Transaction decision-making method and system in market joint clearing environment

Abstract

The invention discloses a transaction decision-making method and system in a market joint clearing environment, and relates to the technical field of electric power spot market transaction, and the method comprises the steps: S1, obtaining a predicted value of a node marginal electricity price, and constructing a quotation optimization model of each power generation main body; s2, designing an agent, and training the agent by taking the quotation optimization model as a reward function; s3, generating quotation data based on the trained agents, and constructing a joint clearing model of all power generation subjects; s4, calculating the actual value of the node marginal electricity price based on the joint clearing model, and calculating the deviation between the actual value and the predicted value; s5, judging whether the deviation exceeds a preset threshold value or not, if so, taking the actual value as a predicted value, and returning to the step S1 for closed-loop learning; and if not, taking the quotation data generated by each agent as a final quotation strategy. According to the invention, global optimization and intelligent collaboration of power generation subject transaction decisions can be realized in a complex market environment.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of electricity spot market transaction, and in particular to a transaction decision-making method and system in a market joint clearing environment. BACKGROUND

[0002] With the deepening of the reform of the electricity market, the electricity market system has developed from a single electricity market to a complex form of "medium and long-term - spot - ancillary service" multi-market coordinated operation. Under this background, market players such as power generation enterprises and energy storage operators are facing the challenge of optimizing decision-making in multiple markets to maximize their profits.

[0003] Currently, the technical methods applied to electricity transaction decision-making have mainly undergone two generations of evolution. The first generation is based on traditional optimization algorithms, using cost-plus models or linear programming methods based on a single electricity price prediction curve. This kind of method relies on accurate market environment prediction and has certain applicability in the early period of relatively stable market structure. The second generation of technology begins to introduce artificial intelligence technology, mainly focusing on strategy optimization in a single market environment, such as using Q-learning algorithm to solve the multi-stage profit optimization problem under incomplete information, applying deep deterministic policy gradient algorithm to process continuous bidding space, and using multi-agent reinforcement learning to simulate the game behavior between market players.

[0004] When these existing technologies are applied to the joint clearing environment of electricity and ancillary service markets, their inherent limitations gradually emerge. At the methodological level, existing technologies regard the electricity market and the ancillary service market as independent decision-making environments and adopt a sequential decision-making mode of "first electricity, then ancillary services". This fragmented decision-making method cannot accurately quantify the opportunity cost between the electricity market and the ancillary service market, ignores the mutually exclusive relationship between the two markets in capacity allocation, and leads the power generation subject to fall into the "local optimal trap" when allocating resources. At the algorithm implementation level, existing technologies face serious technical bottlenecks. First, the prediction vulnerability problem, the existing method relies too much on a single electricity price prediction trajectory, when the actual price deviates from the prediction due to factors such as renewable energy output fluctuation and load change, the decision-making effect will deteriorate sharply. Secondly, the dimension disaster problem, in a high-dimensional decision space, traditional algorithms are difficult to obtain high-quality solutions within a limited bidding time. In addition, in a multi-agent game environment, the algorithm often falls into local optimum or oscillation convergence, which seriously affects the decision-making effect. From the perspective of system operation, existing intelligent decision-making technologies have inherent defects. In the early stage of market operation, intelligent agents lack prior knowledge, leading to drastic fluctuations in strategy, which in turn causes market price fluctuations, forming a negative feedback loop and exacerbating strategy instability. Since a long "trial and error learning" phase is needed to reach a stable state, this process not only affects the stability of market players' profits, but also threatens the smooth operation of the electricity market.

[0005] The above limitations not only restrict the competition ability of the power generation subject in the complex market environment, but also hinder the improvement of the overall operation efficiency of the power market, and need to be solved by technical innovation. SUMMARY

[0006] The purpose of the present application is to overcome the deficiencies in the prior art, provide a transaction decision method and system in a market joint clearing environment, which can realize the global optimization and intelligent collaboration of the transaction decision of the power generation subject in the complex environment of joint clearing of the electricity market and auxiliary service market, improve the competition ability of the power generation subject in the complex market environment, and effectively improve the overall operation efficiency of the power market.

