An electric-gas integrated energy system optimal operation method

Abstract

The present application relates to a kind of electro-gas integrated energy system operation optimization method, including input optimization period parameter, establish proton exchange membrane electrolytic cell model and its start-stop constraint and operation constraint, establish hydrogen fuel cell model and its operation constraint, establish hydrogen storage tank model and its charge-discharge power constraint, establish methane reactor model and its operation constraint, establish gas turbine model and its operation constraint, establish the objective function with minimum integrated energy system cost as optimization target, establish the electric power balance constraint of system, gas power balance constraint, establish power interaction constraint of grid, gas network and upper main network, and decision-making step is carried out by setting agent using deep Q network algorithm.Full consideration hydrogen energy's dispatchable capacity and heat energy's utilization, can realize the efficient use of energy, effectively improve the operation efficiency and operation economy of integrated energy system, reduce system operation cost.

Description

Technical Field

[0001] This invention relates to the field of integrated energy technology, and in particular to a method for optimizing the operation of an integrated electric-gas energy system. Background Technology

[0002] In recent years, due to the increasing demand for low-carbon and clean energy in national economic and social development, various green, clean, and renewable energy sources have continued to develop and gradually become the main source of national energy demand. However, with the continuous expansion of the proportion of renewable energy, the supply-demand imbalance caused by its inherent strong uncertainty and volatility will seriously affect the operation and control of the power system, such as the curtailment of wind and solar power. Therefore, the concept of integrated energy systems provides a new solution to the above problems. By realizing the diversified conversion of energy supply through subsystems of various energy forms, it is conducive to improving the operational stability and economy of each system. Moreover, with the continuous development of power-to-gas (P2G) technology and the continuous increase in the installed capacity of gas turbines, the deep coupling of the power system and the natural gas network, and the mutual conversion of electricity with hydrogen, natural gas, and other forms of energy, will help strengthen the construction of integrated energy systems. Research on the optimized operation of integrated energy systems will also become a hot topic in the future.

[0003] Power-to-Generation (P2G), as a crucial link connecting the power and natural gas systems, is a key area of ​​research in integrated power-gas energy systems. While some scholars have conducted research and summarized the P2G process, most studies model it directly as an electrical load and represent the conversion process through conversion efficiency, without fully considering the hydrogen production stage and the utilization of hydrogen energy. This model neglects the coupling between electrical, hydrogen, and natural gas energy. Furthermore, studies considering the intermediate process of power-to-hydrogen (P2H) have largely failed to provide detailed modeling of the electrolyzer as a critical conversion device in the P2H stage. In actual operation, the electrolyzer's operating state is affected by input power and ambient temperature. Considering that some energy is converted into heat during the conversion process, it is necessary to consider the electrolyzer's heat generation characteristics and fully utilize waste heat for detailed modeling. Existing research has not comprehensively addressed the coupling issues among various energy sources in the P2G process.

[0004] In an integrated electric-gas energy system, P2G equipment plays a crucial role in the coordinated operation of the electric and gas networks. By fully considering the hydrogen and heat production characteristics of P2G equipment during operation, and through the rational scheduling and utilization of hydrogen and heat energy, the economic efficiency of the integrated energy system can be improved, which is of great significance for building a clean and environmentally friendly system. Summary of the Invention

[0005] To address the aforementioned issues, this invention proposes an optimized operation method for an integrated electric-gas energy system, aiming to improve the operating efficiency and economic viability of the integrated energy system while reducing its operating costs.

[0006] To achieve the above objectives, the present invention proposes an optimized operation method for an integrated electric-gas energy system, comprising the following steps:

[0007] 1) Input information on wind turbines, photovoltaic units, electricity load, gas load, proton exchange membrane electrolyzer, heat exchanger, hydrogen fuel cell, hydrogen storage tank model, methane reactor and gas turbine parameters within the optimization period;

[0008] 2) Establish a proton exchange membrane electrolyzer model and its start-up and shutdown constraints and operating constraints; establish a hydrogen fuel cell model and its operating constraints; establish a hydrogen storage tank model and its charging and discharging power constraints; establish a methane reactor model and its operating constraints; and establish a gas turbine model and its operating constraints.

