CN-121998688-A - Quotation decision-making system and method suitable for sustainable development mechanism

CN121998688ACN 121998688 ACN121998688 ACN 121998688ACN-121998688-A

Abstract

The description provides a quotation decision-making system and a quotation decision-making method suitable for a sustainable development mechanism, wherein a simulation subject of each quotation participant is automatically generated according to information, quotation space constraint and inventory space constraint of each quotation participant, quotation and inventory of each simulation subject are cleared based on an clearing mechanism when each iteration is performed, iteration is performed for a plurality of times until iteration cut-off conditions are reached, each simulation subject provides quotation and inventory when each iteration is performed, after each simulation subject provides quotation and inventory, the clearing simulation unit clears the quotation and inventory of each simulation subject, and quotation is determined according to the income of at least one simulation subject after the clearing is performed for a plurality of times. The scheme provides scientific and reasonable decision references for the main bodies participating in bidding, and can reduce the risk of virtual high-output or ultra-low price loss of the bidding, so that the resource allocation can be optimized, the new energy consumption efficiency is improved, and the price fluctuation of the bidding market is reduced.

Inventors

LIU ZHEN
HU KUN
LIU JUN
Ren Xiaomang
ZHANG ZHIXIN

Assignees

中电建新能源集团股份有限公司

Dates

Publication Date: 20260508
Application Date: 20251231

Claims (15)

