Power system transaction optimization method based on electricity-carbon cooperation

Abstract

The invention belongs to the technical field of electric power system transaction methods, and particularly discloses an electric power system transaction optimization method based on electricity-carbon cooperation. Carbon transaction costs of a load side subject and an electricity seller are considered in a model that multiple market subjects participate in an electric power market transaction; in the optimization method, the influence of active consumption of green electricity on load sides on carbon emission and economic benefits is synthesized, each load side subject and an electricity seller are regarded as agents, and an optimal electricity transaction strategy of each agent is solved through a multi-agent depth deterministic strategy gradient algorithm, so that collaborative optimization transaction of multiple types of subjects is realized. The power system transaction optimization method provided by the invention can promote cooperative operation and development of a power market and a carbon market, stimulate green power consumption, provide an effective optimization strategy for power market transactions participated by multiple types of load side subjects, and improve the low-carbon operation efficiency of the power market.

Description

Technical Field

[0001] The present invention belongs to the technical field of power system trading methods, and specifically relates to a power system trading optimization method based on electricity-carbon synergy. Background Art

[0002] The large-scale access of new load-side entities, such as electric vehicles and distributed energy producers and consumers, to the power grid presents significant differences in power consumption patterns and response characteristics, placing new demands on the operating rules and trading models of traditional power markets. Against this backdrop, establishing a power trading mechanism that integrates economic efficiency with carbon emission reduction benefits is crucial for guiding load-side entities to actively participate in market operations and enhancing the overall low-carbonization of the power system.

[0003] There has been some exploration into the coordinated operation of electricity and carbon markets. Existing technologies can help market players optimize carbon emission strategies through the design of a linkage between the two markets, fostering a virtuous interaction between energy consumption and carbon emission reduction. However, existing research has largely focused on analysis at the power generation and transmission and distribution grid levels, neglecting indirect carbon emission behaviors on the load side and the heterogeneous nature of electricity. It also fails to consider the impact of active green electricity consumption on the load side on carbon emissions and its own economic benefits. Furthermore, current model construction is often limited to a single type of entity, lacking a systematic approach to optimizing power transactions for the coordinated participation of diverse load-side entities in the electricity-carbon market. Summary of the Invention

[0004] The problem to be solved by the present invention is to overcome the defects of the power system transaction optimization method in the existing technology that is limited to a single subject perspective and cannot achieve efficient electricity-carbon collaborative optimization, thereby providing a power system transaction optimization method based on electricity-carbon collaboration.

[0005] A power system transaction optimization method based on electricity-carbon synergy includes the following steps: Obtaining an initial data set for power trading, the data set comprising: main grid power price information, retailer power price information, load-side entity power demand information, carbon trading prices, and thermal power carbon emission factors; the load-side entity power demand information comprising the load-side entity's green power demand and thermal power demand; Constructing a multi-market participant participation model for electricity market transactions, including: an objective function of a load-side participant and an objective function of an electricity seller; the objective functions of the load-side participant and the electricity seller include carbon trading costs; Each load-side entity and electricity seller is regarded as an intelligent agent, the initial electricity trading data set is used as the initial environmental state, and the multi-market entity participation in the electricity market trading model is used as the reward set. The optimal electricity trading strategy of each intelligent agent is solved through a multi-agent deep deterministic policy gradient algorithm.

[0006] Furthermore, the initial data set for power trading includes: main grid power price information, electricity price information of electricity sellers, load-side main power demand information, carbon trading price, and thermal power carbon emission factor; The main grid power price information includes: The price of purchasing green electricity from electricity sellers during the period, the main grid The price of thermal power sold to electricity sellers during the period, the main grid The price of green electricity sold to electricity sellers during the period; The electricity price information of the electricity seller includes: The price of thermal power sold to the load side during the period, and the electricity sales The price of selling green electricity to the load side and the electricity seller The price of green electricity purchased from the load side during the period; The load side main body power demand information includes: residential load Time period electricity demand information, prosumers Time period power demand information, electric vehicle load Power demand information by time period.

[0007] Furthermore, the resident load The electricity demand information during the period includes: Residential load Thermal power demand and residential load during the period Green electricity demand during the time period; The prosumer load The power demand information for each period includes: Thermal power demand and prosumer load during the period Green electricity demand during the time period; The electric vehicle load The power demand information during the period includes: electric vehicle load Thermal power demand and electric vehicle load during the period Green electricity demand during the period.

