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

Abstract

The invention provides a quotation decision-making system and method suitable for a sustainable development mechanism. The method comprises the following steps: automatically generating a simulation subject of each quotation participant according to information of each quotation participant, quotation spatial constraints and report spatial constraints; based on a clearing mechanism, the quotation and the report quantity of each simulation subject are cleared during each iteration, multiple iterations are carried out until an iteration cut-off condition is reached, each simulation subject provides the quotation and the report quantity during each iteration, and after each simulation subject provides the quotation and the report quantity, a clearing simulation unit clears the quotation and the report quantity of each simulation subject; and determining a quotation according to the income of the at least one simulation subject after multiple iteration clearing. According to the scheme, a scientific and reasonable decision reference is provided for each subject participating in bidding, and the risk of virtual high offer or ultra-low price loss can be reduced, so that resource configuration can be optimized, the new energy consumption efficiency can be improved, and price fluctuation of a quotation market can be reduced.

Description

Technical Field

[0001] This application relates to the field of power energy technology, and in particular to a pricing decision-making system and method applicable to sustainable development mechanisms. Background Technology

[0002] The mechanism for allowing all electricity generated from new energy sources to enter the market requires that, in principle, all electricity generated by new energy projects such as wind and solar power should enter the electricity market, with the on-grid price determined through market transactions. To ensure the sustainable development of the new energy industry and reduce the impact of market price fluctuations on project profitability, a sustainable development settlement mechanism has been established as a buffer guarantee mechanism for the transition of new energy from fixed electricity prices to full marketization.

[0003] Under the framework of the sustainable development settlement mechanism, a power generation project has only one successful bidding opportunity. Each bidding participant (mainly new energy investment companies) must submit their mechanism electricity price and corresponding mechanism electricity volume through market-based bidding to participate in the allocation competition for the mechanism electricity volume. Currently, the bidding clearing follows the core principles of "lowest price priority, marginal pricing, and total volume control": the energy authorities first determine the total annual mechanism electricity volume based on regional new energy consumption responsibility weights and the affordability of industrial and commercial users; each new energy investment company submits its mechanism electricity price and mechanism electricity volume declaration plan based on its own cost per kilowatt-hour and market expectations; the power trading center sorts compliant projects according to their declared mechanism electricity price from low to high, accumulating the mechanism electricity volume sequentially until the annual total volume is reached. The highest bid of the selected projects becomes the unified mechanism electricity price, and all selected projects participate in the price difference settlement based on this price.

[0004] The sustainable development settlement mechanism provides enterprises with relatively stable profit expectations through compensation for the price difference between the mechanism-based electricity price and the market electricity price, becoming crucial for ensuring the return on investment of projects. In this context, the bidding strategy for the mechanism-based electricity price is critical to the returns of new energy investment enterprises: if the bid price is too high, they will lose their competitive advantage in the "lowest price first" ranking rule, making it difficult to be included in the mechanism's electricity volume range and unable to enjoy the price difference settlement guarantee; if the bid price is too low, although it can increase the probability of selection, it will lower the unified mechanism-based electricity price formed by marginal pricing, meaning that even if successfully selected, the project may not achieve reasonable profitability because the mechanism-based electricity price is lower than the expected profit threshold.

[0005] Currently, participating bidders often submit bids blindly based on their own costs.

[0006] The current bidding method has the following problems: 1. It easily leads to some efficient and low-cost projects not obtaining sufficient quota electricity due to conservative bidding strategies, while some inefficient and high-cost projects squeeze out resources through vicious low-price competition, resulting in the inability of quota electricity resources to be concentrated on high-quality projects and an imbalance in resource allocation; 2. Unreasonable bidding strategies can easily lead to a deviation between the quota electricity price and the actual market supply and demand. When the quota electricity price is too high, it will reduce the willingness of grid companies and users to share the burden. When the quota electricity price is too low, it will dampen the enthusiasm of enterprises to participate in market transactions. Both will affect the consumption channels of new energy electricity and reduce the overall consumption efficiency; 3. Due to the lack of scientific bidding decision-making basis, the bid electricity price is disorderly and scattered, which can easily aggravate the price fluctuations in the bidding market. This is not conducive to the formation of a stable and reasonable quota electricity price by the power trading center, and also increases the uncertainty of the industry's revenue.

[0007] In summary, under the background of the full entry of renewable energy into the market and the sustainable development settlement mechanism, the existing mechanism for electricity price declaration has problems such as imbalanced resource allocation, low renewable energy consumption efficiency, and excessive price fluctuations in the bidding market. These problems seriously affect the implementation effect of the sustainable development settlement mechanism and the high-quality development of the renewable energy industry. Therefore, there is an urgent need for a scientific and reasonable optimization scheme for mechanism electricity price declaration to solve the above-mentioned technical defects. Summary of the Invention

[0008] This specification provides a pricing decision-making system and method applicable to sustainable development mechanisms, in order to address the problems that existing pricing methods under sustainable development mechanisms easily lead to imbalances in resource allocation, low efficiency in the absorption of new energy sources, and excessive price fluctuations in the pricing market.

[0009] To address the aforementioned technical problems, this specification provides a first aspect of a bidding decision-making system applicable to a sustainable development mechanism, comprising: a data acquisition unit for acquiring information on each bidding participant, the bidding space constraints of each bidding participant, and the quantity quotation space constraints of each bidding participant; a subject generation unit for generating simulated subjects of each bidding participant based on the content acquired by the data acquisition unit; each simulated subject is used to provide a bid and quantity quotation within its own bidding space constraints and quantity quotation space constraints based on a bidding strategy in each iteration; the bid is the mechanism electricity price declared under the sustainable development mechanism, and the quantity quotation is the mechanism electricity volume declared under the sustainable development mechanism; a clearing simulation unit for clearing the bids and quantities quotation of each simulated subject in each iteration based on a clearing mechanism; a driving engine for driving the clearing simulation unit and each simulated subject to perform multiple iterations until the iteration deadline is reached; in each iteration, each simulated subject provides a bid and quantity quotation, and after each simulated subject provides a bid and quantity quotation, the clearing simulation unit clears the bids and quantities quotation of each simulated subject; and a determination unit for determining a bid based on the revenue of at least one simulated subject after multiple iterations of clearing.

[0010] In some embodiments, the cost of the power generation project of the target bidding participant is known; the data acquisition unit also acquires probability estimates of the cost of the power generation projects of other bidding participants besides the target bidding participant; accordingly, the subject generation unit generates N cost estimates for the simulated subjects of the other bidding participants by random sampling based on the probability estimates of the power generation project costs; M bidding scenarios are formed by combining the N cost estimates of the other bidding participants; multiple proposed bidding strategies are generated for the subject generation unit, and the expected value of the revenue of each proposed bidding strategy under the M bidding scenarios is calculated to obtain the expected revenue value corresponding to each proposed bidding strategy; the proposed bidding strategy with the largest expected revenue value is used as the final bidding strategy determined for the target bidding participant in a single iteration; the bidding strategy includes the declared mechanism electricity price and mechanism electricity volume; where N and M are natural numbers.