[0009] 3) Establish an objective function with the goal of minimizing the cost of the integrated energy system;

[0010] 4) Establish power balance constraints for the integrated energy system, power balance constraints for the gas grid, power grid and the upper-level main grid;

[0011] 5) Establish an optimization operation framework based on deep Q-networks, construct the state space, action space, action strategy and reward space of the integrated energy system, improve data utilization through experience replay pool, and obtain the optimal operation scheme through Q target network and Q estimation network.

[0012] This invention establishes models and constraints for each energy conversion device, with the goal of minimizing the cost of the integrated energy system. It obtains the optimal operating scheme through a deep Q-network optimization method, fully considering the dispatchability of hydrogen energy and the utilization of thermal energy. This enables efficient energy utilization, effectively improves the operating efficiency and economy of the integrated energy system, and reduces system operating costs. Attached Figure Description

[0013] Figure 1 This is a schematic diagram of the structure of an integrated electric-gas energy system;

[0014] Figure 2 Optimization framework diagram for deep Q-networks;

[0015] Figure 3 A flowchart for optimizing the operation of an integrated electric-gas energy system. Detailed implementation method:

[0016] It should be noted that the following detailed description is illustrative and intended to provide further explanation of the invention. Unless otherwise specified, all technical and scientific terms used in this invention have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains.

[0017] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of exemplary embodiments according to the invention. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0018] The structure of the electric-gas integrated energy system of this invention is as follows: Figure 1 As shown, the system includes a proton exchange membrane electrolyzer, a heat exchanger, a hydrogen fuel cell, a hydrogen storage tank, a methane reactor, and a gas turbine. The proton exchange membrane electrolyzer is connected to the upstream power grid, wind turbines, and photovoltaic (PV) units. Electrical energy from these units is input into the proton exchange membrane electrolyzer, where it undergoes electrolysis, converting electrical energy into hydrogen and heat energy for output. The heat energy output from the proton exchange membrane electrolyzer enters the heat exchanger. Part of the heat energy from the heat exchanger is used by the proton exchange membrane electrolyzer to maintain the temperature required for normal operation, and the other part is supplied to the heating network. Part of the hydrogen energy output from the proton exchange membrane electrolyzer enters the hydrogen fuel cell, part is stored in the hydrogen storage tank to provide time-delayed energy for system operation, and the other part enters the methane reactor. The hydrogen fuel cell uses the hydrogen energy output from the proton exchange membrane electrolyzer to generate electrical energy and heat energy. The electrical energy is supplied to the electrical load through energy output, and the heat energy is supplied to the thermal load through the heat exchanger and meets its own heat requirements. The hydrogen stored in the hydrogen storage tank is used partly for the methane reactor, which can then be utilized by the gas grid, and partly for the hydrogen fuel cell, which can then be used by the grid, providing power support for both networks. The methane reactor uses hydrogen supplied by the proton exchange membrane electrolyzer and the hydrogen storage tank for methanation, providing gas energy to the upstream gas grid's natural gas system. The gas turbine is directly connected to both the gas grid and the grid; when purchasing energy from the gas grid is cheaper, supplying power to the grid via the gas turbine can be considered.

[0019] See Figure 3 The optimized operation method of the integrated electric-gas energy system includes the following steps.

[0020] 1) Input the information on wind turbines, photovoltaic units, electricity load, and gas load during the optimization period, as well as the parameter information of proton exchange membrane electrolyzers, heat exchangers, hydrogen fuel cells, hydrogen storage tank models, methane reactors, and gas turbines. Specific information is described in detail in the following steps.

[0021] 2) Establish a proton exchange membrane electrolyzer model and its start-up and shutdown constraints and operating constraints, establish a hydrogen fuel cell model and its operating constraints, establish a hydrogen storage tank model and its charging and discharging power constraints, establish a methane reactor model and its operating constraints, and establish a gas turbine model and its operating constraints.

[0022] (1) Establishing a proton exchange membrane electrolyzer model and its start-up and shutdown constraints and operating constraints.