[0008] Furthermore, the carbon trading cost of the load-side entity is: ; ; in, represents the carbon trading price, Indicates the actual carbon emissions, represents the allocated free carbon emission allowance, represents the carbon emission factor of thermal power generation, Indicates the purchase amount of thermal power.

[0009] Furthermore, the objective function of the resident load in the load side is expressed as: ; in, Indicates residential load In the period The total benefit, Indicates residential load In the period The utility function of Indicates residential load The electricity cost, Indicates residential load In the period carbon trading costs; The objective function of the prosumer load in the load-side entity is expressed as: ; in, Prosumers In the period The total benefit, Prosumers In the period The electricity utility function is Prosumers With the retailer Income from session trading, Prosumers In the period The cost of power generation, Prosumers In the period carbon trading costs; The objective function of the electric vehicle load in the load side is expressed as: ; in, Indicates electric vehicle load In the period The total benefit, Indicates electric vehicle load In the period The utility function of Indicates electric vehicle load In the period The charging cost, Indicates electric vehicle load In the period carbon trading costs.

[0010] The objective function of the retailer is expressed as: ; in, express The total profit of the time-slot seller, express The transaction income between electricity sellers and load side during the period, express Transaction revenue between time-slot sellers and the main network.

[0011] Furthermore, in the multi-agent deep deterministic policy gradient algorithm, the environment state Expressed as: ; ; ; ; ; in, represents the observation of resident load, Indicates residential load In the period The total electricity demand, and Represents electricity sellers Thermal power prices and green power prices during the time period; Represents the observation of prosumers, and Producers and consumers The power generation and total electricity demand during the period, and Represents electricity sellers Thermal power sales price, green power sales price and power purchase price during the time period; represents the observation of electric vehicle load, Indicates electric vehicle load The total electricity demand during the period, and Represents electricity sellers Thermal power prices and green power prices during the time period; Represents the observation of e-sellers, Represents the residential load In the period thermal power and green power demand, prosumers In the period Thermal power and green power demand, electric vehicle load In the period The demand for thermal power and green electricity, Represents the main network respectively Green electricity purchase price, thermal power sales price, and green electricity sales price during the period.

[0012] Furthermore, in the multi-agent deep deterministic policy gradient algorithm, the action space of the agent Expressed as: ; ; ; ; ; in, Indicates the action of the resident load, and Represents the residential load In the period The demand for thermal power and green electricity; Indicates the actions of prosumers, and Represents the prosumer load In the period The demand for thermal power and green electricity; Indicates the action of electric vehicle load, and Represents electric vehicle load In the period Demand for thermal power and green electricity; Indicates the actions of the seller. and Represents electricity sellers The thermal power sales price, green power sales price and electricity purchase price during the period.

[0013] A computer device comprises: a processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor. When the computer device is running, the processor and the memory communicate via the bus, and when the machine-readable instructions are executed by the processor, the steps of the above-mentioned power system transaction optimization method are performed.

[0014] A computer-readable storage medium stores a computer program, which, when executed by a processor, executes the steps of the above-mentioned power system transaction optimization method.

[0015] Beneficial effects: The present invention discloses a power system transaction optimization method based on electricity-carbon synergy, which considers the carbon transaction costs of load-side entities and electricity sellers in the power market transaction model with multiple market entities. In the optimization method, the impact of the load-side active consumption of green electricity on carbon emissions and its own economic benefits is comprehensively considered, and each load-side entity and electricity seller is regarded as an intelligent agent. The optimal power transaction strategy of each intelligent agent is solved through a multi-agent deep deterministic policy gradient algorithm, thereby realizing collaborative optimization transactions of multiple types of entities. The power system transaction optimization method provided by the present invention can promote the coordinated operation and development of the power market and the carbon market, stimulate green electricity consumption, and provide effective optimization strategies for power market transactions involving multiple types of load-side entities, thereby improving the low-carbon operation efficiency of the power market. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0017] Figure 1 It is a flow chart of the power system transaction optimization method of the present invention. DETAILED DESCRIPTION

[0018] To make the above-mentioned objects, features, and advantages of the present application more clearly understood, the specific embodiments of the present application are described in detail below with reference to the accompanying drawings. The following description sets forth many specific details to facilitate a full understanding of the present application. However, the present application can be implemented in many other ways than those described herein, and those skilled in the art can make similar improvements without violating the scope of the present application. Therefore, the present application is not limited to the specific embodiments disclosed below.