[0011] In some embodiments, the data acquisition unit includes: a crawling subunit, used to crawl information of each bidding participant, the bidding space constraints of each bidding participant, and the quotation space constraints of each bidding participant from the target website; and an evaluation subunit, used to evaluate the probability estimate of the power generation project cost of each bidding participant based on the information of each bidding participant.

[0012] In some embodiments, the evaluation subunit includes an artificial intelligence network whose inputs are the installed capacity of the bidding participant's power generation project and at least one of the following: annual power generation and unit investment cost range, and whose output is a probability estimate of the cost of the bidding participant's power generation project.

[0013] In some embodiments, each simulated subject is automatically generated based on a preset template of the simulated subject, information of each bidding participant, bidding space constraints of each bidding participant, and quotation space constraints of each bidding participant; wherein, the template of the simulated subject includes: the identifier of the bidding participant, the bidding space constraints of each bidding participant, the quotation space constraints of each bidding participant, and the bidding iteration adjustment strategy.

[0014] In some embodiments, the bid iterative adjustment strategy uses the Q-learning algorithm to determine the bid and quantity for the next iteration, and uses the bid and quantity for the first iteration as the action in the Q-learning algorithm, and the profit after clearing as the state in the Q-learning algorithm.

[0015] In some embodiments, the Q-learning algorithm is used to determine the bid and quantity for the next iteration, including: before the first iteration of bidding, initializing parameters, creating a Q-table, and randomly initializing the state s; during each iteration of bidding, performing the following operations: using an ε-greedy strategy to determine the bid and quantity as an action, and executing the action a; after simulating clearing in the clearing simulation unit, calculating the revenue of the bidding participants as a reward value r; updating the new state s' after the action a according to the reward value r; updating the Q-table; and updating the current iteration's state to the new state s' in preparation for the next iteration.

[0016] In some embodiments, the bidder's payout is calculated according to the following formula: R i =R i jz +R i zc -C i dd Q i total ,

[0017] Among them, R i R represents the profit of bidder i, expressed in yuan. i jz R represents the electricity revenue earned by bidder i, expressed in yuan. i zc This represents the long-term and spot electricity revenue for bid participant i, in yuan; C i dd Q represents the cost per kilowatt-hour for bidding participant i, expressed in yuan / MWh; i total This represents the total power generation of bidder i over its entire lifecycle, in MWh.

[0018] ,

[0019] ,

[0020] Among them, P jz The price is expressed as a fixed price per MWh; P t avg Q represents the average spot market electricity price in year t, expressed in yuan / MWh; i,t jz P represents the amount of electricity won by bidder i in the t-th year, in MWh; i,t mar The price quoted by participant i in the electricity spot market transaction in year t is yuan / MWh; P i,t zc This indicates the price (in yuan / MWh) for participant i in the medium-to-long-term electricity trading in year t; Qi,t total Q represents the total electricity generation of bidder i in year t, in MWh; i,t zc Let T be the amount of electricity traded by bidder i in the medium- and long-term electricity transactions in year t, in MWh; jz The execution period for the electricity volume of the bidding participant i mechanism is one year; T l The lifespan of the bidding participants.

[0021] In some embodiments, an ε-greedy strategy is used to determine the bid and quantity, specifically including: randomly selecting an action with a probability of ε, and selecting the joint action with the largest Q value in state s from the current Q-table with a probability of 1-ε.

[0022] In some embodiments, the clearing mechanism includes: determining the selected power generation projects by ranking the bids from lowest to highest, and determining the mechanism price based on the highest bid among the selected power generation projects.

[0023] In some embodiments, the clearing mechanism includes: when multiple bidding participants have bids equal to the mechanism electricity price, allocating the mechanism electricity volume according to the proportion of the bid volumes of the multiple bidding participants.

[0024] The second aspect of this specification provides a bidding decision-making method applicable to a sustainable development mechanism, comprising: acquiring information on each bidding participant, the bidding space constraints of each bidding participant, and the quantity quotation space constraints of each bidding participant; generating simulated entities for each bidding participant based on the acquired information; each simulated entity providing a bid and quantity quotation within its own bidding space constraints and quantity quotation space constraints based on a bidding strategy in each iteration; the bid is the mechanism electricity price declared under the sustainable development mechanism, and the quantity quotation is the mechanism electricity volume declared under the sustainable development mechanism; clearing the bids and quantities quotation of each simulated entity in each iteration based on a clearing mechanism; performing multiple iterations until the iteration deadline is reached; in each iteration, each simulated entity provides a bid and quantity quotation, and after each simulated entity provides a bid and quantity quotation, clearing the bids and quantities quotation of each simulated entity; determining the bid based on the revenue of at least one simulated entity after multiple iterations of clearing.

[0025] In some embodiments, the cost of the power generation project of the target bidding participant is known. The method further includes: obtaining probabilistic estimates of the costs of the power generation projects of each bidding participant other than the target bidding participant; generating N cost estimates for the simulated entities of the other bidding participants through random sampling based on the probabilistic estimates of the power generation project costs; forming M bidding scenarios based on the N cost estimates of the other bidding participants; generating multiple proposed bidding strategies and calculating the expected revenue of each proposed bidding strategy under the M bidding scenarios, obtaining the expected revenue value corresponding to each proposed bidding strategy; and selecting the proposed bidding strategy with the largest expected revenue value as the final bidding strategy determined for the target bidding participant in a single iteration; the bidding strategy includes the declared mechanism electricity price and mechanism electricity volume; wherein N and M are natural numbers.

[0026] In some embodiments, the method further includes: crawling information of each bidding participant, the bidding space constraints of each bidding participant, and the quotation space constraints of each bidding participant from a target website; and evaluating a probability estimate of the power generation project cost of each bidding participant based on the information of each bidding participant.

[0027] In some embodiments, the probability estimate of the power generation project cost of each bidding participant is evaluated based on the information of each bidding participant, including using an artificial intelligence network to estimate the probability estimate of the power generation project cost of each bidding participant. The input of the artificial intelligence network is the installed capacity of the power generation project of the bidding participant and at least one of the following: annual power generation and unit investment cost range. The output is the probability estimate of the power generation project cost of the bidding participant.

[0028] In some embodiments, generating a simulated subject for each bidding participant based on the acquired content includes: automatically generating each simulated subject based on a preset template of the simulated subject, information of each bidding participant, bidding space constraints of each bidding participant, and quotation space constraints of each bidding participant; wherein, the template of the simulated subject includes: the identifier of the bidding participant, the bidding space constraints of each bidding participant, the quotation space constraints of each bidding participant, and the bidding iteration adjustment strategy.

