustness into consideration is difficult to form, and particularly under the condition that user load randomness and multiple uncertainty of the market are superposed, reasonable distribution of risks between the electricity selling company and users is difficult to realize in the existing mechanism design. In addition, existing research on cost-effectiveness measurement of electricity-selling companies is mostly based on deterministic models or traditional risk measurement methods, and tail risks in extreme market situations cannot be fully reflected. Partial achievements are introduced into medium-long term and spot market electricity purchasing cost analysis, but when the double uncertainties of electricity price and load are processed, linear prediction or simplifying assumptions are still relied on, and the cost benefit deviation under nonlinear and multi-scene is difficult to accurately describe. How to construct a price conduction and cost optimization model integrating advanced risk measurement tools such as conditional risk values and integrating multi-market and multi-cycle transaction data is a key problem to be broken through in the current research. Disclosure of Invention In order to solve the defects in the prior art, the invention provides a wholesale retail market price conduction optimization method considering risk allocation based on a power wholesale-retail market price conduction mechanism, which introduces CVaR to perform cost deviation measurement and calculation of risk adjustment and constructs an analysis framework covering multidimensional influence factors such as policies, markets, operation and the like, and the technical scheme is as follows: S1, constructing a double-layer optimization model framework for wholesale retail market price conduction: The framework comprises an upper model and a lower model, wherein the upper model takes the total electricity purchasing cost of an electricity selling company as a decision target, and the lower model takes the retail benefit of the electricity selling company as a decision target; S2, processing uncertainty of markets and loads: selecting a typical daily curve from the scene curves by adopting a K-means clustering method, decomposing the middle-long term contract electric quantity based on the typical daily load curve, and obtaining the middle-long term decomposition electric quantity under the typical day; S3, constructing a wholesale retail market price conduction optimization model considering risk allocation: Based on the frame of the step S1 and the data processed in the step S2, a specific double-layer optimized mathematical model is established; The objective function of the upper model is to minimize the total electricity purchasing cost of an electricity selling company, wherein the total electricity purchasing cost comprises middle-long-term contract electricity purchasing cost, daily market electricity purchasing cost and real-time market electricity purchasing cost, and condition risk value CVaR is introduced to carry out risk quantification and adjustment on cost deviation; The objective