[0019] Reference Figure 1 As shown, this embodiment provides a power system transaction optimization method based on electricity-carbon synergy, including the following steps: Step S1: Obtaining an initial data set for power trading, which includes: main grid power price information, retailer power price information, load-side entity power demand information, carbon trading price, and thermal power carbon emission factor; the load-side entity power demand information includes the load-side entity's green power demand and thermal power demand; Step S2: Constructing a model for multiple market entities to participate in electricity market transactions, including: an objective function of the load-side entity and an objective function of the electricity seller; the objective functions of the load-side entity and the electricity seller include carbon trading costs; Step S3: Consider each load-side entity and electricity seller as an intelligent agent, the initial data set of electricity trading as the initial environmental state, and the model of multiple market entities participating in electricity market trading as the reward set, and solve the optimal electricity trading strategy of each intelligent agent through a multi-agent deep deterministic policy gradient algorithm.

[0020] Specifically, in step S1, the initial data set of power transaction Including: main grid electricity price information , electricity price information of electricity retailers , load-side main power demand information , carbon trading price, thermal power carbon emission factor; The main grid electricity price information Including mainnet Electricity price information for each period, including: Green electricity purchase price during the period , Mainnet Thermal power sales price during the period , Mainnet Green electricity sales price during the period ; The electricity price information of the electricity seller Including retailer Electricity price information for the time period, including: electricity sellers Thermal power price during the period , retail stores Green electricity price during the period , retail stores Electricity purchase price during the period ; The load side main body power demand information Contains power demand information of three types of load-side entities, including: residential load Power demand information by time period , Prosumers Power demand information by time period , electric vehicle load Power demand information by time period .

[0021] Specifically, the residential load The electricity demand information during the period includes: Residential load In the period Thermal power demand , Residential load In the period Green electricity demand ; The prosumer The electricity demand information for each period includes: In the period Thermal power demand , Prosumers In the period Green electricity demand ; The electric vehicle load The power demand information during the period includes: electric vehicle load In the period Thermal power demand , electric vehicle load In the period Green electricity demand .

[0022] Specifically, in step S2, considering carbon trading costs not only promotes green electricity consumption but also facilitates more direct user participation in the carbon market, thereby promoting the achievement of carbon emission reduction targets. In this embodiment, the electricity market includes both thermal power and green power, and load-side entities can choose to purchase electricity with different attributes based on their needs.

[0023] The carbon trading cost of the load-side entity is: ; ; in, represents the carbon trading price, Indicates the actual carbon emissions, represents the allocated free carbon emission allowance, represents the carbon emission factor of thermal power generation, Indicates the purchase amount of thermal power.

[0024] For the resident load in the load side, the objective function is expressed as: ; in, Indicates residential load In the period The total benefit, Indicates residential load In the period The utility function of Indicates residential load The electricity cost, Indicates residential load In the period carbon trading costs.

[0025] Specifically, Residential load The utility function Expressed as: ; in, Represents a user In the period The total electricity consumption, , Indicates residential load In the period The demand for thermal power, Indicates residential load In the period Green electricity demand; Indicates residential load Initial electricity consumption intention; is a fixed constant.

[0026] Residential load The electricity cost is expressed as: ; in Representative of retailer The thermal power price during the period, Representative of retailer The price of green electricity during the period.

[0027] Carbon trading costs are the cost of residential loads The cost or income generated by carbon trading on the carbon trading platform. In this embodiment, under the green electricity carbon emission offset mechanism, the green electricity price has both environmental value and electric energy value, which can be used to offset carbon emissions. Electricity carbon emission quota , indirect carbon emissions from actual electricity consumption and carbon trading costs Expressed as: ; ; ; in, Indicates residential load The carbon trading costs, Indicates residential load carbon trading revenue; Indicates the carbon emission quota per unit of electric power.

[0028] As a preference of this embodiment, the resident load needs to meet the following constraints: ; in, Indicates residential load In the period The minimum power requirement, Indicates residential load In the period The maximum power demand.

[0029] The objective function of the prosumer load in the load-side entity is expressed as: ; in, Prosumers In the period The total benefit, Prosumers In the period The electricity utility function is Prosumers With the retailer Income from session trading, Prosumers In the period The cost of power generation, Prosumers In the period carbon trading costs.