[0029] In some embodiments, the bid iterative adjustment strategy uses the Q-learning algorithm to determine the bid and quantity for the next iteration, and uses the bid and quantity for the first iteration as the action in the Q-learning algorithm, and the profit after clearing as the state in the Q-learning algorithm.

[0030] In some embodiments, the Q-learning algorithm is used to determine the bid and quantity for the next iteration, including: before the first iteration of bidding, initializing parameters, creating a Q-table, and randomly initializing the state s; during each iteration of bidding, performing the following operations: using an ε-greedy strategy to determine the bid and quantity as an action, and executing the action a; after simulating clearing in the clearing simulation unit, calculating the revenue of the bidding participants as a reward value r; updating the new state s' after the action a according to the reward value r; updating the Q-table; and updating the current iteration's state to the new state s' in preparation for the next iteration.

[0031] In some embodiments, the bidder's payout is calculated according to the following formula: R i =R i jz +R i zc -C i dd Q i total , where R i R represents the profit of bidder i, expressed in yuan. i jz R represents the electricity revenue earned by bidder i, expressed in yuan. i zc This represents the long-term and spot electricity revenue for bid participant i, in yuan; C i dd Q represents the cost per kilowatt-hour for bidding participant i, expressed in yuan / MWh; i total This represents the total power generation of bidder i over its entire lifecycle, in MWh.

[0032] ,

[0033] ,

[0034] Where Pjz represents the mechanism-based electricity price, in yuan / MWh; P t avg Q represents the average spot market electricity price in year t, expressed in yuan / MWh; i,t jz P represents the amount of electricity won by bidder i in the t-th year, in MWh; i,t mar The price quoted by participant i in the electricity spot market transaction in year t is yuan / MWh; P i,t zc This indicates the price (in yuan / MWh) for participant i in the medium-to-long-term electricity trading in year t; Q i,t total Q represents the total electricity generation of bidder i in year t, in MWh;i,t zc Let T be the amount of electricity traded by bidder i in the medium- and long-term electricity transactions in year t, in MWh; jz The execution period for the electricity volume of the bidding participant i mechanism is one year; T l The lifespan of the bidding participants.

[0035] In some embodiments, an ε-greedy strategy is used to determine the bid and quantity, specifically including: randomly selecting an action with a probability of ε, and selecting the joint action with the largest Q value in state s from the current Q-table with a probability of 1-ε.

[0036] In some embodiments, the clearing mechanism includes: determining the selected power generation projects by ranking the bids from lowest to highest, and determining the mechanism price based on the highest bid among the selected power generation projects.

[0037] In some embodiments, the clearing mechanism includes: when multiple bidding participants have bids equal to the mechanism electricity price, allocating the mechanism electricity volume according to the proportion of the bid volumes of the multiple bidding participants.

[0038] A third aspect of this specification provides an electronic device, comprising: a memory and a processor, the processor and the memory being communicatively connected to each other, the memory storing computer instructions, and the processor executing the computer instructions to implement the pricing decision-making system applicable to sustainable development mechanisms as described in any one aspect of the first aspect.

[0039] A fourth aspect of this specification provides a computer storage medium storing computer program instructions that, when executed, implement the pricing decision-making system applicable to sustainable development mechanisms as described in any one of the first aspects.

[0040] The fifth aspect of this specification provides a computer program product comprising a computer program that, when executed by a processor, implements the pricing decision system applicable to sustainable development mechanisms as described in any of the first aspects.

[0041] The bidding decision-making system and method applicable to sustainable development mechanisms provided in this specification automatically generate simulated entities for each bidding participant based on their information, bidding space constraints, and quantity space constraints. Each simulated entity provides bids and quantities within its own bidding space and quantity constraints based on its bidding strategy during each iteration. A clearing simulation unit clears the bids and quantities of each simulated entity during each iteration based on a clearing mechanism. A driving engine drives the clearing simulation unit and each simulated entity through multiple iterations until the iteration deadline is reached. During each iteration, each simulated entity provides bids and quantities, and after each entity provides bids and quantities, the clearing simulation unit clears the bids and quantities. Finally, the bid is determined based on the revenue of at least one simulated entity after multiple iterations of clearing. This scheme assumes that all bidding participants are rational, aiming to maximize their own interests, and that there is no collusion among them. It iteratively simulates the bidding process of each participant multiple times. The bid is determined based on the profit of at least one simulated entity after multiple iterations and clearing. A strategy interaction model is constructed based on the analysis of the economic behavior of power trading entities, realizing dynamic game theory. By finding the equilibrium point, the optimal bid for each participant is determined, providing a scientific and reasonable decision-making reference for all participants. This reduces the risk of being eliminated due to excessively high bids or losing profits due to excessively low bids, thereby optimizing resource allocation, improving the efficiency of new energy consumption, and reducing price fluctuations in the bidding market. Attached Figure Description

[0042] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this application. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0043] Figure 1 This is a schematic diagram of a pricing decision-making system applicable to sustainable development mechanisms, as provided in this specification.

[0044] Figure 2 This is a flowchart illustrating the method for determining the price and quantity for the next iteration using the Q-learning algorithm.

[0045] Figure 3 A diagram illustrating the initial bids, final bids, and final clearing price for each entity;

[0046] Figure 4 A schematic diagram illustrating the winning bid volumes for various entities under the clearing mechanism electricity price;

[0047] Figure 5 A diagram illustrating the revenue of various stakeholders before and after participating in the electricity price bidding mechanism;

[0048] Figure 6 This is a flowchart illustrating a pricing decision-making method applicable to sustainable development mechanisms, as provided in this specification.

[0049] Figure 7 This is a schematic diagram of the electronic device provided in this specification. Detailed Implementation

[0050] To enable those skilled in the art to better understand the technical solutions in this application, the technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this application.

[0051] This specification provides a bidding decision system applicable to sustainable development mechanisms. This system can be used by power trading companies to determine clearing prices for optimizing sustainable development mechanisms, and also by new energy investment companies, as target bidding participants, to determine their bidding strategies under sustainable development mechanisms.

[0052] The pricing strategy in this manual includes the declared mechanism-based electricity price and the declared mechanism-based electricity volume. In this manual, the declared mechanism-based electricity price may also be referred to as the quoted price, and the declared mechanism-based electricity volume may also be referred to as the quoted volume.

[0053] like Figure 1 As shown, it includes a data acquisition unit 10, a main body generation unit 20, a clearing simulation unit 30, a drive engine 40, and a calculation unit 50.

[0054] The data acquisition unit 10 is used to acquire information about each bidding participant, the bidding space constraints of each bidding participant, and the quotation space constraints of each bidding participant.

[0055] Before bidding participants submit their bids, they typically need to register with the energy authority. The energy authority will first publish the information of all registered parties, and then allow each bidding participant to submit their bids. Data acquisition unit 10 can obtain more comprehensive information about the registered parties based on the information published by the energy authority.

