Intelligent edge control method for information energy system based on adaptive dynamic programming

Abstract

The present invention provides an intelligent edge control method for an information energy system based on adaptive dynamic programming, and relates to the technical field of information energy system optimization control. Compared with existing control technologies, the present invention can automatically adjust the control strategy according to the real-time changes of the system state and environment. Through online learning and dynamic programming principles, the control strategy is continuously optimized iteratively, so that the system can always pursue optimal performance in a changing environment. Compared with static preset strategies, it can better cope with complex dynamic scenarios and ensure the stability and control effect of the system even when parameters change. Moreover, the adaptive dynamic programming method has shown good performance when applied to the solution of nonlinear system optimization problems, and can perfectly overcome the adverse effects of the "curse of dimensionality", which is undoubtedly closer to actual industrial needs.

Description

Technical Field

[0001] The present invention belongs to the technical field of information energy system optimization control, and in particular relates to an information energy system intelligent edge control method based on adaptive dynamic programming. Background Art

[0002] By integrating various distributed renewable energy sources, the Energy Internet can achieve large-scale development and utilization of clean energy, reduce carbon emissions, and contribute to climate change mitigation. Its emergence and development reflects both the inherent development trends of energy systems and external demands placed on them. These factors, combined, give it distinct contemporary significance and enormous potential for development. The optimization and control of integrated energy systems is essentially an extension of smart grid optimization technology, reflecting the intersection and integration of the two disciplines of automatic control and energy applications. Optimizing the control of the Energy Internet has garnered widespread attention.

[0003] Energy Internet system optimization typically involves analyzing the system's steady-state model and finding a near-optimal solution. For complex static models, an iterative approach is often used to find the near-optimal solution. During each iteration, information such as the gradient or Jacobian matrix of the objective function and constraints is calculated based on the current solution. This information is used to update the decision variables (i.e., control parameters), and the global or local optimal solution is gradually approached through iteration. The new solution is checked to ensure that it satisfies all constraints. If not, projection operations or other correction mechanisms may be required to ensure the solution's validity. Iterations are terminated when a preset convergence criterion is reached, and the final solution obtained is considered the near-optimal solution.

[0004] In the modeling and analysis of integrated energy systems, characterizing them using algebraic equations is a common approach to solving multi-energy flow network optimization scheduling problems. Integrated energy systems involve the conversion, transmission, and storage of multiple energy forms, including electricity, heat, cooling, and gas. These processes can be described and solved using a series of algebraic equations, yielding optimal scheduling strategies for the integrated energy system, ensuring efficient, economical, and environmentally friendly operation. For each node (e.g., power plant, substation, user terminal), based on the principle of energy conservation, a balance equation is established between the supply and consumption of various energy types (electricity, heat, etc.). For various energy networks (e.g., power grid, heat grid, natural gas pipeline network), the continuous energy flow equations can be discretized into algebraic form based on the principles of energy transmission. The operating characteristics of equipment (e.g., generators, boilers, energy storage devices, etc.) can be represented using algebraic equations, including but not limited to the relationship between the output power and input fuel, equipment performance curves, startup and shutdown logic, minimum stable operating levels, and maximum power limits.

[0005] Define one or more objective functions, typically minimizing total cost, minimizing carbon emissions, maximizing energy efficiency, and so on. These objective functions are also expressed in algebraic expressions. When faced with uncertainties such as energy prices, demand forecasts, and weather conditions, introduce probability distributions or interval variables to transform the uncertain problem into a deterministic optimization problem.

[0006] Linearization methods are often used to solve these problems. These methods are primarily used to simplify complex nonlinear models and facilitate the rapid and efficient solution of optimization problems. For example, economic scheduling problems can often be transformed into linear programming problems, where the goal is to minimize total cost or maximize revenue while satisfying various constraints (such as supply and demand balance, upper and lower limits on equipment operation, and environmental emission limits).

[0007] Most current research is based on steady-state or quasi-steady-state models of the Energy Internet, with limited analysis of dynamic models of integrated energy systems. Optimization and regulation of Energy Internet systems typically involves analyzing steady-state models and seeking approximate optimal solutions. However, these solutions are of limited utility in the ever-changing world of energy networks. Because the dynamics of energy systems in real-world scenarios are subject to change due to the influence of production and daily life, optimization solutions based on steady-state models are difficult to adapt to new scenarios and environments.

[0008] Moreover, most of the existing literature uses algebraic equations to characterize the integrated energy system, while there is little research on the construction of the system's state space model, especially the interconnected large-scale system model of the integrated energy system has not yet been constructed. The research mainly focuses on the dynamic model of independent generators, wind turbines, and energy storage batteries, etc., and does not fully consider the interconnection effects between multiple energy sources. As a result, it is impossible to construct an interconnected network dynamic model that can reflect the multi-energy coupling relationship within the integrated energy system, which restricts the development of related control technologies.

[0009] Numerous studies have considered the Energy Internet as a multi-agent system, but these approaches often rely on iterative approaches to optimal solutions, making them difficult to implement online. Using iterative approaches to approximate optimal solutions for static models is ill-suited to the dynamic and highly random nature of the Energy Internet. Furthermore, as the Energy Internet's system becomes increasingly complex and the number of nodes continues to grow, the computational burden will increase exponentially, making solutions extremely difficult and the "curse of dimensionality" increasingly prominent.

[0010] While using linearization methods to solve the problem reduces the difficulty and computational complexity to a certain extent, the linearization process often overlooks some nonlinear characteristics of the original problem, which can lead to inaccurate solutions, especially when nonlinear characteristics significantly affect system performance. Many dynamic characteristics of the Energy Internet (such as fluctuations in renewable energy output, load demand uncertainty, and equipment efficiency curves) cannot be fully captured by linear models. Linear models are unable to effectively address the impact of environmental changes and uncertainties, especially the random characteristics and unexpected events faced by the Energy Internet. Simplified models fail to truly reflect system characteristics, and the resulting computational errors are also a problem that cannot be ignored.