[0007] To achieve the above purpose, the present application adopts the following technical solutions:

[0008] In a first aspect, the present application provides a transaction decision method in a market joint clearing environment, comprising:

[0009] Step S1, obtaining the predicted value of the node marginal price, and constructing a bidding optimization model of each power generation subject with the goal of maximizing the comprehensive income of the power generation subject;

[0010] Step S2, designing an agent capable of directly outputting continuous bidding for each power generation subject, and training the agent with the bidding optimization model as the reward function;

[0011] Step S3, generating bidding data of each power generation subject based on the trained agent, and constructing a joint clearing model of all power generation subjects with the goal of minimizing the total cost of the whole society;

[0012] Step S4, calculating the actual value of the node marginal price based on the joint clearing model, and calculating the deviation between the actual value and the predicted value of the node marginal price;

[0013] Step S5, judging whether the deviation exceeds a preset threshold, if yes, taking the actual value of the node marginal price as its predicted value, and returning to step S1 for closed-loop learning; if not, taking the bidding data generated by each agent as the final bidding strategy.

[0014] Optionally, the bidding optimization model is:

[0015] In the formula, are the medium and long-term electricity income and the base electricity income of the unit corresponding to the t period;

[0016] is the winning electricity quantity income of the unit in the day-ahead market of the electricity spot market corresponding to the t period, and is the winning electricity quantity income of the unit in the real-time market of the electricity spot market corresponding to the t period; k is the prediction curve of the kth node marginal electricity price;

[0017] is the compensation fee and the auxiliary service market income of the unit corresponding to the t period, respectively;

[0018] is the generation cost, the start-stop cost and the shutdown penalty cost of the unit corresponding to the t period, respectively, is the generation power of the unit corresponding to the t period;

[0019] is the confidence probability of the prediction curve of the kth node marginal electricity price, T is the total period number, and K is the total number.

[0020] Optionally, the winning electricity quantity income in the day-ahead market of the electricity spot market is:

[0021]

[0022] In the formula, is the winning electricity in the day-ahead market of the t period, is the prediction value of the node electricity price in the day-ahead market of the t period, is the house service rate of the unit;

[0023] The winning electricity quantity income in the real-time market of the electricity spot market is:

[0024]

[0025] In the formula, is the winning electricity in the real-time market of the t period, is the prediction value of the node electricity price in the real-time market of the t period.

[0026] Optionally, the constraint condition for the agent to generate a decision includes:

[0027] a bidding upper and lower limit constraint and a physical operation constraint, the physical operation constraint including a unit output upper and lower limit constraint, a unit output ramping rate constraint, a minimum start-stop time constraint and a state of charge constraint of an energy storage unit.

[0028] Optionally, the joint clearing model is:

[0029]

[0030] In the formula, is the operation cost, the start cost and the generation power of the ith unit corresponding to the t period, respectively; respectively, are the positive reserve offer and the winning positive reserve capacity of the i-th unit at the t-th time interval; N is the total number of units, and T is the total number of time intervals;

[0031] respectively, are the operating cost and the charge / discharge power of the e-th energy storage unit at the t-th time interval, respectively, are the positive reserve offer and the winning positive reserve capacity of the e-th energy storage unit at the t-th time interval; is the total number of energy storage units;

[0032] respectively, are the operating cost and the generation / pumping power of the p-th pumped storage unit at the t-th time interval, respectively, are the positive reserve offer and the winning positive reserve capacity of the p-th pumped storage unit at the t-th time interval; is the total number of pumped storage units;

[0033] is the network flow constraint relaxation penalty factor for market clearing optimization, respectively, are the positive and negative flow relaxation variables of the l-th line, and NL is the total number of lines; respectively, are the positive and negative flow relaxation variables of the s-th section, and NS is the total number of sections.

[0034] Optionally, the constraint condition based on the actual value of the node marginal electricity price calculated by the joint clearing model comprises:

[0035] a power balance constraint, a unit operating characteristic constraint, a unit winning electricity quantity constraint, a network flow constraint, and an ancillary service demand constraint.

[0036] In a second aspect, the present application provides a transaction decision system in a market joint clearing environment, comprising:

[0037] a bid optimization model module configured to obtain a predicted value of a node marginal electricity price, and to construct a bid optimization model of each power generation subject with the goal of maximizing the comprehensive revenue of the power generation subject;

[0038] an agent design module configured to design an agent capable of directly outputting continuous bids for each power generation subject, and to train the agent with the bid optimization model as a reward function;

[0039] a joint clearing model module configured to generate bid data of each power generation subject based on the trained agent, and to construct a joint clearing model of all power generation subjects with the goal of minimizing the total cost of the whole society;

[0040] an electricity price deviation calculation module configured to calculate the actual value of the node marginal electricity price based on the joint clearing model, and to calculate the deviation between the actual value and the predicted value of the node marginal electricity price;

[0041] The closed-loop learning module is configured to determine whether the deviation exceeds a preset threshold value, if yes, take the actual value of the node marginal price as the predicted value thereof, and return to the bidding optimization model module for closed-loop learning, and if no, take the bidding data generated by each agent as the final bidding strategy.