[0056] Information about the bidding participants may include: the number of power generation projects participating in the bidding, as well as the identification, installed capacity, annual power generation, and unit investment cost range of each power generation project.

[0057] When formulating their bidding strategies (i.e., determining their bids and quantities), each bidding participant is unaware of the bids and quantities submitted by other participants. A single submission determines the final clearing mechanism price and the winning (selected) electricity volume for each participant. Since each participant's bid aims to maximize their own profitability, their bids are inevitably influenced by their costs. However, these costs are not publicly disclosed. Participants cannot know the investment and operation costs, power generation capacity, or other information of other bidders; they can only speculate based on industry standards and historical experience. Essentially, the bidding process is a static game of incomplete information.

[0058] To more accurately simulate the bidding decisions of bidding participants, the cost of their power generation projects can be probabilistically estimated based on information about the bidding participants.

[0059] Before organizing bidding, energy authorities typically announce the upper and lower limits for bids under the mechanism tariff, as well as the upper limit for each power generation project to be included in the mechanism tariff, i.e.: P min jz ≤P i jz ≤P max jz , 0≤Q i m ≤w×Q i total , where P i jz For the declared mechanism electricity price, P max jz P min jz These are the upper and lower limits of the declared mechanism electricity price, Q i m For the declared electricity volume, Q i total To obtain the total power generation of the power generation project, w is the upper limit of the application ratio for the power generation project.

[0060] For bidding participants, the upper and lower limits of the mechanism electricity price they submit are the bidding space constraints, and the upper and lower limits of the mechanism electricity volume they submit are the reporting volume space constraints.

[0061] In some embodiments, the data acquisition unit also acquires probability estimates of the cost of the power generation project for each bidding participant. Correspondingly, the subject generation unit generates a simulated subject for each bidding participant based on the probability estimates of the power generation project cost through random sampling, randomly selecting N cost estimates for each bidding participant. When the simulated subject of each bidding participant provides a bid and quantity within its own bid space constraints and quantity reporting constraints based on its bidding strategy, it determines the proposed bid and quantity based on the N cost estimates respectively. The weighted sum of the N proposed bids is used as the final bid for a single iteration, and the weighted sum of the N proposed quantities is used as the final quantity reported for a single iteration; where N is a natural number.

[0062] When N cost estimates are randomly selected for each bidding participant, these cost estimates generally conform to the rules of probability estimation. For example, if the cost range is 40-60, then the probability of a value around 50 is usually higher, while the probability around 40 and 60 is lower. The randomly selected N cost estimates conform to this characteristic. The larger the value of N, the more the distribution of the randomly selected cost estimates reflects the characteristics of probability distribution. When each simulation subject determines its bid and quantity in each iteration, it can determine the proposed bid and quantity based on each cost estimate. Finally, these proposed bids are weighted and summed to obtain the final bid for this iteration, and these proposed quantities are weighted and summed to obtain the final quantity for this iteration. The bid and quantity calculated in this way are closer to the bid and quantity determined by the bidding participants based on the actual costs, resulting in better simulation effects and a more scientific and reasonable final bid and quantity output by the bidding decision system.

[0063] In some embodiments, the cost of the power generation project of the target bidding participant is known. The data acquisition unit 10 also acquires probability estimates of the costs of the power generation projects of each bidding participant other than the target bidding participant.

[0064] Accordingly, the subject generation unit 20 generates N cost estimates for the simulated subjects of the other bidding participants by random sampling based on the probability estimate of the cost of the power generation project; M bidding scenarios are formed by combining the N cost estimates of the other bidding participants; multiple proposed bidding strategies are generated for the subject generation unit, and the expected value of the revenue of each proposed bidding strategy under the M bidding scenarios is calculated to obtain the expected revenue value corresponding to each proposed bidding strategy; the proposed bidding strategy with the largest expected revenue value is used as the final bidding strategy determined for the target bidding participant in a single iteration.

[0065] For example, consider three bidding participants, A, B, and C, where A is the target bidding participant. Assuming cost estimates b1 and b2 are generated for bidding participant B, and cost estimates c1 and c2 are generated for other bidding participants, then four (M=4) bidding scenarios can be obtained: b1+c1, b1+c2, b2+c1, and b2+c2. Three bidding strategies a1, a2, and a3 are generated for the target bidding participant A. The payoff e for each of the four bidding scenarios can be calculated when target bidding participant A uses bidding strategy a1. a11 e a12 e a13 e a14 And calculate the payoff e corresponding to pricing strategy a1. a11 e a12 e a13 e a14 Expected value E1; calculate the revenue e for each of the four bidding scenarios when the target bidding participant A adopts bidding strategy a2. a21 e a22 e a23 e a24 And calculate the payoff e corresponding to pricing strategy a2. a21 e a22 e a23 e a24 The expected value E2; calculate the revenue e for each of the four bidding scenarios when the target bidding participant A adopts bidding strategy a3. a31 e a32 e a33 e a34 And calculate the payoff e corresponding to pricing strategy a3. a31 e a32 e a33 e a34 The expected value E3; calculate the revenue e corresponding to the four bidding scenarios when the target bidding participant A adopts bidding strategy a4. a41 e a42 e a43 e a44 And calculate the payoff e corresponding to pricing strategy a4. a41 e a42 e a43 e a44 The expected value is E4. Finally, the bidding strategy corresponding to the largest expected value among the expected values ​​E1, E2, E3, and E4 is used as the final bidding strategy determined for the target bidding participant in a single iteration.

[0066] In some embodiments, the data acquisition unit includes a crawling subunit and an evaluation subunit.

[0067] The crawling sub-unit is used to crawl information about each bidding participant, their bidding space constraints, and their bid volume constraints from the target website. The information about the bidding participants can refer to relevant company introductions.

[0068] The evaluation subunit is used to assess the probability estimate of the power generation project cost for each bidding participant based on the information provided by each bidding participant.

[0069] In some embodiments, the evaluation subunit may be an artificial intelligence network, whose inputs are the installed capacity of the bidding participant's power generation project and at least one of the following: annual power generation and unit investment cost range, and whose output is a probability estimate of the cost of the bidding participant's power generation project.

[0070] The subject generation unit 20 is used to generate simulated subjects of each bidding participant based on the content obtained by the data acquisition unit; each simulated subject is used to provide bids and quantities within its own bid space constraints and quantity space constraints based on the bidding strategy in each iteration.

[0071] Each simulated entity can be automatically generated based on the preset template of the simulated entity, the information of each bidding participant, the bidding space constraints of each bidding participant, and the quotation space constraints of each bidding participant.

[0072] The template for the simulation subject may include: the identifier of the bidding participants, the probability estimate of the power generation project cost of each bidding participant, the bidding space constraints of each bidding participant, the quotation space constraints of each bidding participant, and the bidding iteration adjustment strategy.

