Electric heat storage multi-energy cooperation consistent energy flow intelligent control method and system

Abstract

The invention discloses an electric heat storage multi-energy cooperative consistent energy flow intelligent control method and system, and relates to the technical field of electric power system intelligent control, and the method comprises the steps: carrying out the multi-energy system state modeling of an electric system, a heat energy system and an energy storage system; an adaptive deep Q network architecture is introduced, an evolutionary strategy is optimized, and local optimal traps are reduced; designing a dynamic weight adjustment mechanism, and dynamically adjusting each index in the reward function according to different energy flow cooperative control states; and dynamically correcting control parameters through a closed-loop correction mechanism fed back in real time. According to the invention, rapid convergence is realized under the condition of system load fluctuation, and peak-valley difference adjustment and voltage adjustment can be completed within 200 seconds. Through an energy flow consistency optimization model, the consumption rate of distributed energy such as photovoltaic energy is improved. Through an adaptive learning rate and a dynamic weight adjustment mechanism, the response speed of the system is improved, and power fluctuation is remarkably reduced. The dynamic adjustment of the control strategy enhances the stability and power supply quality of the power system.

Description

Technical Field

[0001] The present invention relates to the field of intelligent control technology for power systems, and in particular to an intelligent control method and system for coordinated energy flow of multiple energy sources such as electrical thermal storage. Background Art

[0002] In modern power systems, the coordinated optimization control of electrical, thermal and energy storage systems is the key to achieving stable and efficient operation of active distribution networks. However, traditional control methods, such as linear optimal control or fixed threshold control, have difficulty in achieving optimal control effects in multiple energy types and highly uncertain environments. Although the existing deep reinforcement learning (DRL) control methods have achieved nonlinear modeling and adaptive control to a certain extent, they still have limitations in accommodating a variety of distributed energy sources and improving system stability. For example, the learning rate and exploration strategy of traditional DRL methods are usually fixed, resulting in slow convergence and poor adaptability of the control system under dynamic load conditions.

[0003] This paper proposes an adaptive deep reinforcement learning (ADRL) control method by improving the traditional deep reinforcement learning algorithm to achieve multi-energy coordinated consistent energy flow control. ADRL can converge quickly under a wider range of system conditions by adaptively adjusting the learning rate, introducing a dynamic weight update mechanism, and integrating energy flow consistency and stability optimization modules, thereby improving the absorption rate of distributed energy such as photovoltaics and reducing system power fluctuations and energy losses. Summary of the invention

[0004] In view of the above-mentioned problems, the present invention is proposed.

[0005] Therefore, the problem to be solved by the present invention is: how to solve the problem that the learning rate and exploration strategy of the traditional DRL method are usually fixed, resulting in slow convergence speed and poor adaptability of the control system under dynamic load conditions.

[0006] To solve the above technical problems, the present invention provides the following technical solutions: an intelligent control method for the coordinated energy flow of electrical, thermal and energy storage systems, comprising: modeling the state of the multi-energy system of electrical, thermal and energy storage systems; introducing an adaptive deep Q network architecture and evolutionary strategy optimization to reduce local optimal traps; designing a dynamic weight adjustment mechanism to dynamically adjust various indicators in the reward function according to different energy flow coordinated control states; and dynamically correcting control parameters through a closed-loop correction mechanism with real-time feedback.

[0007] As a preferred solution of the method for intelligent control of electric, thermal and energy storage multi-energy coordinated consistent energy flow described in the present invention, wherein: the state modeling of the multi-energy system includes modeling the electrical, thermal and energy storage systems as a multi-agent Markov decision process, wherein each agent represents an energy type and interacts with different energy flows in the distribution network, including state space, action space, reward function and state transition probability; the state space refers to the system state vector containing the power state, voltage state and environmental information of electrical, thermal and energy storage, expressed as:

[0008] S={P e , Q e , P h , Q h , E s , V, T}

[0009] Among them, P e , Q e are electrical power and reactive power respectively, P h , Q h are thermal power and reactive power, E s is the energy storage capacity, V is the system voltage state, T is the ambient temperature, and S is the state space.