[0042] In a third aspect, the present application provides an electronic device, comprising a processor and a storage medium.

[0043] The storage medium is used for storing instructions.

[0044] The processor is used for operating according to the instructions to perform the steps of the above method.

[0045] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above method.

[0046] In a fifth aspect, the present application provides a computer program product comprising a computer program / instructions, which, when executed by a processor, implements the steps of the above method.

[0047] Compared with the prior art, the present application has the following beneficial effects:

[0048] The market joint clearing environment under the transaction decision method and system provided by the present application, by constructing a double-layer optimization model in the environment of joint clearing of electricity market and auxiliary service market, namely a bidding optimization model and a joint clearing model, in the bidding optimization of power generation subjects, based on the multi-element subject bidding strategy generation algorithm of reinforcement learning, a local optimal decision is obtained, then through the iteration feedback and convergence mechanism between the double-layer optimization models, the selfish decision of the power generation subject is coordinated with the safe and economic operation of the power system, and finally the market is guided to reach or approach the equilibrium state. BRIEF DESCRIPTION OF DRAWINGS

[0049] Figure 1 is a flowchart of the market joint clearing environment under the transaction decision method provided by the embodiment of the present application;

[0050] Figure 2 is a comparison diagram of the actual value and the predicted value of the node marginal price and the deviation rate thereof;

[0051] Figure 3 is a block diagram of the market joint clearing environment under the transaction decision system provided by the embodiment of the present application. DETAILED DESCRIPTION

[0052] The present invention will be further described below with reference to the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and should not be used to limit the scope of protection of the present invention.

[0053] Example 1

[0054] In a first aspect, the present invention provides a transaction decision-making method under a market joint clearing environment, comprising:

[0055] Step S1: Obtain the predicted value of the marginal electricity price of the node, and construct a bidding optimization model for each power generation entity with the goal of maximizing the comprehensive income of the power generation entity.

[0056] In this embodiment, an operating day is taken as a cycle and divided into 96 time periods, each time period being 15 minutes. The predicted curve of the node marginal electricity price for the 96 time periods of the operating day, obtained based on historical data and market analysis, is used as the initial predicted value of the node marginal electricity price.

[0057] The pricing optimization model comprehensively considers revenue from medium- and long-term contract electricity, base electricity revenue, spot market revenue, compensation fees, and ancillary service market revenue. The specific expression is as follows:

[0058]

[0059] In the formula, These are the medium- and long-term electricity revenue and the base electricity revenue of the generating unit in time period t, respectively.

[0060] These represent the day-ahead market revenue and real-time market revenue of the generating unit in the spot electricity market corresponding to time period t; k is the prediction curve of the marginal electricity price at node k.

[0061] These represent the compensation costs and ancillary service market revenue for the generating unit during time period t, respectively.

[0062] These represent the power generation cost, start-up and shutdown cost, and shutdown penalty cost of the generating unit during time period t, respectively. The generating capacity of the unit during time period t;

[0063] Let T be the confidence probability of the predicted curve of the marginal electricity price of the k-th node, where T is the total number of time periods and K is the total number of nodes.

[0064] Specifically, in this embodiment, the revenue from day-ahead electricity bids in the electricity spot market is as follows:

[0065]

[0066] In the formula, is the winning power in the day-ahead market of period t, is the predicted value of the node price in the day-ahead market of period t, is the unit consumption rate of the unit;

[0067] The winning power in the real-time market is:

[0068]

[0069] In the formula, is the winning power in the real-time market of period t, is the predicted value of the node price in the real-time market of period t.

[0070] The compensation cost is:

[0071]

[0072]

[0073] In the formula, is the market electricity compensation cost, is the operation compensation cost; is the unit start-up compensation cost; is the market electricity cost subsidy price.