[0030] Specifically, prosumers Electricity utility function , which can be expressed as: ; in, Prosumers In the period The total electricity consumption, , Prosumers In the period The demand for thermal power, Prosumers In the period Green electricity demand; Prosumers Initial electricity consumption intention.

[0031] Prosumers With the retailer Income from session trading Expressed as: ; in, , Prosumers In the period Power generation; Represents e-commerce sellers The electricity purchase price during the period.

[0032] Prosumers The cost of power generation Expressed as: ; in, Prosumers The power generation cost coefficient.

[0033] Prosumers Carbon trading costs Expressed as: ; ; ; in, Prosumers exist Initial carbon quota for the period, Prosumers exist The actual carbon emissions from electricity consumption during the period.

[0034] As a preferred embodiment of this invention, the producer and consumer The electricity consumption must also meet the basic demand and maximum demand constraints: .

[0035] in, Prosumers In the period The minimum power requirement, Prosumers In the period The maximum power demand.

[0036] The objective function of the electric vehicle load in the load side is expressed as: ; in, Indicates electric vehicle load In the period The total benefit, Indicates electric vehicle load In the period The utility function of Indicates electric vehicle load In the period The charging cost, Indicates electric vehicle load In the period carbon trading costs.

[0037] Specifically, electric vehicle load The utility function The formula is: ; in, Indicates electric vehicle load In the period The total electricity consumption, , Indicates electric vehicle load In the period The demand for thermal power, Indicates electric vehicle load In the period Green electricity demand; Prosumers Electric vehicle load Initial electricity consumption intention.

[0038] Electric vehicle load In the period Charging cost The formula is: ; Electric vehicle load In the period Carbon trading costs In this embodiment, the baseline method is adopted, with the carbon emissions of traditional fuel vehicles as the benchmark, and the difference in carbon emissions between electric vehicles and traditional fuel vehicles under the same driving distance is used as the initial carbon quota obtained by the car owner.

[0039] Therefore, electric vehicle load In the period Carbon trading costs Expressed as: ; ; ; in, express Electric vehicle load during the period Initial carbon allowances held; Indicates electric vehicle load Mileage per unit of electricity; Indicates the load of electric vehicles The corresponding carbon emissions generated per unit of driving distance of traditional fuel vehicles; express Electric vehicle load during the period The actual carbon emissions generated by charging.

[0040] As a preferred embodiment of this invention, electric vehicle load The following constraints must also be met: ; in, Indicates electric vehicle load In the period Maximum power demand; Indicates electric vehicle load In the period The minimum power requirement.

[0041] The objective function of the retailer is expressed as: ; in, express The total profit of the time-slot seller, express The transaction income between electricity sellers and load side during the period, express Transaction revenue between time-slot sellers and the main network.

[0042] Specifically, Transaction revenue between electricity sellers and load-side during the period Expressed as: ; in, represents the number of residential loads, represents the number of electric vehicle loads, , It refers to the prosumer whose power generation is less than the green electricity demand. Refers to prosumers whose electricity generation is greater than their green electricity demand.

[0043] express Transaction revenue between time slot sellers and the main network: ; ; in, Indicates the green electricity transaction volume between electricity retailers and the main grid; Indicates the main network The purchase price of green electricity during the period, and Represents the main network respectively The thermal power sales price and green power sales price during the period.

[0044] As a preferred embodiment of this invention, the transaction price between the electricity seller and the load side satisfies the following constraints: ; in, and They represent the lowest price and the highest price that electricity sellers sell thermal power to load side respectively. and They represent the lowest price and the highest price of green electricity sold by electricity sellers to load side respectively; and They respectively represent the minimum and maximum prices for electricity sellers to purchase green electricity from the load side.

[0045] In step S3, each load-side entity and electricity seller is regarded as an intelligent agent, the initial electricity trading data set is used as the initial environmental state, and the multi-market entity participation in the electricity market trading model is used as the reward set. The optimal electricity trading strategy of each intelligent agent is solved through the multi-agent deep deterministic policy gradient algorithm (MADDPG).