[0010] As a preferred solution of the intelligent control method of the electric thermal storage multi-energy coordinated energy flow of the present invention, the action space refers to the control operation of the intelligent agent in each state, including power regulation, voltage adjustment, and temperature control. The action vector is expressed as:

[0011] A={α e , a h , α s}

[0012] Among them, a e , a h , a s are the control actions of electric energy, thermal energy and energy storage respectively, and A is the action vector; the reward function is based on energy flow consistency, peak-valley difference regulation and voltage stability, and is expressed as:

[0013] R=-α·ΔP+β·η-γ·ΔV

[0014] Among them, ΔP is the peak-to-valley difference fluctuation, η is the energy flow consistency, ΔV is the voltage fluctuation, α, β, γ are weight coefficients, and R is the action vector.

[0015] As a preferred solution of the method for intelligent control of multi-energy coordinated energy flow of electrical thermal storage described in the present invention, the adaptive deep Q network architecture includes a Q value update formula and an adaptive learning rate and exploration rate; the Q value update formula using the adaptive learning rate is expressed as:

[0016]

[0017] Among them, α(t) is the learning rate under the time step, and the discount factor γ weighs the long-term and short-term benefits; in each iteration, the learning rate and exploration rate are dynamically adjusted according to environmental changes to optimize the control performance in the distributed energy scenario, which is expressed as:

[0018]

[0019] ∈ t =∈0·exp(-λ·t)

[0020] Among them, α0,∈0 is the initial value, δ,λ are adjustment parameters.

[0021] As a preferred solution of the method for intelligent control of multi-energy coordinated energy flow of electrical thermal storage described in the present invention, the strategy optimization includes fitness calculation of evolution strategy and updating of strategy parameters; the fitness calculation of evolution strategy includes defining fitness function expressed as:

[0022]

[0023] Among them, R t is the reward value, θ is the strategy parameter, and the evolution strategy selects the best strategy by maximizing the fitness; the strategy parameter update includes updating the strategy parameters using the mean and variance estimation, which is expressed as:

[0024]

[0025] Among them, η1 is the learning rate, σ is the standard deviation of the strategy update, ∈ i is the noise vector.

[0026] As a preferred solution of the method for intelligent control of multi-energy coordinated energy flow of electrical thermal storage described in the present invention, the dynamic weight adjustment includes dynamically adjusting various indicators in the reward function according to different energy flow coordinated control states, and comprehensively controlling the power consumption, stability, and peak-to-valley difference, which is expressed as:

[0027] w t =w0·exp(-λ q ·|ΔP t |)

[0028] Among them, ω0 is the initial weight, λ q is the weight adjustment factor, ΔP t Represents the peak-to-valley difference fluctuation; the system energy flow consistency control model is based on the multi-energy coordinated consumption and balanced distribution goals. The impact of system fluctuations on operation is reduced by optimizing the objective function. The optimization goal is expressed as:

[0029]

[0030] Where T is the control period, ΔP t Indicates the peak-to-valley fluctuation, ΔV t represents voltage fluctuation, λ c is the penalty coefficient, and the optimization algorithm uses adaptive gradient descent method combined with evolutionary strategy to optimize the objective function in real time.

[0031] As a preferred solution of the intelligent control method of the electric thermal storage multi-energy coordinated consistent energy flow of the present invention, wherein: the closed-loop correction mechanism of real-time feedback dynamically corrects the control parameters, including dynamically correcting the control parameters through the closed-loop correction mechanism of real-time feedback to adapt to the actual working condition changes in the operation of the active distribution network, and the real-time feedback is expressed as:

[0032] α t =α t +k·(P target -P current )

[0033] Among them, k is the feedback gain. During the operation process, through real-time analysis of state data, abnormal fluctuations are detected and responded to in time to ensure the stability and safety of the system, and the control strategy is quickly corrected in the case of load fluctuations and interference.

[0034] Another object of the present invention is to provide a system for intelligent control method of multi-energy coordinated consistent energy flow of electrical thermal storage, which can solve the problem of intelligent control of multi-energy coordinated consistent energy flow of electrical thermal storage by constructing an intelligent control system.