[0074] The auxiliary service market income is:

[0075]

[0076] In the formula, is the expected income of the unit participating in the frequency modulation market, is the auxiliary service coefficient;

[0077] The unit start-stop cost and shutdown penalty cost are:

[0078]

[0079]

[0080] In the formula, is the start-up cost of the unit, and the start-up action period is included in the cost; is the shutdown cost of the unit, and the shutdown action period is included in the cost; if the user expects to start up, a shutdown penalty cost is generated for the shutdown period , and the shutdown penalty cost is set to a large value.

[0081] Step S2: Design an agent capable of directly outputting continuous bids for each power generation subject, and use the bid optimization model as a reward function to train the agent.

[0082] Specifically in the embodiment, a deep reinforcement learning algorithm, in particular a deep deterministic policy gradient algorithm, is adopted, an environment model containing market states and unit states is constructed, an agent capable of directly outputting continuous bidding actions is designed, and the agent and the environment model are repeatedly interacted and learned for training.

[0083] The configuration of the environment model includes market rule model configuration, power system model construction, market boundary condition preparation, subject cost and prediction model configuration, and is configured based on existing technologies, which will not be described in detail in the application.

[0084] In the strategy generation process, the upper and lower limits of the bid and various physical operation constraints must be strictly considered, such as:

[0085] Upper and lower limits of unit output , Upper and lower limits of unit output is a start-stop state variable, and a value of 1 indicates that the unit is started, and a value of 0 indicates that the unit is stopped;

[0086] Ramp rate constraints of unit output , are upper and lower limits of the ramp rate;

[0087] Minimum start-stop time constraints, state of charge constraints of energy storage units, and the like.

[0088] Step S3, based on the trained agents, generating bidding data of each power generation subject, and constructing a joint clearing model of all power generation subjects with the goal of minimizing the total social cost.

[0089] Specifically in the embodiment, the joint clearing model is:

[0090]

[0091] In the formula, are respectively the operating cost, the start cost and the power generation power of the i-th unit at the t-th period; are respectively the positive reserve bid and the winning positive reserve capacity of the i-th unit at the t-th period; N is the total number of units, and T is the total number of periods;

[0092] are respectively the operating cost and the charge / discharge power of the e-th energy storage unit at the t-th period, are respectively the positive reserve bid and the winning positive reserve capacity of the e-th energy storage unit at the t-th period; is the total number of energy storage units;

[0093] are respectively the operating cost and the power generation / pumping power of the p-th pumped storage unit at the t-th period, respectively, are the positive reserve offer and the winning positive reserve capacity of the pth pumped storage unit at the t period; is the total number of pumped storage units;

[0094] is the network flow constraint relaxation penalty factor for market clearing optimization, respectively, are the positive and negative flow relaxation variables of the lth line, and NL is the total number of lines; respectively, are the positive and negative flow relaxation variables of the s th section, and NS is the total number of sections.

[0095] In the clearing process, related constraints must be strictly met, such as power balance constraints, unit operation characteristic constraints , is the maximum power generation, is the coefficient parameter, and the winning power of the unit is constrained , is the reserve capacity requirement, the network flow constraint, and the auxiliary service requirement constraint, etc.

[0096] Step S4, calculate the actual value of the nodal marginal price based on the joint clearing model, and calculate the deviation between the actual value and the predicted value of the nodal marginal price.

[0097] Specifically, in this embodiment, the average deviation rate is taken as the deviation index, as shown in Figure 2 In other alternative embodiments, other deviation indicators can also be selected.

[0098] Step S5, judge whether the deviation exceeds the preset threshold, if yes, take the actual value of the nodal marginal price as its predicted value, and return to step S1 for closed-loop learning; if not, take the offer data generated by each agent as the final offer strategy.

[0099] In the closed-loop learning of returning to step S1, based on the actual value of the feedback nodal marginal price, the offer strategy is re-adjusted and optimized to form a new round of decision scheme. Repeat steps S1 to S5 to establish a closed-loop learning mechanism of "strategy generation-market clearing-result feedback-strategy optimization". After multiple iterations, when the deviation meets the requirements, it is determined that the system has reached a stable state, and the final optimized offer strategy is output for actual market declaration.