[0073] In some embodiments, the bid iteration adjustment strategy may be based on rules. For example, if the clearing price after one iteration is x, and the current bid is higher (lower) than the clearing price x, then in the next iteration, the bid will be reduced (increased) by a fixed percentage, such as 20%.

[0074] In other embodiments, the bid iteration adjustment strategy can use the Q-learning algorithm to determine the bid and quantity for the next bid, and use the bid and quantity at the time of the first iteration as action 'a' in the Q-learning algorithm, and use the revenue after clearing as state 's' in the Q-learning algorithm.

[0075] Q-learning algorithms can approximate the optimal Q-function by continuously updating the Q-value according to an iterative strategy based on the external environment (one iteration here is...). Specifically, for example... Figure 2 As shown, the Q-learning algorithm is used to determine the price and quantity for the next iteration, including the following steps SA1 to SA9.

[0076] SA1: Before the first iteration of bidding, initialize the parameters, create the Q-table, and randomly initialize the state s.

[0077] During each iteration of the bidding process, SA2 to SA9 are executed as follows.

[0078] SA2: Employs an ε-greedy strategy to determine the price and quantity as actions, and executes combined actions.

[0079] A joint action refers to the set of actions 'a' of all bidding participants. For example, "Company A bids 0.31 yuan / kWh + Company B bids 0.32 yuan / kWh + Company C bids 0.33 yuan / kWh + Company D bids 0.30 yuan / kWh".

[0080] The ε-greedy strategy is used to determine the price and quantity, which includes: randomly selecting an action with a probability of ε, and selecting the joint action with the largest Q value in the current Q-table with a probability of 1-ε.

[0081] SA3: The clearing simulation unit simulates clearing.

[0082] SA4: Calculate the profit of the bidding participants as the reward value r.

[0083] The revenue of the bidding participants can be calculated using the following formula: R i =R i jz +R i zc -C i dd Q i total , where R i R represents the profit of bidder i, expressed in yuan. i jz R represents the electricity revenue earned by bidder i, expressed in yuan. i zc C represents the medium- and long-term electricity revenue and spot electricity revenue (i.e., non-mechanism electricity revenue) for bid participant i, in yuan; i dd Q represents the cost per kilowatt-hour for bidding participant i, expressed in yuan / MWh; i total This represents the total electricity generation of bidder i over its entire life cycle, in MWh.

[0084] ,

[0085] ,

[0086] Among them, P jz The price is expressed as a fixed price per MWh; P tavg Q represents the average spot market electricity price in year t, expressed in yuan / MWh; i,t jz P represents the amount of electricity won by bidder i in the t-th year, in MWh; i,t mar The price quoted by participant i in the electricity spot market transaction in year t is yuan / MWh; P i,t zc This indicates the price (in yuan / MWh) for participant i in the medium-to-long-term electricity trading in year t; Q i,t total Q represents the total electricity generation of bidder i in year t, in MWh; i,t zc Let T be the amount of electricity traded by bidder i in the medium- and long-term electricity transactions in year t, in MWh; jz The execution period for the electricity volume of the bidding participant i mechanism is one year; T l The lifespan of the bidding participants is typically 20-25 years.

[0087] SA5: Update the new state s' after the action a based on the reward value r.

[0088] SA6: Update Q-table.

[0089] It can be based on the Q value of the Bellman equation, specifically, Q(s,a)←Q(s,a)+α[r+γ·max] a’ Q(s',a')-Q(s,a)], where Q(s,a) is the Q value after taking action a in state s, α is the learning rate, r is the reward value, γ is the discount factor, s' is the new state after taking action a, and a' is the next action to be chosen in the new state s'.

[0090] SA7: Update the current iteration's state to the new state s' in preparation for the next iteration.

[0091] SA8: Determine if the termination state has been reached. If yes, execute SA9; otherwise, jump to SA2 to continue execution.

[0092] SA9: Determine if ε is less than or equal to a preset threshold (e.g., 0.05) and the policy is stable. If yes, output the Nash equilibrium policy.

[0093] The use of Q-learning algorithms can make the simulated subject closer to the intelligent agent, thereby enabling more accurate simulation of the trading behavior of each pricing participant.

[0094] The clearing simulation unit 30 is used to clear the bids and quantities of each simulation subject in each iteration based on the clearing mechanism.

[0095] Clearing refers to the core process in the electricity market where, by matching supply and demand and following trading rules, it ultimately determines which bidding participants will be traded, the amount of electricity traded, and at what price. Clearing simulation unit 30 can simulate the trading matching logic of electricity trading companies.

[0096] The clearing mechanism may include: selecting power generation projects based on their bids from lowest to highest, with the mechanism price determined by the highest bid among the selected projects. That is, P jz =max{θ i P i jz}, i=1,…,N; θ represents a Boolean variable, if bid participant i is included in the mechanism's electricity volume, then θ i The value is 1 if it is not 0 otherwise.

[0097] The clearing mechanism may include: when multiple bidding participants submit bids equal to the mechanism electricity price, allocating the mechanism electricity volume according to the proportion of the bids submitted by the multiple bidding participants. Specifically, the clearing mechanism determines the mechanism electricity volume for each bidding participant based on the following formula:

[0098] ,

[0099] ,

[0100] ,

[0101] ,

[0102] Among them, Q Total Let M represent the total amount of electricity subject to the mechanism, and M be the number of bidders whose bids equal the mechanism price. θ represents a Boolean variable; if bidder i is included in the mechanism's electricity supply, then θ... i The value is 1 if it is not 0 otherwise.

[0103] The electricity supply is subject to total quantity control, and to ensure sufficient competition, a reporting sufficiency rate is set, namely:

[0104] ,

[0105] Among them, Q Total Let represent the total amount of the mechanism's electricity, in MWh; θ represents a Boolean variable, which is determined if bid participant i is included in the mechanism's electricity. i The value is 1 if it is not 0 otherwise.

[0106] ,

[0107] Where η represents the sufficiency rate of the electricity declared by the mechanism.

[0108] The driving engine 40 is used to drive the clearing simulation unit and each simulation subject to perform multiple iterations until the iteration deadline is reached. During each iteration, each simulation subject provides a price quote and quantity quote. After each simulation subject provides a price quote and quantity quote, the clearing simulation unit clears the price quotes and quantities quotes of each simulation subject.

[0109] The iteration cutoff condition can be reaching a preset iteration coefficient, or the variance of the bids or bids of each bidding participant being less than a preset threshold.

[0110] The determining unit 50 is used to determine the bid based on the revenue of at least one simulated subject after multiple iterations of clearing.

[0111] The bid can be determined based on the returns of the simulated entity corresponding to a target bidder in each iteration, or it can be determined based on the returns of the simulated entities corresponding to all bidders in each iteration.