[0035] To solve the above technical problems, the present invention provides the following technical solutions: an intelligent control system for electric thermal storage multi-energy coordinated consistent energy flow, including an adaptive deep reinforcement learning control module, an energy flow consistency optimization module, a dynamic weight update module and a real-time data feedback and correction module; the adaptive deep reinforcement learning control module is used to automatically adjust the learning rate and exploration rate to adapt to different loads and system fluctuations, and improve the convergence and adaptability of the control system; the energy flow consistency optimization module is used to improve the absorption rate of distributed energy and reduce system power fluctuations through an energy flow consistency algorithm; the dynamic weight update module is used to adjust the weight of the control strategy in real time according to load fluctuations to achieve optimal control at different times and scenarios; the real-time data feedback and correction module is used to use the real-time data of the system to update the control parameters and optimize the operating state of the multi-energy coordinated system.

[0036] A computer device comprises a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of the above-mentioned method for intelligent control of electrical thermal storage multi-energy coordinated energy flow are implemented.

[0037] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the above-mentioned method for intelligent control of electrical thermal storage multi-energy coordinated consistent energy flow.

[0038] The beneficial effects of the present invention are as follows: the real-time adaptive optimization control of the electric heat storage multi-energy coordinated consistent energy flow intelligent control method provided by the present invention: rapid convergence is achieved under system load fluctuation conditions, and peak-to-valley difference adjustment and voltage regulation can be completed within 200 seconds. Improve the absorption rate of distributed energy: through the energy flow consistency optimization model, the absorption rate of distributed energy such as photovoltaics is increased to 99%. Improve the system response speed: through the adaptive learning rate and dynamic weight adjustment mechanism, the system response speed is increased by more than 20%, significantly reducing power fluctuations. Reduce system losses: the dynamic adjustment of the control strategy reduces the regional peak-to-valley difference by 10% and the line loss by 5%, enhancing the stability of the power system and the power supply quality. Adaptability and scalability: the adaptive adjustment capability of the system ensures its robustness and scalability under different load fluctuation scenarios, and can adapt to the efficient coordinated control needs of multiple energy systems in the future. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative labor.

[0040] Figure 1 A flow chart of a method for intelligently controlling the coordinated energy flow of multiple energy storage devices for electrical heat storage provided in the first embodiment of the present invention.

[0041] Figure 2 A core framework diagram combining adaptive deep reinforcement learning with evolutionary strategy optimization for the intelligent control method of multi-energy coordinated consistent energy flow of electrical thermal storage provided in the first embodiment of the present invention.

[0042] Figure 3 A structural diagram of an electrical thermal storage multi-energy coordinated consistent energy flow intelligent control system provided for the second embodiment of the present invention. DETAILED DESCRIPTION

[0043] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0044] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0045] Example 1, reference Figure 1 and Figure 2 , which is the first embodiment of the present invention, provides an intelligent control method for the coordinated consistent energy flow of electrical, thermal and energy storage systems, including: modeling the state of the multi-energy system of electrical, thermal and energy storage systems; introducing an adaptive deep Q network architecture and evolutionary strategy optimization to reduce local optimal traps; designing a dynamic weight adjustment mechanism to dynamically adjust various indicators in the reward function according to different energy flow coordinated control states; and dynamically correcting control parameters through a closed-loop correction mechanism with real-time feedback.

[0046] The present invention proposes an intelligent control method for the coordinated consistent energy flow of multi-energy electrical thermal storage based on adaptive deep reinforcement learning and evolutionary strategy optimization algorithm, aiming to solve the collaborative optimization problem of multi-energy systems in active distribution networks to cope with the complex regulation and control needs of distributed energy under uncertain environments. Based on the combined optimization of Adaptive Deep Q-Network (ADQN) and Evolution Strategies (ES) in reinforcement learning, efficient and stable control of the system is achieved through adaptive adjustment strategies and rapid evolutionary search strategies. The method mainly includes state modeling, intelligent agent action generation, reward function design, and a core framework combining adaptive deep reinforcement learning with evolutionary strategy optimization. Figure 2 shown.