1. A bid decision system adapted for use with a sustainable development mechanism, comprising: the data acquisition unit is used for acquiring information of each quotation participant, quotation space constraint of each quotation participant and volume reporting space constraint of each quotation participant; The system comprises a main body generation unit, a main body generation unit and a main body generation unit, wherein the main body generation unit is used for generating a simulation main body of each quotation participant according to the content acquired by the data acquisition unit, and each simulation main body is used for providing quotation and reporting amount in the own quotation space constraint and reporting amount space constraint based on a quotation strategy when each iteration is performed; The clearing simulation unit is used for clearing quotations and amortization of each simulation main body based on a clearing mechanism during each iteration; The system comprises a simulation main body, a driving engine, a clearing simulation unit and a display unit, wherein the simulation main body is used for providing quotation and volume reporting for each simulation main body; and the determining unit is used for determining the quotation according to the income of the at least one simulation subject after the clearing of a plurality of iterations.
2. The system of claim 1, wherein the cost of the power generation project of the target bidding party is known, the data acquisition unit further acquiring a probabilistic estimate of the cost of the power generation project of each bidding party other than the target bidding party; The main body generating unit generates N cost estimated values for the simulation main body of each other quotation participant through random sampling based on probability estimation of the cost of the power generation project, M quotation scenes are formed by combining the N cost estimated values of each other quotation participant, a plurality of quasi-quotation strategies are generated for the main body generating unit, expected values of profits of each quasi-quotation strategy in the M quotation scenes are calculated, expected value of profits corresponding to each quasi-quotation strategy is obtained, the quasi-quotation strategy with the largest expected value of profits is used as a final quotation strategy determined for the target quotation participant in a single iteration, the quotation strategy comprises declared mechanism electricity price and mechanism electricity quantity, and N, M is a natural number.
3. The system of claim 2, wherein the data acquisition unit comprises: The crawling subunit is used for crawling information of each quotation participant, quotation space constraint of each quotation participant and volume reporting space constraint of each quotation participant from the target website; And the evaluation subunit is used for evaluating the probability estimation of the power generation project cost of each quotation participant according to the information of each quotation participant.
4. A system according to claim 3, wherein the evaluation subunit comprises: an artificial intelligence network inputs the installed capacity of the power generation project of the quoting party and at least one of annual energy generation capacity and unit investment cost interval, and outputs probability estimation of the power generation project cost of the quoting party.
5. The system of claim 1, wherein each simulated body is automatically generated based on a template of a preset simulated body, information of each bidding participant, bidding space constraints of each bidding participant, and inventory space constraints of each bidding participant; The templates of the simulation main body comprise identifications of quotation participants, quotation space constraints of all quotation participants, quantitative space constraints of all quotation participants and quotation iteration adjustment strategies.
6. The system of claim 5, wherein the bid iteration adjustment strategy uses a Q-learning algorithm to determine a bid and a bid for a next iteration, and uses the bid and the bid for a single iteration as actions in the Q-learning algorithm and the clear benefit as states in the Q-learning algorithm.
7. The system of claim 6, wherein determining the bid, the bid amount at the next iteration using the Q-learning algorithm comprises: Initializing parameters, creating a Q-table and randomly initializing a state s before the first iteration bidding; The method comprises the following steps of determining quotation and inventory as actions by adopting an epsilon-greedy strategy and executing joint actions when bidding is performed in a previous iteration, calculating the benefits of quotation participants as a reward value r after the clearing simulation unit simulates clearing, updating a new state s 'after joint actions according to the reward value r, updating a Q-table, and updating the state of the current iteration into the new state s' to prepare for the next iteration.
8. The system of claim 7, wherein the revenue for the bidding party is calculated according to the following formula: R i =R i jz +R i zc -C i dd Q i total , Wherein R i represents the income of the quoting party i, R i jz represents the mechanism electric quantity income of the quoting party i, R i zc represents the long-term and spot electric quantity income of the quoting party i, C i dd represents the electricity-measuring cost of the quoting party i, element/MWh, and Q i total represents the total electric quantity generated by the quoting party i in the whole life cycle, MWh; , , Wherein P jz represents a mechanism electricity price, meta/MWh, P t avg represents an average trading electricity price of a spot market in the T-th year, meta/MWh, Q i,t jz represents a decomposition amount of the winning electricity in the T-th year of the mechanism electricity amount of a quoting party i, MWh, P i,t mar represents a trading price of the power spot market of the quoting party i in the T-th year, meta/MWh, P i,t zc represents a medium-long term trading price of the power of the quoting party i in the T-th year, meta/MWh, Q i,t total represents a total electricity generation amount of the quoting party i in the T-th year, MWh, Q i,t zc represents an execution period of the mechanism electricity amount of the quoting party i in the T-th year, MWh, T jz represents a service life cycle of the quoting party i, and T l represents a service life cycle of the quoting party.
9. The system of claim 7, wherein the determining of the bid amount using the epsilon-greedy strategy comprises randomly selecting actions with epsilon probability and selecting joint actions with 1-epsilon probability with the highest Q value in the current Q-table under the state s.
10. The system of claim 1, wherein the clearing mechanism includes determining the selected power generation items in a low to high ranking of offers, and wherein the price of mechanical power is determined in accordance with the highest offer in each of the selected power generation items.
11. The system of claim 1, wherein the clearing mechanism includes distributing the amount of electricity by a proportion of the amount of credit of the plurality of bidding participants in the event that there are a plurality of bidding participants whose bids are equal to the price of electricity by the mechanism.
12. A bid decision method suitable for a sustainable development mechanism, comprising: acquiring information of each quotation participant, quotation space constraint of each quotation participant and volume reporting space constraint of each quotation participant; The method comprises the steps of obtaining content of a user, generating simulation subjects of each quotation participant according to the obtained content, wherein each simulation subject is used for providing quotation and reporting volume in a quotation space constraint of the simulation subjects based on a quotation strategy in each iteration, and the quotation is a mechanism electricity price declared under a sustainable development mechanism and the reporting volume is a mechanism electricity quantity declared under the sustainable development mechanism; Based on the clearing mechanism, clearing quotations and amortization of each simulation subject at each iteration; performing multiple iterations until reaching the iteration cut-off condition, wherein each simulation subject provides quotation and volume reporting, and clearing the quotation and volume reporting of each simulation subject after each simulation subject provides quotation and volume reporting; And determining the quotation according to the benefits of the at least one simulation subject after the clearing of the plurality of iterations.
13. An electronic device, comprising: A memory and a processor in communication with each other, the memory having stored therein computer instructions which, upon execution, implement the quotation decision-making system for a sustainable development mechanism of any one of claims 1-11.
14. A computer storage medium having stored thereon computer program instructions which, when executed, implement the quotation decision system of any one of claims 1-11 adapted for use with a sustainable development mechanism.
15. A computer program product comprising a computer program which, when executed by a processor, implements the quotation decision system of any one of claims 1-11 adapted for sustainable development mechanisms.