[0023] The operation of a proton exchange membrane electrolyzer can be divided into an electrolysis process and a heat transfer process. During electrolysis, the electrical energy input to the electrolyzer is converted into hydrogen energy and heat energy. The input AC power P of the electrolyzer is determined according to the electrochemical process. ec Actual input power Output hydrogen power and output heat power They are respectively:

[0024]

[0025]

[0026]

[0027] In the formula: The AC / DC conversion efficiency of the electrolytic cell; i ec U is the current in the electrolytic cell; ec (i ec T ec Let be the voltage function of the electrolytic cell, which is related to the current i. ec and temperature T ec The function, T ec，t U represents the temperature of the electrolytic cell at time t. tn (T ec ) is a thermally neutral voltage function, which is related to temperature T. ec related; These represent the hydrogen production and heat generation efficiencies of the electrolyzer, respectively.

[0028] Linearizing the relationship between the electrolyzer's output hydrogen power, thermal power, input AC power, and temperature yields the following expression:

[0029]

[0030]

[0031] In the formula: μ1, υ1, μ2, and υ2 are coefficients after linearizing the operating region of the electrolytic cell; σ ec，t P is a 0-1 variable representing the operating state of the electrolytic cell at time t; ec，t Let be the input AC power of the electrolytic cell at time t.

[0032] During the heat transfer process, some of the heat generated by the electrolyzer is lost, while the rest enters the heat exchanger. Of the heat entering the heat exchanger, a portion is supplied to the proton exchange membrane electrolyzer to maintain the temperature required for normal operation, and the other portion is transferred to the heating network to supply the heat load. The heat output power of the heat exchanger is then... It can be represented as:

[0033]

[0034] In the formula: For lost heat energy; η represents the actual heat energy input to the heat exchanger. he The heat exchange efficiency of the heat exchanger.

[0035] The temperature change of the electrolytic cell during heat transfer is determined by the following steady-state model:

[0036]

[0037]

[0038] In the formula: C ec T represents the lumped heat capacity of the electrolytic cell. out，t R represents the ambient temperature. ec This represents the lumped thermal resistance of the electrolytic cell.

[0039] The start-up and shutdown constraints for the proton exchange membrane electrolyzer are:

[0040]

[0041] In the formula: These represent the start-up and shutdown actions of the electrolytic cell, respectively. These represent the maximum daily start-up and shutdown limits for the electrolytic cell; These represent the shortest working and downtime of the electrolytic cell, respectively; the last constraint ensures that the working state of the electrolytic cell is consistent at the end and beginning of each day.

[0042] The operating constraints of the proton exchange membrane electrolyzer are:

[0043]

[0044] In the formula: C cp，ec This refers to the installation capacity of the electrolytic cell; These are the upper and lower limits of the electrolytic cell load rate, respectively. This represents the maximum ramping power of the electrolytic cell; These are the upper and lower limits of the operating temperature of the electrolytic cell.

[0045] To ensure full energy utilization in the P2G process, a refined model of the proton exchange membrane electrolyzer is constructed. During the electro-to-gas process, the electrolyzer inputs electrical energy and outputs hydrogen and heat. Hydrogen can be stored in a hydrogen storage tank to provide time-transferable energy for system operation, or it can be directly converted into electrical energy through a hydrogen fuel cell. Since the electrolyzer requires a certain temperature for normal operation, the generated heat energy is fully considered for utilization within the electrolyzer itself. A steady-state model of the electrolyzer's temperature change is established, and the remaining heat energy is used to supply the heating load through a heat exchanger to improve operational economy.

[0046] (2) Establishing a hydrogen fuel cell model and its operational constraints

[0047] After hydrogen is produced through the electrolysis of water in a proton exchange membrane electrolyzer, a hydrogen fuel cell can use hydrogen energy to generate electricity and heat, supplying electrical and thermal loads and meeting its own heat requirements. Compared to gas turbines, hydrogen fuel cells can directly convert hydrogen energy into electricity, eliminating the need for methanation of hydrogen before supplying it to a gas turbine, thus significantly improving the energy efficiency of the integrated energy system.