[0047] S1. Model the state of the multi-energy system by combining electrical, thermal and energy storage systems.

[0048] The electrical, thermal, and energy storage systems are modeled as a multi-agent Markov decision process (MDP), where each agent represents an energy type (such as electrical energy, thermal energy, and energy storage) and interacts with different energy flows in the distribution network. The state space, action space, reward function, and state transition probability are described as follows:

[0049] State space S: The system state vector contains the power state, voltage state and environmental information of electrical, thermal and energy storage, and is defined as:

[0050] S={P e , Q e , P h , Q h , E s , V, T}

[0051] Among them, P e , Q e are electrical power and reactive power respectively, P h , Q h are thermal power and reactive power, E s is the energy storage capacity, V is the system voltage state, T is the ambient temperature, and S is the state space.

[0052] Action space A: The control operations of the agent in each state, including power regulation, voltage adjustment, temperature control, etc. The action vector is defined as:

[0053] A={α e , a h , α s}

[0054] Among them, a e , a h , a s They are the control actions for electric energy, thermal energy and energy storage respectively, and A is the action vector.

[0055] Reward function R: The reward function is designed based on energy flow consistency, peak-to-valley difference regulation, and voltage stability, and is specifically defined as

[0056] R=-α·ΔP+β·η-γ·ΔV

[0057] Among them, ΔP is the peak-to-valley difference fluctuation, η is the energy flow consistency, ΔV is the voltage fluctuation, α, β, γ are weight coefficients used to balance the optimization objectives, and R is the action vector.

[0058] S2. Introduce adaptive deep Q network architecture and evolutionary strategy optimization to reduce local optimal traps.

[0059] The traditional deep Q network (DQN) has poor adaptability in distributed multi-energy systems. Therefore, the adaptive deep Q network (ADQN) architecture is introduced to improve the real-time and adaptability of the control system by dynamically adjusting the learning rate and exploration rate. The specific steps are as follows:

[0060] S2.1, Q value update formula: The Q value update formula using adaptive learning rate is as follows:

[0061]

[0062] Among them, α(t) is the learning rate under the time step, and the discount factor γ weighs the long-term and short-term benefits.

[0063] S2.2, Adaptive learning rate and exploration rate: In each iteration, the learning rate and exploration rate are dynamically adjusted according to environmental changes to optimize control performance in complex distributed energy scenarios:

[0064]

[0065] ∈ t =∈0·exp(-λ·t)

[0066] Among them, α0,∈0 is the initial value, δ,λ are adjustment parameters.

[0067] In order to further improve the global search capability of ADQN under different environmental conditions, the present invention introduces the evolutionary strategy optimization (ES) method to improve the strategy exploration breadth of the system through random search and reduce local optimal traps.

[0068] Fitness calculation of evolution strategy: Define the fitness function:

[0069]

[0070] Among them, R t is the reward value, θ is the strategy parameter, and the evolution strategy selects the best strategy by maximizing fitness.

[0071] Strategy parameter update: Use mean and variance estimation to update strategy parameters so that the strategy can dynamically adapt to changes in complex environments. The update formula is as follows:

[0072]

[0073] Among them, η1 is the learning rate, σ is the standard deviation of the strategy update, ∈ i is the noise vector.

[0074] S3. Design a dynamic weight adjustment mechanism to dynamically adjust the various indicators in the reward function according to different energy flow collaborative control states.

[0075] In order to meet the multi-objective control requirements in the active distribution network multi-energy system, the present invention designs a dynamic weight adjustment mechanism to dynamically adjust the indicators in the reward function according to different energy flow collaborative control states to achieve comprehensive control of power consumption, stability, and peak-to-valley difference. The dynamic adjustment formula is as follows:

[0076] w t =w0·exp(-λ q ·|ΔP t |)

[0077] Among them, w0 is the initial weight, λ q is the weight adjustment factor, ΔP t Indicates the peak-to-valley fluctuation. By flexibly adjusting the weight coefficient, the optimal effect of the control strategy under different load conditions is ensured.