[0046] Step S3.1: Initialize the critic network and actor network parameters of each agent; Step S3.2: Initialize the environment state, the environment state consists of each agent's observations, Environmental status Expressed as: ; ; ; ; ; in, represents the observation of resident load, Indicates residential load In the period The total electricity demand, and Represents electricity sellers Thermal power prices and green power prices during the time period; Represents the observation of prosumers, and Producers and consumers The power generation and total electricity demand during the period, and Represents electricity sellers Thermal power sales price, green power sales price and power purchase price during the time period; represents the observation of electric vehicle load, Indicates electric vehicle load The total electricity demand during the period, and Represents electricity sellers Thermal power prices and green power prices during the time period; Represents the observation of e-sellers, Represents the residential load In the period thermal power and green power demand, prosumers In the period Thermal power and green power demand, electric vehicle load In the period The demand for thermal power and green electricity, Represents the main network respectively Green electricity purchase price, thermal power sales price, and green electricity sales price during the period.

[0047] Step 3.3: The electricity supplier agent and the three types of load-side agents select actions and make decisions in turn. The action space of the agent Expressed as: ; ; ; ; ; in, Indicates the action of the resident load, and Represents the residential load In the period The demand for thermal power and green electricity; Indicates the actions of prosumers, and Represents the prosumer load In the period The demand for thermal power and green electricity; Indicates the action of electric vehicle load, and Represents electric vehicle load In the period Demand for thermal power and green electricity; Indicates the actions of the seller. and Represents electricity sellers The thermal power sales price, green power sales price and electricity purchase price during the period.

[0048] Step S3.4: Get the new environment state based on the action selected by the agent and reward collection , reward collection That is the objective function value of the electricity seller and each load-side entity, which can be expressed as follows: ; Step 3.5: Update the environment state and return to step 3.3. Continue iterating until the actions selected by each agent no longer change, and obtain the optimal power trading decision for each agent.

[0049] Thus, the optimal electricity trading strategy of electricity sellers and three types of load-side entities is solved by the multi-agent deep deterministic policy gradient algorithm.

[0050] This embodiment also provides a computer device, including: a processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor. When the computer device is running, the processor and the memory communicate through the bus, and when the machine-readable instructions are executed by the processor, the steps of the above-mentioned power system transaction optimization method are performed.

[0051] This embodiment further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the above-mentioned power system transaction optimization method are executed.

[0052] This embodiment provides a power system transaction optimization method based on electricity-carbon collaboration. In the power market transaction model with multiple market entities, the carbon transaction costs of the load-side entities and electricity sellers are considered. In the optimization method, the impact of the active consumption of green electricity on carbon emissions and its own economic benefits by the load-side entity is comprehensively considered. Each load-side entity and electricity seller is regarded as an intelligent agent. The optimal power transaction strategy of each intelligent agent is solved through a multi-agent deep deterministic policy gradient algorithm, thereby realizing collaborative optimization transactions of multiple types of entities. The power system transaction optimization method provided by the present invention can promote the coordinated operation and development of the power market and the carbon market, stimulate green electricity consumption, and provide effective optimization strategies for power market transactions involving multiple types of load-side entities, thereby improving the low-carbon operation efficiency of the power market.

[0053] The technical features of the above-mentioned embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the technical features in the above-mentioned embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0054] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the patent application. It should be noted that a person of ordinary skill in the art may make various modifications and improvements without departing from the spirit of the present application, and these modifications and improvements fall within the scope of protection of the present application. Therefore, the scope of protection of the present patent application shall be determined by the appended claims.

Claims

1. A power system transaction optimization method based on electricity-carbon synergy, characterized in that: The following steps are involved: Obtaining an initial data set for power trading, the data set comprising: main grid power price information, retailer power price information, load-side entity power demand information, carbon trading prices, and thermal power carbon emission factors; the load-side entity power demand information comprising the load-side entity's green power demand and thermal power demand; Constructing a multi-market participant participation model for electricity market transactions, including: an objective function of a load-side participant and an objective function of an electricity seller; the objective functions of the load-side participant and the electricity seller include carbon trading costs; Each load-side entity and electricity seller is regarded as an intelligent agent, the initial electricity trading data set is used as the initial environmental state, and the multi-market entity participation in the electricity market trading model is used as the reward set. The optimal electricity trading strategy of each intelligent agent is solved through a multi-agent deep deterministic policy gradient algorithm.