[0048] The model for a hydrogen fuel cell is as follows:

[0049]

[0050] The operating constraints of hydrogen fuel cells are:

[0051]

[0052] In the formula: P HFC，t This refers to the output electrical power of the fuel cell; For electrical conversion efficiency; Q represents the hydrogen power input to the fuel cell. HFC，t The output thermal power of the fuel cell; The thermal conversion coefficient; and These are the upper and lower limits of the electro-thermal conversion coefficients, respectively; k e，max k h，max These are the slopes corresponding to the operating boundaries of the highest electrical and thermal efficiencies, respectively. η he and These represent the thermal power supplied by the fuel cell to the heat load, the heat exchanger conversion efficiency, and the thermal power entering the heat exchanger at time t, respectively; T HFC，t T out，t These are the fuel cell temperature and the outdoor temperature, respectively. σ represents the lost heat power; HFC，t The variable is 0-1, representing the operating state of the fuel cell at time t; These represent the upper and lower limits of the fuel cell load rate, respectively; Cap，HFC For the installed capacity of fuel cells; This represents the maximum ramp power of the fuel cell; These are the upper and lower limits of the fuel cell operating temperature, respectively.

[0053] (3) Establishing a hydrogen storage tank model and its charging and discharging power constraints

[0054] Similar to battery energy storage, hydrogen storage tanks can provide a stable and time-shiftable scheduling resource for processes and equipment such as hydrogen fuel cells and methanation by storing hydrogen.

[0055] The model for the hydrogen storage tank is as follows:

[0056]

[0057] In the formula: E hs，0 E hs，t C represents the initial capacity of the hydrogen storage tank at time t; cp，hc The installation capacity of the hydrogen storage tank; The self-discharge rate of the hydrogen storage tank; These are the charging and discharging efficiencies of the hydrogen storage tank, respectively. and These represent the input power and output power of the hydrogen storage tank at time t, respectively.

[0058] Δt is the time interval; These are the upper and lower limits of hydrogen storage tank capacity constraints, respectively. These are 0-1 variables, representing the charging and discharging state variables of the hydrogen storage tank at time t. At the same time, the charging and discharging state variables cannot both be 1, meaning that the charging and discharging processes cannot occur simultaneously. This is the maximum charge / discharge rate coefficient for the hydrogen storage tank.

[0059] The hydrogen storage tank also needs to ensure that the energy at the beginning and end of each day is equal. The corresponding charging and discharging power constraints are as follows:

[0060] E hs，0 =E hs，T (14)

[0061] (4) Establishing a methane reactor model and its operating constraints

[0062] The hydrogen output from the electrolyzer can be methanated in a methane reactor to provide gas energy for the natural gas system, completing the hydrogen-to-gas (H2G) process. The model of the methane reactor is as follows:

[0063]

[0064] The operating constraints of the methane reactor are:

[0065]

[0066] In the formula: These represent the input hydrogen power and output gas power of the methane reactor, respectively; η mr The methanation efficiency of the methane reactor; The low calorific value of natural gas; δ mol This is the molar conversion factor for hydrogen to methane; The mass of methane per unit volume; These are the upper and lower limits of the hydrogen input power for the methanation unit, respectively. These represent the upper and lower limits of the ramp-up power for the methanation equipment.

[0067] (5) Establishing the gas turbine model and its operating constraints

[0068] The gas turbine consumes natural gas to generate electricity, and together with the P2G equipment, completes the bidirectional coupling of the electro-gas system. The gas turbine model is as follows:

[0069]

[0070] The operating constraints of the gas turbine are:

[0071]

[0072] In the formula: These represent the input gas power and output electrical power of the gas turbine, respectively; η gt λ represents the conversion efficiency of the gas turbine. gt The rate limit for unloading the gas turbine; Δt is the time interval.

[0073] Considering the load demand on each side of the integrated electric-gas energy system, we should make full use of renewable energy, improve the renewable energy absorption capacity, and leverage the "bridge" role of hydrogen energy between the electric-gas system through a refined P2G model, so as to improve the economic efficiency of integrated energy operation and strengthen the construction of environmentally friendly integrated electric-gas energy systems.

[0074] 3) Establish an objective function with the goal of minimizing the cost of the integrated energy system.

[0075] In this embodiment, the integrated energy system cost includes energy purchase and sales cost, equipment operation and maintenance cost, and carbon emission cost. The objective function for optimization is the sum of the energy purchase and sales cost, equipment operation and maintenance cost, and carbon emission cost.

[0076] The objective function of this invention considers both the operating cost of the integrated energy system and the carbon emission cost. The optimization objective is to minimize the sum of these two costs.