[0112] When the bidding decision system can be used by power trading companies to determine clearing prices to optimize the sustainable development mechanism, the bid determined by determination unit 50 can be used as the clearing price. When the bidding decision system is used by new energy investment companies as target bidding participants to determine their bidding strategies under the sustainable development mechanism, the bid determined by determination unit 50 can be used as the mechanism price to be declared.

[0113] The bidding decision-making system for sustainable development mechanisms provided in this specification automatically generates simulated entities for each bidding participant based on their information, bidding space constraints, and quantity space constraints. Each simulated entity provides bids and quantities within its own bidding space and quantity constraints based on its bidding strategy during each iteration. A clearing simulation unit clears the bids and quantities of each simulated entity during each iteration based on a clearing mechanism. A driving engine drives the clearing simulation unit and each simulated entity through multiple iterations until the iteration deadline is reached. During each iteration, each simulated entity provides bids and quantities, and after each entity provides bids and quantities, the clearing simulation unit clears the bids and quantities. Finally, the bid is determined based on the revenue of at least one simulated entity after multiple iterations of clearing. This scheme assumes that all bidding participants are rational, aiming to maximize their own interests, and that there is no collusion among them. It iteratively simulates the bidding process of each participant multiple times. The bid is determined based on the profit of at least one simulated entity after multiple iterations and clearing. A strategy interaction model is constructed based on the analysis of the economic behavior of power trading entities, realizing dynamic game theory. By finding the equilibrium point, the optimal bid for each participant is determined, providing a scientific and reasonable decision-making reference for all participants. This reduces the risk of being eliminated due to excessively high bids or losing profits due to excessively low bids, thereby optimizing resource allocation, improving the efficiency of new energy consumption, and reducing price fluctuations in the bidding market.

[0114] To verify the effectiveness of the above system, simulation analysis was conducted using historical average data to simulate information from 50 new energy power stations (i.e., the main entities in Table 1 below, i.e., the bidding participants). Some information is shown in Table 1 below. The relevant boundary conditions for market bidding are shown in Table 2 below.

[0115] ,

[0116] ,

[0117] The simulation involved bidding behavior from 50 entities, with 1000 iterations. In the first half of the iteration, all entities continuously lowered their bids. In the second half, low-priced and high-priced entities gradually converged to the cost per kilowatt-hour, while entities with bids close to the clearing price readjusted their bids based on each clearing price. Entities at the margin (those with bids equal to the clearing price) would raise their bids to pursue excess profits, while those with bids slightly above the clearing price would drastically lower their bids to the cost per kilowatt-hour to secure the bid, and this cycle continued.

[0118] In the simulated scenario, each participant can adjust their bid according to the situation. However, in the actual bidding, there is only one opportunity to bid. Therefore, for most participants, using the cost per kilowatt-hour as the bidding strategy is the most scientific and reasonable choice. The initial bids, final bids, and final clearing prices of each participant are as follows: Figure 3 As shown, all entities initially quoted prices higher than their own costs, but eventually converged to near the cost per kilowatt-hour, with the clearing price at 0.3098 yuan / kWh.

[0119] Under the clearing mechanism pricing, the electricity volumes won by each entity under the mechanism are as follows: Figure 4 As shown, the winning bid volume for each entity does not exceed 80%, and the remaining quota volume for the four marginal entities is allocated according to the proportion of the declared volume.

[0120] The benefits of each participant in the electricity price bidding mechanism before and after the bidding process are as follows: Figure 5 As shown in the diagram, the main revenue streams included in the mechanism have all seen significant increases, providing a certain level of protection for project profitability, especially for low-priced projects where revenue has increased exponentially. Conversely, high-priced projects, due to their excessively high bids, are either excluded from the mechanism's revenue stream or, even if included, their revenue remains negative, facing a significant risk of being phased out. This aligns perfectly with the design objectives of the sustainable development mechanism: to eliminate high-priced, inefficient, and substandard projects while ensuring the availability of low-priced, high-efficiency, high-quality projects, thereby optimizing resource allocation.

[0121] This specification also provides a pricing decision-making method suitable for sustainable development mechanisms, which can be implemented using the aforementioned pricing decision-making system suitable for sustainable development mechanisms. For example... Figure 6 As shown, the method includes the following steps S10 to S50.

[0122] S10: Obtain information on each bidding participant, the bidding space constraints of each bidding participant, and the quotation space constraints of each bidding participant.

[0123] S20: Generate simulated entities for each bidding participant based on the acquired content; each simulated entity is used to provide bids and quantities within its own bid space constraints and quantity space constraints based on the bidding strategy in each iteration. The bid is the mechanism electricity price declared under the sustainable development mechanism, and the quantity is the mechanism electricity volume declared under the sustainable development mechanism.

[0124] S30: Based on the clearing mechanism, the bids and quantities of each simulated entity are cleared in each iteration.

[0125] S40: Perform multiple iterations until the iteration deadline is reached; in each iteration, each simulated entity provides a price quote and quantity quote, and after each simulated entity provides a price quote and quantity quote, the price quotes and quantities quotes of each simulated entity are cleared.

[0126] S50: Determine the bid based on the revenue of at least one simulated entity after multiple iterations of clearing.

[0127] In some embodiments, the cost of the power generation project of the target bidding participant is known. The method further includes: obtaining probabilistic estimates of the costs of the power generation projects of each bidding participant other than the target bidding participant; generating N cost estimates for the simulated entities of the other bidding participants through random sampling based on the probabilistic estimates of the power generation project costs; forming M bidding scenarios based on the N cost estimates of the other bidding participants; generating multiple proposed bidding strategies and calculating the expected revenue of each proposed bidding strategy under the M bidding scenarios, obtaining the expected revenue value corresponding to each proposed bidding strategy; and selecting the proposed bidding strategy with the largest expected revenue value as the final bidding strategy determined for the target bidding participant in a single iteration; the bidding strategy includes the declared mechanism electricity price and mechanism electricity volume; wherein N and M are natural numbers.

[0128] In some embodiments, the method further includes: crawling information of each bidding participant, the bidding space constraints of each bidding participant, and the quotation space constraints of each bidding participant from the target website; and evaluating a probability estimate of the power generation project cost of each bidding participant based on the information of each bidding participant.

[0129] In some embodiments, the probability estimate of the power generation project cost of each bidding participant is evaluated based on the information of each bidding participant, including using an artificial intelligence network to estimate the probability estimate of the power generation project cost of each bidding participant. The input of the artificial intelligence network is the installed capacity of the power generation project of the bidding participant and at least one of the following: annual power generation and unit investment cost range. The output is the probability estimate of the power generation project cost of the bidding participant.

[0130] In some embodiments, generating a simulation subject for each bidding participant based on the acquired content includes: automatically generating each simulation subject based on a preset simulation subject template, information of each bidding participant, probability estimates of the power generation project cost of each bidding participant, bidding space constraints of each bidding participant, and reporting volume space constraints of each bidding participant; wherein, the simulation subject template includes: the identifier of the bidding participant, the probability estimate of the power generation project cost of each bidding participant, the bidding space constraints of each bidding participant, the reporting volume space constraints of each bidding participant, and the bidding iteration adjustment strategy.