[0100] Embodiment two

[0101] The embodiment of the application provides a transaction decision system in a market joint clearing environment, comprising:

[0102] The offer optimization model module is configured to obtain the predicted value of the nodal marginal price, and to maximize the comprehensive income of the power generation subject, and to construct an offer optimization model of each power generation subject;

[0103] An agent design module is configured to design an agent capable of directly outputting continuous bidding for each power generation subject, and train the agent by taking the bidding optimization model as a reward function;

[0104] A joint clearing model module is configured to generate bidding data of each power generation subject based on the trained agent, and construct a joint clearing model of all power generation subjects with the minimum total cost of the whole society as the target;

[0105] A price deviation calculation module is configured to calculate the actual value of the node marginal price based on the joint clearing model, and calculate the deviation between the actual value and the predicted value of the node marginal price;

[0106] A closed-loop learning module is configured to judge whether the deviation exceeds a preset threshold, if yes, take the actual value of the node marginal price as its predicted value, and return to the bidding optimization model module for closed-loop learning; if no, take the bidding data generated by each agent as the final bidding strategy.

[0107] As shown in Figure 3 The transaction decision system can also be divided into a power generation subject bidding decision layer and a market clearing layer according to the power generation subjects and market clearing. Through artificial intelligence algorithms and iterative feedback mechanisms, the power generation subjects can maximize their profits in a complex multi-market environment. This method coordinates the selfish decisions of power generation subjects with the safe and economic operation of the power system, and ultimately guides the market to reach or approach an equilibrium state.

[0108] Embodiment three

[0109] Based on the transaction decision method provided in embodiment one, the electronic device provided in the embodiment of the present application includes a processor and a storage medium;

[0110] The storage medium is used to store instructions;

[0111] The processor is used to operate according to the instructions to perform the steps of the above method.

[0112] Embodiment four

[0113] Based on the transaction decision method provided in embodiment one, the computer readable storage medium provided in the embodiment of the present application has a computer program stored thereon, and the program is executed by the processor to realize the steps of the above method.

[0114] Embodiment five

[0115] Based on the transaction decision method provided in embodiment one, the computer program product provided in the embodiment of the present application includes a computer program / instruction, and the computer program / instruction is executed by the processor to realize the steps of the above method.

[0116] Those skilled in the art will appreciate that embodiments of the present application can be readily used as software, hardware, or a combination of software and hardware. In a software embodiment, the methods can be tangibly embodied in a machine-readable storage medium having stored thereon instructions that can be used to program a computer to perform any of the methods. The software implementation can be initialized by loading and executing a set of instructions arranged to perform one of the methods into the computer's memory. Alternatively, hard-wired circuitry can be used in place of, or in combination with, software instructions. Thus, the

[0117] The present application is described in reference to the flowchart and / or block diagrams of the methods, apparatus (systems) and computer program products according to embodiments of the application. It will be understood that each block of the flowchart and / or block diagrams, and combinations of blocks in the flowchart and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general purpose computer, special purpose computer, embedded processing device, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart and / or block diagram block or blocks. Figure 1 one or more functions specified in one or more of the flowchart or block diagram block or blocks. Figure 1 means for performing the function specified by the flowchart or block diagram block or blocks.

[0118] These computer program instructions can also be stored in a computer- readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instructions which implement the flowchart or block diagram block or blocks. Figure 1 one or more functions specified in one or more of the flowchart or block diagram block or blocks. Figure 1 means for performing the function specified by the flowchart or block diagram block or blocks.

[0119] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the flowchart or block diagram block or blocks. Figure 1 one or more functions specified in one or more of the flowchart or block diagram block or blocks. Figure 1 means for performing the function specified by the flowchart or block diagram block or blocks.

[0120] The above description is only preferred embodiments of the present application. It is obvious that those skilled in the art can make some improvements and modifications without departing from the technical principles of the present application. These improvements and modifications should also be considered as falling within the protection scope of the present application.

Claims

1. A transaction decision-making method under a market joint clearing environment, characterized in that, include: Step S1: Obtain the predicted value of the marginal electricity price of the node, and construct a bidding optimization model for each power generation entity with the goal of maximizing the comprehensive benefits of the power generation entity. Step S2: Design an agent for each power generation entity that can directly output continuous bids, and train the agent using the bid optimization model as the reward function; Step S3: Based on the trained agents, generate the bidding data of each power generation entity, and construct a joint clearing model for all power generation entities with the goal of minimizing the total cost of the whole society. Step S4: Calculate the actual value of the node marginal electricity price based on the joint clearing model, and calculate the deviation between the actual value and the predicted value of the node marginal electricity price; Step S5: Determine whether the deviation exceeds a preset threshold. If it does, use the actual value of the node marginal electricity price as its predicted value and return to step S1 for closed-loop learning. If the bid is not exceeded, the bid data generated by each of the aforementioned agents will be used as the final bid strategy.