[0077]

[0078] In the formula: f w For system operating costs; Cost of carbon emissions.

[0079] The system operating cost includes energy purchase and sales costs. w1 and equipment operation and maintenance costs f w2 .

[0080] f w =f w1 +f w2 (20)

[0081]

[0082] In the formula: T is the day-ahead scheduling period, where t is the unit scheduling time period; These are the unit's purchase price of electricity, purchase price of gas, sales price of electricity, sales price of gas, and sales price of heat, respectively; P e，t Electricity purchased from the superior power grid at time t; P g，t The amount of gas purchased from the gas source at time t; The electrical energy sold to the upper-level power grid at time t; The amount of natural gas sold to the higher-level gas network at time t.

[0083]

[0084] In the formula: C wt C pv C ec C my C HFC C gt C hs These are the unit operating and maintenance costs for wind turbines, photovoltaic units, electrolyzers, methane reactors, hydrogen fuel cells, gas turbines, and hydrogen storage tanks, respectively; P wt，t P represents the power output of the wind turbine at time t. pv，t Let t be the power output of the photovoltaic unit.

[0085] Carbon emission costs primarily consider the emission costs caused by carbon dioxide produced by methane reactors and gas turbines.

[0086]

[0087] In the formula: This represents the unit CO2 emission factor of the methane reactor; This represents the unit CO2 emission factor of the gas turbine.

[0088] In the coupled links of the integrated energy system, carbon emissions from the methane reactor and gas turbine are considered, so the carbon emission costs during the operation of these two devices are included in the objective function. Since hydrogen energy utilization is relatively clean, the carbon emission costs of other devices are ignored. The total system operating cost, in addition to the system's energy purchase and sale costs, also considers the operating losses and start-up / shutdown losses of each conversion device, representing them collectively as equipment operation and maintenance costs. This approach helps protect equipment and extend its lifespan.

[0089] 4) Establish power balance constraints for electricity and gas in the integrated energy system, and establish power interaction constraints between the power grid, gas grid and the upper-level main grid.

[0090] Power balance constraints of integrated energy systems:

[0091]

[0092] In the formula: Let be the electrical load power at time t.

[0093] Gas power balance constraints of integrated energy systems:

[0094]

[0095] In the formula: Let be the gas load power at time t.

[0096] Power interaction constraints between the power grid, gas grid, and upstream main grid:

[0097]

[0098] In the formula: These represent time t and the upper limit of power purchase and sale from the upstream power grid and gas grid, respectively.

[0099] 5) Establish an optimization operation framework based on deep Q-networks, construct the state space, action space, action strategy and reward space of the integrated energy system, improve data utilization through experience replay pool, and obtain the optimal operation scheme through Q target network and Q estimation network. Figure 2 Optimization framework diagram for deep Q-networks.

[0100] Since the increased number of control parameters in the refined P2G model leads to a more complex optimization model, various continuous variables and action spaces will cause the curse of dimensionality. Therefore, this invention introduces the deep Q-network algorithm and proposes an optimization operation method for an integrated electric-gas energy system based on the deep Q-network algorithm. The agent in the deep Q-network interacts with the corresponding integrated energy system operation environment, and the optimal strategy is obtained through a series of exploratory and trial-and-error behaviors.

[0101] An optimization framework based on deep Q-networks is established. The framework is constructed based on Markov decision processes, comprising a state space S, an action space A, an action policy π, and a reward space R. In each learning phase, the agent observes the current state space s∈S, selects an action a∈A according to the observed state using policy π, obtains the reward r for taking that action in that state, and then transitions to the next state s′∈S.

[0102] ① Construct the state space S

[0103] The state space includes the operating state parameters of each device during the operation of the integrated system, including the input power, temperature, start-up and shutdown status of the proton exchange membrane electrolyzer, the input power, temperature, and operating status of the hydrogen fuel cell, the gas storage capacity of the hydrogen storage tank, the input power of the methane reactor and gas turbine, and the power and time of purchasing and selling electricity and gas from the upstream network.

[0104]

[0105] ② Constructing Action Space A

[0106] The action space includes the input power adjustment value and start / stop actions of the proton exchange membrane electrolyzer, the input power adjustment value and action status of the hydrogen fuel cell, the charging and discharging power of the hydrogen storage tank, the input power adjustment value of the methane reactor and gas turbine, and the power adjustment value for purchasing and selling electricity and gas from the upstream network. Among them, the power adjustment value is set as a discrete action variable.