[0131] In some embodiments, the bid iterative adjustment strategy uses the Q-learning algorithm to determine the bid and quantity for the next iteration, and uses the bid and quantity for the first iteration as the action in the Q-learning algorithm, and the profit after clearing as the state in the Q-learning algorithm.

[0132] In some embodiments, the Q-learning algorithm is used to determine the bid and quantity for the next iteration, including: before the first iteration of bidding, initializing parameters, creating a Q-table, and randomly initializing the state s; during each iteration of bidding, performing the following operations: using an ε-greedy strategy to determine the bid and quantity as an action, and executing the action a; after simulating clearing in the clearing simulation unit, calculating the revenue of the bidding participants as a reward value r; updating the new state s' after the action a according to the reward value r; updating the Q-table; and updating the current iteration's state to the new state s' in preparation for the next iteration.

[0133] In market bidding, there are many bidding participants and the game process is relatively complex. Using the traditional Q-learning algorithm may cause the curse of dimensionality. To address this, factorization dimensionality reduction technology can be used to optimize the algorithm and improve its applicability.

[0134] In some embodiments, the bidder's payout is calculated according to the following formula: R i =R i jz +R i zc -C i dd Q i total ,

[0135] Among them, R i R represents the profit of bidder i, expressed in yuan. i jz R represents the electricity revenue earned by bidder i, expressed in yuan. i zc This represents the long-term and spot electricity revenue for bid participant i, in yuan; C i dd Q represents the cost per kilowatt-hour for bidding participant i, expressed in yuan / MWh; i total This represents the total electricity generation of bidder i over its entire life cycle, in MWh.

[0136] ,

[0137] ,

[0138] Where Pjz represents the mechanism-based electricity price, in yuan / MWh; P t avg Q represents the average spot market electricity price in year t, expressed in yuan / MWh; i,t jz P represents the amount of electricity won by bidder i in the t-th year, in MWh; i,t marThe price quoted by participant i in the electricity spot market transaction in year t is yuan / MWh; P i,t zc This indicates the price (in yuan / MWh) for participant i in the medium-to-long-term electricity trading in year t; Q i,t total Q represents the total electricity generation of bidder i in year t, in MWh; i,t zc Let T be the amount of electricity traded by bidder i in the medium- and long-term electricity transactions in year t, in MWh; jz The execution period for the electricity volume of the bidding participant i mechanism is one year; T l The lifespan of the bidding participants.

[0139] In some embodiments, an ε-greedy strategy is used to determine the bid and quantity, specifically including: randomly selecting an action with a probability of ε, and selecting the joint action with the largest Q value in state s from the current Q-table with a probability of 1-ε.

[0140] In some embodiments, the clearing mechanism includes: determining the selected power generation projects by ranking the bids from lowest to highest, and determining the mechanism price based on the highest bid among the selected power generation projects.

[0141] In some embodiments, the clearing mechanism includes: when multiple bidding participants have bids equal to the mechanism electricity price, allocating the mechanism electricity volume according to the proportion of the bid volumes of the multiple bidding participants.

[0142] The descriptions and functions of the above methods can be found in the section on pricing decision-making systems applicable to sustainable development mechanisms, and will not be repeated here.

[0143] This invention also provides an electronic device, such as... Figure 7 As shown, the electronic device may include a processor 701 and a memory 702, wherein the processor 701 and the memory 702 may be connected via a bus or other means. Figure 7 Taking the example of a connection between China and Israel via a bus.

[0144] Processor 701 can be a central processing unit (CPU). Processor 701 can also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or combinations of the above types of chips.

[0145] Memory 702, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs, non-transitory computer-executable programs, and modules, such as the program instructions / modules corresponding to the pricing decision system applicable to the sustainable development mechanism in this embodiment of the invention (e.g., Figure 1 The data acquisition unit 10, the main generation unit 20, the clearing simulation unit 30, the drive engine 40, and the calculation unit 50 are shown in the diagram. The processor 701 executes various functional applications and data processing by running non-transitory software programs, instructions, and modules stored in the memory 702, thereby realizing the bidding decision system suitable for sustainable development mechanisms in the above system embodiment.

[0146] The memory 702 may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created by the processor 701, etc. Furthermore, the memory 702 may include high-speed random access memory and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, the memory 702 may optionally include memory remotely located relative to the processor 701, and these remote memories may be connected to the processor 701 via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0147] The one or more modules are stored in the memory 702 and, when executed by the processor 701, implement the above-mentioned pricing decision system applicable to sustainable development mechanisms.

[0148] The specific details of the above-mentioned electronic devices can be understood by referring to the relevant descriptions and effects in the system embodiments, and will not be repeated here.

[0149] This specification also provides a computer storage medium storing computer program instructions that, when executed, implement the aforementioned pricing decision-making system applicable to sustainable development mechanisms.

[0150] This specification also provides a computer program product comprising a computer program that, when executed by a processor, implements the aforementioned pricing decision system applicable to sustainable development mechanisms.

[0151] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), random access memory (RAM), flash memory, hard disk drive (HDD), or solid-state drive (SSD), etc.; the storage medium can also include combinations of the above types of memory.

[0152] The various embodiments in this specification are described in a progressive manner. For the same or similar parts between the various embodiments, please refer to each other. The focus of each embodiment is to describe the differences from other embodiments.

[0153] The systems, devices, modules, or units described in the above embodiments can be implemented by computer chips or entities, or by products with certain functions.

[0154] For ease of description, the above devices are described separately by function as various units. Of course, in implementing this application, the functions of each unit can be implemented in one or more software and / or hardware.

[0155] As can be seen from the above description of the embodiments, those skilled in the art can clearly understand that this application can be implemented by means of software plus necessary general-purpose hardware platforms. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute certain parts of the methods of various embodiments of this application.

[0156] This application can be used in a wide variety of general-purpose or special-purpose computer system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics devices, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, etc.

[0157] This application can be described in the general context of computer-executable instructions, such as program modules, that are executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform a specific task or implement a specific abstract data type. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0158] Although this application has been described through embodiments, those skilled in the art will know that this application has many modifications and variations without departing from the spirit of this application, and it is intended that the appended claims cover such modifications and variations without departing from the spirit of this application.