[0078] The system energy flow consistency control model is based on the goals of multi-energy coordinated consumption and balanced distribution, and reduces the impact of system fluctuations on operation through objective function optimization. The optimization goals are as follows:

[0079]

[0080] Where T is the control period, ΔP t Indicates the peak-to-valley fluctuation, Δv t represents voltage fluctuation, λ c is the penalty coefficient. The optimization algorithm uses the adaptive gradient descent method combined with the evolutionary strategy to optimize the objective function in real time to ensure the efficient convergence of the control process.

[0081] S4. Dynamically correct control parameters through a closed-loop correction mechanism with real-time feedback.

[0082] This method dynamically corrects the control parameters through a closed-loop correction mechanism with real-time feedback to adapt to the actual operating conditions in the operation of the active distribution network.

[0083] S4.1, Real-time feedback formula:

[0084] a t =a t +k·(P target -P ccurrent )

[0085] Among them, K is the feedback gain, and precise regulation of power, heat and other energies can be achieved by adjusting the control strategy.

[0086] S4.2. Abnormal detection and response mechanism: During operation, the system detects abnormal fluctuations and responds promptly through real-time analysis of status data, thereby ensuring the stability and safety of the system, especially in the case of load fluctuations and interference, the control strategy can be quickly corrected.

[0087] Example 2, reference Figure 3 , which is the second embodiment of the present invention. It is different from the previous embodiment in that it provides an intelligent control system for electrical thermal storage multi-energy coordinated consistent energy flow, including: an adaptive deep reinforcement learning control module 100, an energy flow consistency optimization module 200, a dynamic weight update module 300 and a real-time data feedback and correction module 400.

[0088] The adaptive deep reinforcement learning control module 100 is used to automatically adjust the learning rate and the exploration rate to adapt to different loads and system fluctuations, and to improve the convergence and adaptability of the control system.

[0089] The energy flow consistency optimization module 200 is used to improve the absorption rate of distributed energy and reduce system power fluctuations through an energy flow consistency algorithm.

[0090] The dynamic weight updating module 300 is used to adjust the weight of the control strategy in real time according to load fluctuations to achieve optimal control under different times and scenarios.

[0091] The real-time data feedback and correction module 400 is used to update control parameters using the real-time data of the system and optimize the operating state of the multi-energy cooperative system.

[0092] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., which can store program codes.

[0093] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0094] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0095] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0096] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. An intelligent control method for the coordinated energy flow of multiple energy storage systems, characterized in that: include, Modeling the state of multi-energy systems including electrical, thermal and energy storage systems; Introducing adaptive deep Q network architecture and evolutionary strategy optimization to reduce local optimal traps; Design a dynamic weight adjustment mechanism to dynamically adjust the indicators in the reward function according to different energy flow collaborative control states; The control parameters are dynamically corrected through a closed-loop correction mechanism with real-time feedback.

2. The method for intelligent control of multi-energy coordinated energy flow of electrical thermal storage according to claim 1, characterized in that: The multi-energy system state modeling includes modeling electrical, thermal and energy storage systems as a multi-agent Markov decision process, where each agent represents an energy type and interacts with different energy flows in the distribution network, including state space, action space, reward function and state transition probability; The state space refers to the system state vector containing the power state, voltage state and environmental information of electrical, thermal and energy storage, which is expressed as: S={P e ,Q e ,P h ,Q h ,E s V,T} Among them, P e , Q e are electrical power and reactive power respectively, P h , Q h are thermal power and reactive power, E s is the energy storage capacity, V is the system voltage state, T is the ambient temperature, and S is the state space.

3. The method for intelligent control of multi-energy coordinated energy flow of electrical thermal storage according to claim 2, characterized in that: The action space refers to the control operation of the agent in each state, including power regulation, voltage adjustment, and temperature control. The action vector is expressed as, A={a e ,a h ,a s } Among them, a e , a h , a s They are the control actions of electric energy, thermal energy, and energy storage, respectively, and A is the action vector; The reward function is based on energy flow consistency, peak-to-valley difference regulation and voltage stability, and is expressed as: R=-α·AP+β·η-γ·ΔV Among them, AP is the peak-to-valley difference fluctuation, η is the energy flow consistency, AV is the voltage fluctuation, α, β, γ is the weight coefficient, and R is the action vector.