Description

Quotation decision-making system and method suitable for sustainable development mechanism Technical Field The application relates to the technical field of electric power energy, in particular to a quotation decision-making system and method suitable for a sustainable development mechanism. Background The new energy full-electric-quantity marketing mechanism requires that the online electric quantity of new energy projects such as wind power, solar power generation and the like enter an electric market in principle, and the online electricity price is formed through market transaction. In order to ensure sustainable development of new energy industry and reduce impact of market price fluctuation on project income, a sustainable development settlement mechanism is established and is used as a buffer guarantee mechanism for transition of new energy from fixed electricity price to complete marketization. Under the operation framework of a sustainable development settlement mechanism, one power generation project only has one opportunity of successful bidding, and each quotation participant (mainly a new energy investment enterprise) needs to declare the mechanism electricity price and the corresponding mechanism electricity quantity in a marketing bidding mode to participate in the allocation competition of the mechanism electricity quantity. The current bidding clearing follows the core principles of low price priority, marginal pricing and total amount control, namely an energy management department firstly checks the total scale of annual mechanism electric quantity according to regional new energy consumption responsibility weights, industrial and commercial user bearing capacity and the like, each new energy investment enterprise submits mechanism electric prices and mechanism electric quantity reporting schemes in combination with self electricity price, market expectation and the like, an electric power trading center sorts the pair reporting items according to the reporting mechanism electric prices from low to high, sequentially accumulates the mechanism electric quantity until the annual total scale is reached, the highest quotation of the selected items is the unified mechanism electric price, and all the selected items participate in differential price clearing according to the electric price. The sustainable development settlement mechanism provides relatively stable income expectation for enterprises through the differential compensation of mechanism electricity price and market electricity price, and becomes a key for guaranteeing project investment return. Under the background, the declaration strategy of the mechanism electricity price is crucial to the income of new energy investment enterprises, if the declaration electricity price is too high, the competitive advantage is lost in the sequencing rule of low price priority, the declaration electricity price is difficult to be brought into the mechanism electricity quantity range, and the price difference settlement guarantee cannot be enjoyed, and if the declaration electricity price is too low, the probability of entry can be improved, the unified mechanism electricity price formed by marginal pricing can be pulled down, so that even if the successful entry is successful, reasonable profit of the project cannot be realized because the mechanism electricity price is lower than the expected income threshold. Currently, each quotation participant is often blindly declared based on the enterprise's own costs. The current quotation method has the following problems that 1, a part of high-efficiency low-cost projects are easy to obtain insufficient mechanism electric quantity due to conservation of quotation strategies, a part of low-efficiency high-cost projects occupy resources through low-price malignant competition, mechanism electric quantity resources cannot be concentrated to high-quality projects, and resource allocation is unbalanced, 2, unreasonable quotation strategies easily cause deviation of mechanism electric quantity and actual supply and demand of markets, when the mechanism electric quantity is too high, the allocation will of power grid enterprises and users can be reduced, when the mechanism electric quantity is too low, enthusiasm of enterprises for participating in market trading can be hit, the way of new energy electric quantity can be influenced, the overall absorption efficiency is reduced, 3, due to lack of scientific quotation decision basis, the reporting electric quantity is in a disordered dispersion state, price fluctuation of competitive markets is easy to be aggravated, stable and reasonable mechanism electric quantity is not favorable for an electric quantity forming in an electric power trading center, and profit uncertainty of the whole industry is increased. In summary, under the background of the new energy full-electricity marketing and sustainable development settlement mechanism, the existing mechanism el