Claims

1. A pricing decision-making system suitable for sustainable development mechanisms, characterized in that, include: The data acquisition unit is used to acquire information about each bidding participant, the bidding space constraints of each bidding participant, and the quotation space constraints of each bidding participant. The subject generation unit is used to generate simulated subjects of each bidding participant based on the content obtained by the data acquisition unit; each simulated subject is used to provide a bid and a quantity based on its own bid space constraints and quantity space constraints in each iteration; the bid is the mechanism electricity price declared under the sustainable development mechanism, and the quantity is the mechanism electricity volume declared under the sustainable development mechanism; The clearing simulation unit is used to clear the bids and quantities of each simulation subject in each iteration based on the clearing mechanism; The driving engine is used to drive the clearing simulation unit and each simulation subject to perform multiple iterations until the iteration cutoff condition is reached. In each iteration, each simulation entity provides a price quote and quantity quote. After each simulation entity provides a price quote and quantity quote, the clearing simulation unit clears the price quotes and quantities quotes of each simulation entity. A determining unit is used to determine the bid based on the revenue of at least one simulated entity after multiple iterations of clearing.

2. The system according to claim 1, characterized in that, The cost of the power generation project of the target bidder is known; the data acquisition unit also acquires probability estimates of the cost of the power generation projects of other bidders besides the target bidder. Accordingly, the subject generation unit generates N cost estimates for the simulated subjects of the other bidding participants by random sampling based on the probability estimate of the power generation project cost; M bidding scenarios are formed by combining the N cost estimates of the other bidding participants; multiple proposed bidding strategies are generated for the subject generation unit, and the expected value of the revenue of each proposed bidding strategy under the M bidding scenarios is calculated to obtain the expected revenue value corresponding to each proposed bidding strategy; the proposed bidding strategy with the largest expected revenue value is used as the final bidding strategy determined for the target bidding participant in a single iteration; the bidding strategy includes the declared mechanism electricity price and mechanism electricity volume; where N and M are natural numbers.

3. The system according to claim 2, characterized in that, The data acquisition unit includes: The crawling sub-unit is used to crawl information about each bidding participant, the bidding space constraints of each bidding participant, and the bidding volume space constraints of each bidding participant from the target website. The evaluation subunit is used to assess the probability estimate of the power generation project cost for each bidding participant based on information from each bidding participant.

4. The system according to claim 3, characterized in that, The evaluation subunit includes: An artificial intelligence network takes as input the installed capacity of the power generation projects of bidding participants and at least one of the following: annual power generation and unit investment cost range, and outputs a probability estimate of the power generation project cost of the bidding participants.

5. The system according to claim 1, characterized in that, Based on the pre-set template of the simulated subject, the information of each bidding participant, the bidding space constraints of each bidding participant, and the quotation space constraints of each bidding participant, each simulated subject is automatically generated. The template for the simulated subject includes: the identifier of the bidding participants, the bidding space constraints of each bidding participant, the quotation space constraints of each bidding participant, and the bidding iteration and adjustment strategy.

6. The system according to claim 5, characterized in that, The proposed price iterative adjustment strategy uses the Q-learning algorithm to determine the price and quantity for the next iteration, and uses the price and quantity at the time of the first iteration as the action in the Q-learning algorithm, and the profit after clearing as the state in the Q-learning algorithm.

7. The system according to claim 6, characterized in that, The Q-learning algorithm is used to determine the price and quantity for the next iteration, including: Before the first iteration of bidding, initialize the parameters, create the Q-table, and randomly initialize the state s; During each iteration of the bidding process, the following operations are performed: the ε-greedy strategy is used to determine the bid and quantity as actions, and the joint action is executed; after the clearing simulation unit simulates clearing, the revenue of the bidding participants is calculated as the reward value r; the new state s' after the joint action is updated according to the reward value r; the Q-table is updated; the state of the current iteration is updated to the new state s' in preparation for the next iteration.

8. The system according to claim 7, characterized in that, Calculate the bidder's payout using the following formula: R i =R i jz +R i zc -C i dd Q i total , Among them, R i R represents the profit of bidder i, expressed in yuan. i jz R represents the electricity revenue earned by bidder i, expressed in yuan. i zc This represents the long-term and spot electricity revenue for bid participant i, in yuan; C i dd Q represents the cost per kilowatt-hour for bidding participant i, expressed in yuan / MWh; i total This represents the total power generation of bidder i over its entire lifecycle, in MWh. ， ， Among them, P jz This indicates the price of electricity from the grid, expressed in yuan / MWh; P t avg Q represents the average spot market electricity price in year t, expressed in yuan / MWh; i,t jz P represents the amount of electricity won by bidder i in the t-th year, in MWh; i,t mar The price quoted by participant i in the electricity spot market transaction in year t is yuan / MWh; P i,t zc This indicates the price (in yuan / MWh) for participant i in the medium-to-long-term electricity trading in year t; Q i,t total Q represents the total electricity generation of bidder i in year t, in MWh; i,t zc Let T be the amount of electricity traded by bidder i in the medium- and long-term electricity transactions in year t, in MWh; jz The execution period for the electricity volume of the bidding participant i mechanism is one year; T l The lifespan of the bidding participants.

9. The system according to claim 7, characterized in that, The ε-greedy strategy is used to determine the price and quantity, which includes: randomly selecting an action with a probability of ε, and selecting the joint action with the largest Q value in the current Q-table with a probability of 1-ε.

10. The system according to claim 1, characterized in that, The clearing mechanism includes: determining the selected power generation projects by ranking the bids from lowest to highest, and determining the mechanism price based on the highest bid among the selected power generation projects.

11. The system according to claim 1, characterized in that, The clearing mechanism includes: when multiple bidding participants submit bids equal to the mechanism electricity price, allocating the mechanism electricity volume according to the proportion of the bids submitted by the multiple bidding participants.

12. A pricing decision-making method applicable to sustainable development mechanisms, characterized in that, include: Obtain information about each bidding participant, their bidding space constraints, and their quotation space constraints; Based on the acquired content, simulated entities for each bidding participant are generated; each simulated entity is used to provide a bid and quantity within its own bid space constraints and quantity space constraints based on the bidding strategy in each iteration; the bid is the mechanism electricity price declared under the sustainable development mechanism, and the quantity is the mechanism electricity volume declared under the sustainable development mechanism. The clearing mechanism clears out the bids and quantities of each simulated entity in each iteration; Perform multiple iterations until the iteration deadline is reached; In each iteration, each simulated entity provides a price quote and quantity quote. After each simulated entity provides a price quote and quantity quote, the price quotes and quantities quotes of each simulated entity are cleared. The bid is determined based on the revenue of at least one simulated entity after multiple iterations of clearing.

13. An electronic device, characterized in that, include: A memory and a processor are communicatively connected, the memory storing computer instructions, and the processor executing the computer instructions to implement the pricing decision system applicable to sustainable development mechanisms as described in any one of claims 1-11.

14. A computer storage medium, characterized in that, The computer storage medium stores computer program instructions, which, when executed, implement the pricing decision-making system applicable to sustainable development mechanisms as described in any one of claims 1-11.

15. A computer program product, characterized in that, It includes a computer program that, when executed by a processor, implements the pricing decision system applicable to sustainable development mechanisms as described in any one of claims 1-11.

Images