[0107]

[0108] ③ Constructing Strategy π

[0109] The agent obtains the reward for taking a certain action in the current state by interacting with the environment. The policy π updates the state-value function based on the obtained reward through differential learning, thereby obtaining the final optimal policy. The differential iteration formula used to update the policy π is:

[0110] Q t (S t a t )=Q t (S t a t )+α[r t+1 +γmax Q t (S t+1 a t )-Q t (S t a t (29)

[0111] In the formula: Q t (S t at ) is in state S t Take action a at that time t The value obtained; r t+1 To take action a t The immediate reward obtained afterward; γ is the reward decay value; α is the learning rate.

[0112] ④ Construct reward r

[0113] The reward *r* obtained after taking an action is defined as the difference between the total cost in the previous time step and the total cost in the current time step. The agent receives a positive reward as the total cost gradually decreases.

[0114] r t =f T-t -f T-t+1 (30)

[0115] In the formula: f T-t f represents the total operating cost of the system up to time t. T-t+1 The total operating cost of taking action for the intelligent agent in the next moment.

[0116] ⑤ Experience Replay Pool

[0117] Since the data learned in deep reinforcement learning are all interconnected, and in order to learn from previous data better, the deep Q-network algorithm uses an experience replay pool. The dataset obtained from each interaction is put into the replay pool, and random sampling is performed during subsequent network training. This can both decouple the data and improve the efficiency of dataset utilization.

[0118] ⑥ Set the target network

[0119] Because the error target is updated as the network parameters are updated during differential learning, which affects the instability of neural network training, the deep Q-network algorithm uses two networks, a Q-target network and a Q-estimation network, for learning. The parameters of the Q-target network remain unchanged within a set time interval and are updated by the parameters of the Q-estimation network at the end of the interval.

[0120] This invention considers the deep coupling process of electricity, hydrogen, and gas energy in a refined P2G model. In the P2G process, hydrogen energy, as a secondary energy source, strengthens the coupling relationship between electricity, gas, and other energy sources through various conversion devices, achieving efficient energy utilization and improving the overall energy system's operating efficiency. First, a refined model of a proton exchange membrane electrolyzer is established. During the P2H process, the electrolyzer can utilize the heat generated during the chemical process to maintain its operating temperature and transfer the remaining heat energy to the heat load, fully improving heat energy utilization. The output hydrogen energy can be converted into natural gas for use by the gas grid, stored in a hydrogen storage tank, or converted into electricity for use by the power grid. Second, a refined model of a hydrogen fuel cell is established, considering the heat energy generated during operation and fully utilizing it. The proton exchange membrane electrolyzer and hydrogen fuel cell together establish a bidirectional coupling relationship between electrical energy and hydrogen energy. Considering the adjustability of various conversion devices, the conversion and coupling of energy forms are enhanced. The utilization of secondary energy in the electricity-gas system coupling link is fully considered. A time-shiftable and dispatchable hydrogen energy is provided to the system through a hydrogen storage tank, and the heat energy generated during the production process is fully utilized through a heat exchanger. By employing a deep Q-network algorithm to make decisions using an intelligent agent, and interacting with the environment within a framework of an integrated energy system optimization operation model, the goal is to minimize the total operating cost of the integrated energy system, including operating costs and carbon emission costs, ultimately obtaining the optimal operating strategy within the optimization cycle. This approach can significantly improve the system's operational economy, reduce total operating costs, and decrease carbon emissions by considering carbon emission costs, thus contributing to the construction of a cleaner and more environmentally friendly integrated electricity-gas energy system.

[0121] While the specific embodiments of the present invention have been described above in conjunction with the accompanying drawings, this is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art without creative effort based on the technical solutions of the present invention are still within the scope of protection of the present invention.