2. The transaction decision-making method under a market joint clearing environment according to claim 1, characterized in that, The pricing optimization model is as follows: In the formula, These are the medium- and long-term electricity revenue and the base electricity revenue of the generating unit in time period t, respectively. These represent the day-ahead market revenue and real-time market revenue of the generating unit in the spot electricity market corresponding to time period t; k is the prediction curve of the marginal electricity price at node k. These represent the compensation costs and ancillary service market revenue for the generating unit during time period t, respectively. These represent the power generation cost, start-up and shutdown cost, and shutdown penalty cost of the generating unit during time period t, respectively. The generating capacity of the unit during time period t; Let T be the confidence probability of the predicted curve of the marginal electricity price of the k-th node, where T is the total number of time periods and K is the total number of nodes.

3. The transaction decision-making method under a market joint clearing environment according to claim 2, characterized in that, The revenue from day-ahead electricity bids in the aforementioned electricity spot market is: In the formula, For the electricity contracts won in the market before time period t, Let t be the predicted market node electricity price for time period t. The unit's plant power consumption rate; The revenue from real-time market-winning electricity bids in the aforementioned electricity spot market is: In the formula, For the real-time market-winning electricity during time period t, This represents the predicted real-time market node electricity price for time period t.

4. The transaction decision-making method under a market joint clearing environment according to claim 1, characterized in that, The constraints on the agent's generation decision include: The price quotation limits and physical operation constraints include upper and lower limits for unit output, unit output ramp-up rate constraints, minimum start-up and shutdown time constraints, and state of charge constraints for energy storage units.

5. The transaction decision-making method under a market joint clearing environment according to claim 1, characterized in that, The joint clearing model is as follows: In the formula, These represent the operating cost, startup cost, and power generation of the i-th unit during time period t. These represent the positive reserve bid and the awarded positive reserve capacity for the i-th generating unit during time period t, respectively; N is the total number of generating units, and T is the total number of time periods; Let $\frac{e}{t}$ be the operating cost and charge / discharge power of the $e$-th energy storage unit during time period $t$. These are the positive reserve bid and the winning positive reserve capacity for the e-th energy storage unit during time period t; This represents the total number of energy storage units. Let $\frac{p}{t}$ be the operating cost and power generation / pumping capacity of the $p$-th pumped-storage unit during time period $t$. These are the positive reserve bid and the winning positive reserve capacity of the p-th pumped-storage unit during time period t; This represents the total number of pumped storage units; The network flow constraint relaxation penalty factor is used for market clearing optimization. are the forward and reverse power flow relaxation variables for the l-th line, respectively, and NL is the total number of lines; , where are the forward and reverse current relaxation variables for the s-th cross section, and NS is the total number of cross sections.

6. The transaction decision-making method under a market joint clearing environment according to claim 1, characterized in that, The constraints for calculating the actual value of the nodal marginal electricity price based on the joint clearing model include: Constraints include power balance, unit operating characteristics, unit contracted power volume, network power flow, and ancillary service demand.

7. A transaction decision-making system under a market joint clearing environment, characterized in that, include: The pricing optimization model module is configured to obtain the predicted value of the marginal electricity price of the node, and construct the pricing optimization model of each power generation entity with the goal of maximizing the comprehensive benefits of the power generation entity. The agent design module is configured to design agents for each power generation entity that can directly output continuous bids, and train the agents using the bid optimization model as the reward function; The joint clearing model module is configured to generate bid data for each power generation entity based on the trained agents, and construct a joint clearing model for all power generation entities with the goal of minimizing the total social cost. The electricity price deviation calculation module is configured to calculate the actual value of the node marginal electricity price based on the joint clearing model, and to calculate the deviation between the actual value and the predicted value of the node marginal electricity price. The closed-loop learning module is configured to determine whether the deviation exceeds a preset threshold. If it does, the actual value of the node marginal electricity price is used as its predicted value and returned to the pricing optimization model module for closed-loop learning. If the bid is not exceeded, the bid data generated by each of the aforementioned agents will be used as the final bid strategy.

8. An electronic device, characterized in that, Including processor and storage media; The storage medium is used to store instructions; The processor is configured to operate according to the instructions to perform the steps of the method according to any one of claims 1-6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the program implements the steps of the method according to any one of claims 1-6.

10. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the method described in any one of claims 1-6.

Images