4. The method for intelligent control of multi-energy coordinated energy flow of electrical thermal storage as claimed in claim 3, characterized in that: The adaptive deep Q network architecture includes a Q value update formula and adaptive learning rate and exploration rate; The Q value update formula using adaptive learning rate is expressed as, Among them, α(t) is the learning rate under the time step, and the discount factor γ weighs the long-term and short-term benefits; In each iteration, the learning rate and exploration rate are dynamically adjusted according to the environmental changes to optimize the control performance in the distributed energy scenario, which is expressed as, ∈ t =∈0·exp(-λ·t) Among them, α0,∈0 is the initial value, δ,λ are adjustment parameters.

5. The method for intelligent control of multi-energy coordinated energy flow of electrical thermal storage according to claim 4, characterized in that: The strategy optimization includes fitness calculation of evolution strategy and updating of strategy parameters; The fitness calculation of the evolution strategy includes defining a fitness function expressed as: Among them, R t is the reward value, θ is the strategy parameter, and the evolution strategy selects the best strategy by maximizing fitness; The strategy parameter update includes updating the strategy parameters using mean and variance estimates, expressed as, Among them, η1 is the learning rate, σ is the standard deviation of the strategy update, ∈ i is the noise vector.

6. The method for intelligent control of electric thermal storage multi-energy coordinated energy flow according to claim 5, characterized in that: The dynamic weight adjustment includes dynamically adjusting various indicators in the reward function according to different energy flow coordinated control states, and comprehensively controlling power consumption, stability, and peak-to-valley difference, which is expressed as: w t =w0·exp(-λ q ·|ΔP t |) Among them, w0 is the initial weight, λ q is the weight adjustment factor, ΔP t Indicates the peak-to-valley fluctuation; The system energy flow consistency control model is based on the goals of multi-energy coordinated consumption and balanced distribution. The impact of system fluctuations on operation is reduced by optimizing the objective function. The optimization goal is expressed as: Where T is the control period, ΔP t Indicates the peak-to-valley fluctuation, ΔV t represents voltage fluctuation, λ c is the penalty coefficient, and the optimization algorithm uses adaptive gradient descent method combined with evolutionary strategy to optimize the objective function in real time.

7. The method for intelligent control of electric thermal storage multi-energy coordinated energy flow according to claim 6, characterized in that: The real-time feedback closed-loop correction mechanism dynamically corrects the control parameters, including dynamically correcting the control parameters through the real-time feedback closed-loop correction mechanism to adapt to the actual operating conditions in the operation of the active distribution network. The real-time feedback is expressed as: a t =a t +k·(P target -P current ) Where, k is the feedback gain; During operation, the system can ensure stability and safety by analyzing status data in real time, detecting abnormal fluctuations and responding promptly, and quickly correcting control strategies in the event of load fluctuations and interference.

8. A system using the method for intelligent control of electric thermal storage multi-energy coordinated energy flow as claimed in any one of claims 1 to 7, characterized in that: It includes an adaptive deep reinforcement learning control module (100), an energy flow consistency optimization module (200), a dynamic weight updating module (300) and a real-time data feedback and correction module (400); The adaptive deep reinforcement learning control module (100) is used to automatically adjust the learning rate and the exploration rate, adapt to different loads and system fluctuations, and improve the convergence and adaptability of the control system; The energy flow consistency optimization module (200) is used to improve the absorption rate of distributed energy and reduce system power fluctuations through an energy flow consistency algorithm; The dynamic weight updating module (300) is used to adjust the weight of the control strategy in real time according to load fluctuations to achieve optimal control under different times and scenarios; The real-time data feedback and correction module (400) is used to update control parameters using the real-time data of the system and optimize the operating state of the multi-energy cooperative system.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the intelligent control method of multi-energy coordinated energy flow of electrical thermal storage described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the intelligent control method of multi-energy coordinated energy flow of electrical thermal storage described in any one of claims 1 to 7 are implemented.

Images