2. The power system transaction optimization method based on electricity-carbon synergy according to claim 1 is characterized in that: The initial data set for power trading includes: main grid power price information, electricity price information for electricity sellers, load-side power demand information, carbon trading price, and thermal power carbon emission factor; The main grid power price information includes: The price of purchasing green electricity from electricity sellers during the period, the main grid The price of thermal power sold to electricity sellers during the period, the main grid The price of green electricity sold to electricity sellers during the period; The electricity price information of the electricity seller includes: The price of thermal power sold to the load side during the period, and the electricity sales The price of selling green electricity to the load side and the electricity seller The price of green electricity purchased from the load side during the period; The load side main body power demand information includes: residential load Time period power demand information, prosumer load Time period power demand information, electric vehicle load Power demand information by time period.

3. The power system transaction optimization method based on electricity-carbon synergy according to claim 2 is characterized in that: The residential load The electricity demand information during the period includes: Residential load Thermal power demand and residential load during the period Green electricity demand during the time period; The prosumer load The power demand information for each period includes: Thermal power demand and prosumer load during the period Green electricity demand during the time period; The electric vehicle load The power demand information during the period includes: electric vehicle load Thermal power demand and electric vehicle load during the period Green electricity demand during the period.

4. The power system transaction optimization method based on electricity-carbon synergy according to claim 1 is characterized in that: The carbon trading cost of the load-side entity is: ； ； in, represents the carbon trading price, Indicates the actual carbon emissions, represents the allocated free carbon emission allowance, represents the carbon emission factor of thermal power generation, Indicates the purchase amount of thermal power.

5. The power system transaction optimization method based on electricity-carbon synergy according to claim 1 is characterized in that: The objective function of the resident load in the load side is expressed as: ； in, Indicates residential load In the period The total benefit, Indicates residential load In the period The utility function of Indicates residential load The electricity cost, Indicates residential load In the period carbon trading costs; The objective function of the prosumer load in the load-side entity is expressed as: ； in, Prosumers In the period The total benefit, Prosumers In the period The electricity utility function is Prosumers With the retailer Income from session trading, Prosumers In the period The cost of power generation, Prosumers In the period carbon trading costs; The objective function of the electric vehicle load in the load side is expressed as: ； in, Indicates electric vehicle load In the period The total benefit, Indicates electric vehicle load In the period The utility function of Indicates electric vehicle load In the period The charging cost, Indicates electric vehicle load In the period carbon trading costs.

6. The power system transaction optimization method based on electricity-carbon synergy according to claim 1 is characterized in that: The objective function of the retailer is expressed as: ； in, express The total profit of the time-slot seller, express The transaction income between electricity sellers and load side during the period, express Transaction revenue between time-slot sellers and the main network.

7. The power system transaction optimization method based on electricity-carbon synergy according to claim 1 is characterized in that: In the multi-agent deep deterministic policy gradient algorithm, the environment state Expressed as: ； ； ； ； ； in, represents the observation of resident load, Indicates residential load In the period The total electricity demand, and Represents electricity sellers Thermal power prices and green power prices during the time period; Represents the observation of prosumers, and Producers and consumers The power generation and total electricity demand during the period, and Represents electricity sellers Thermal power sales price, green power sales price and power purchase price during the time period; represents the observation of electric vehicle load, Indicates electric vehicle load The total electricity demand during the period, and Represents electricity sellers Thermal power prices and green power prices during the time period; Represents the observation of e-sellers, Represents the residential load In the period thermal power and green power demand, prosumers In the period Thermal power and green power demand, electric vehicle load In the period The demand for thermal power and green electricity, Represents the main network respectively Green electricity purchase price, thermal power sales price, and green electricity sales price during the period.

8. The power system transaction optimization method based on electricity-carbon synergy according to claim 1 is characterized in that: In the multi-agent deep deterministic policy gradient algorithm, the action space of the agent is Expressed as: ； ； ； ； ； in, Indicates the action of the resident load, and Represents the residential load In the period The demand for thermal power and green electricity; Indicates the actions of prosumers, and Represents the prosumer load In the period The demand for thermal power and green power; Indicates the action of electric vehicle load, and Represents electric vehicle load In the period Demand for thermal power and green electricity; Indicates the actions of the seller. and Represents electricity sellers The thermal power sales price, green power sales price and electricity purchase price during the period.

9. A computer device, characterized in that: include: A processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the computer device is running, the processor and the memory communicate via the bus, and when the machine-readable instructions are executed by the processor, the steps of the power system transaction optimization method according to any one of claims 1 to 4 are performed.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, which, when executed by a processor, executes the steps of the power system transaction optimization method according to any one of claims 1 to 4.

Images