Claims

1. A method for optimizing the operation of an integrated electric-gas energy system, characterized in that, Includes the following steps: 1) Input information on wind turbines, photovoltaic units, electricity load, and gas load during the optimization period, as well as parameter information for proton exchange membrane electrolyzers, heat exchangers, hydrogen fuel cells, hydrogen storage tank models, methane reactors, and gas turbines; 2) Establish a proton exchange membrane electrolyzer model and its start-up and shutdown constraints and operating constraints; establish a hydrogen fuel cell model and its operating constraints; establish a hydrogen storage tank model and its charging and discharging power constraints; establish a methane reactor model and its operating constraints; and establish a gas turbine model and its operating constraints. 3) Establish an objective function with the goal of minimizing the cost of the integrated energy system; 4) Establish power balance constraints for the integrated energy system, power balance constraints for the gas grid, power grid and the upper-level main grid; 5) Establish an optimization operation framework based on deep Q-networks, construct the state space, action space, action strategy and reward space of the integrated energy system, improve data utilization through experience replay pool, and obtain the optimal operation scheme through Q target network and Q estimation network learning. In step 5), the state space includes the operating state parameters of each device during the operation of the integrated system, including the input power, temperature, start-up and shutdown status of the proton exchange membrane electrolyzer, the input power, temperature, and operating status of the hydrogen fuel cell, the gas storage capacity of the hydrogen storage tank, the input power of the methane reactor and the gas turbine, and the power and time of purchasing and selling electricity and gas from the upstream network. The operational space includes the input power adjustment value and start-up / stop operation of the proton exchange membrane electrolyzer, the input power adjustment value and operational status of the hydrogen fuel cell, the charging and discharging power of the hydrogen storage tank, the input power adjustment value of the methane reactor and gas turbine, and the power adjustment value for purchasing and selling electricity and gas from the upstream network. Used to update strategy The difference iteration formula is: , In the formula: In the state Take action at the time The value gained; In order to take action The immediate reward received afterward; The reward decay value; The learning rate; Rewards for taking an action The total cost is the difference between the total cost in the previous running state and the total cost in the current running state. The agent receives a positive reward as the total cost gradually decreases. In the formula For the system to run to Total operating cost at any given time. The total operating cost of taking action for the intelligent agent in the next moment.

2. The method for optimizing the operation of an integrated electric-gas energy system according to claim 1, characterized in that, In step 3), an objective function is established with the goal of minimizing the sum of the integrated energy system's energy purchase and sale costs, equipment operation and maintenance costs, and carbon emission costs. ， ， In the formula: To purchase and sell energy costs, For equipment operation and maintenance, Carbon emission costs.

3. The method for optimizing the operation of an integrated electric-gas energy system according to claim 2, characterized in that, In step 3), ， ， ， ， , In the formula: The day-ahead scheduling cycle, of which For unit scheduling time periods, , , , , These are the unit's prices for purchasing electricity, purchasing gas, selling electricity, selling gas, and selling heat. For a moment Purchase electricity from the superior power grid. For a moment Purchase gas from gas sources For a moment Electricity sold to the upper-level power grid For a moment The amount of natural gas sold to the higher-level gas network. , , , , , , The unit operating and maintenance costs are for wind turbines, photovoltaic units, electrolyzers, methane reactors, hydrogen fuel cells, gas turbines, and hydrogen storage tanks, respectively. For a moment The power output of the wind turbine For a moment Power output of photovoltaic units For the electrolytic cell at time Input AC power, For the input hydrogen power of the methane reactor, To input hydrogen power into the fuel cell, This refers to the input gas power of the gas turbine. and The time of the hydrogen storage tank Input power and output power, Unit of methane reactor Emission coefficient; Unit for gas turbine Emission coefficient, For the electrolytic cell at time temperature, and The coefficients are those obtained after linearizing the operating region of the electrolytic cell. , and These represent the thermal power supplied by the fuel cell to the heat load at time t, the heat exchanger conversion efficiency, and the thermal power entering the heat exchanger, respectively.

4. The method for optimizing the operation of an integrated electric-gas energy system according to claim 3, characterized in that, In step 4), The power balance constraint is, , The gas power balance constraint is: , The power interaction constraints between the power grid, gas grid, and the upstream main grid are as follows: ， In the formula: For the output electrical power of the fuel cell, , These are the input gas power and output electrical power of the gas turbine, respectively. For a moment electrical load power, The output gas power of the methane reactor. For a moment air load power, , , , They are time points The upper limit of power purchase and sale with the superior power grid and gas network.

Images