[0011] The optimization problem of the Energy Internet is essentially a constrained optimization problem. Traditional control schemes are not adaptable to changes in the energy system structure and new energy devices, and are unable to meet the current development needs of the Energy Internet. Summary of the Invention

[0012] In view of the shortcomings of the existing technology, the present invention proposes an intelligent edge control method for an information energy system based on adaptive dynamic programming to solve the optimization scheduling problem of the information energy system.

[0013] The technical solution of the present invention is:

[0014] The intelligent edge control method of an information energy system based on adaptive dynamic programming includes the following steps:

[0015] Step 1: Model the gas grid system and power grid system in the electric-gas coupled energy system separately to obtain the dynamic model of the electric-gas coupled energy system. Use neural networks to approximate the function terms in the dynamic model of the electric-gas coupled energy system to obtain an approximate dynamic mathematical model of the electric-gas coupled energy system.

[0016] The electric-gas coupled energy system includes an electric power network system and a gas network system. The gas network system includes several gas network subsystems, and the electric power network system includes several power network subsystems. Each gas network subsystem corresponds to one power network subsystem. Such a group of corresponding gas network subsystems and power network subsystems is called an energy subsystem.

[0017] Step 1.1: Model the gas grid system and obtain the dynamic model of each gas grid subsystem.

[0018] For containing N The electric-gas coupled energy system of each energy subsystem is considered as an energy node. Energy nodes , allowing gas to flow to the energy node All energy nodes are called energy nodes Upstream nodes, energy nodes The upstream node set of , Energy Node The upstream node, is the number of the upstream node in the upstream node set, is the number of upstream nodes, let The upstream nodes Energy nodes The input gas pressure increment is , Represents a set of upstream nodes The The upstream node sends Energy nodes The input gas pressure increment is set from the upstream node The upstream nodes Energy nodes The output gas flow increment is , Represents a set of upstream nodes The The upstream node sends Energy nodes Output gas flow increment.

[0019] Let the gas flow from the energy node All energy nodes that flow to are called nodes Downstream nodes, energy nodes The downstream node set is , Energy Node The downstream node, is the number of the downstream node in the downstream node set, is the number of downstream nodes, let Energy nodes Downstream node collection The gas pressure increment inputted by all downstream nodes in is , Energy Node To the downstream node set The gas pressure increment input from the downstream node is Energy nodes Downstream node collection The gas flow increment output by all downstream nodes is , Energy Node To the downstream node set The gas flow increment output by the downstream node.

[0020] Order The gas grid state vector of energy nodes is , according to the gas network state vector, construct the The dynamic model of a gas grid subsystem is:

[0021] (6);

[0022] in, The first The status of each gas grid subsystem, for The time derivative, for The time derivative, for The time derivative, , , , , , For upstream nodes The gas flow increment input by each energy node, For the The gas pressure increment output by each energy node to the downstream node, , , For upstream nodes and The length of the natural gas pipeline between energy nodes, For the The length of the natural gas pipeline between the energy node and the downstream node, is the flow rate value at the steady-state operating point, is the pressure value at the steady-state operating point, is the speed of sound in natural gas, is the cross-sectional area of the natural gas pipeline, Indicates the friction factor when gas flows in a natural gas pipeline. is the diameter of the natural gas pipeline, where is the interconnected item, .

[0023] Step 1.2: Model the power network system and obtain the dynamic model of each power network subsystem.

[0024] No. The dynamic model of the power network subsystem in the energy subsystem is:

[0025] (8);

[0026] in, For the The frequency increment of the generator set in the power network subsystem in the energy node, for The time derivative, For the The output power increment of the generator set in the power network subsystem in the energy node, for The time derivative, For the The increment of the speed regulator valve opening of the generator set in the power network subsystem in the energy node, for derivative with respect to time; , , , is the control input variable, , For the The state vector of the power network subsystem in each energy node, , , , is the interconnected item, is the global state of the electric-gas coupled energy system, For the The state vector of the energy subsystem, For the The time constant of the generator set in the power network subsystem in each energy node, is the feedback adjustment gain, For the The time constant of the speed regulator in the power network subsystem in each energy node, For the The gain constant of the generator set in the power network subsystem in the energy node, For the The time constant of the steam turbine in the power network subsystem in each energy node, For the The gain constant of the steam turbine in the power network subsystem in the energy node, For the The increase in electric power provided by the power network subsystem in each energy node.

[0027] Step 1.3: Integrate the dynamic model of the gas grid subsystem and the dynamic model of the power grid subsystem to construct a dynamic model of each energy subsystem in the electricity-gas coupled energy system.

[0028] The dynamic model of each energy subsystem in the electric-gas coupled energy system is:

[0029] (9);

[0030] in, for The time derivative, , For the The state vector of the energy subsystem elements, , Indicates the The state vector of the energy subsystem The influence of each element on the dynamic process inside the gas grid subsystem, For interconnected items No. elements, , is the input control vector, is the matrix of interconnected terms.

[0031] Step 1.4: Use a neural network to approximate the function terms of the dynamic model of each energy subsystem in the electric-gas coupled energy system to obtain an approximate dynamic mathematical model of each energy subsystem.

[0032] The approximate dynamic mathematical model of each energy subsystem is:

[0033] (10);

[0034] in, is a stable matrix, The design size is The matrix, For size The zero matrix of , is the ideal value of the neural network weights and , is the activation function, ,in is the approximation error of the neural network.

[0035] Step 2: Design an identifier for the electric-gas coupled energy system based on the approximate dynamic mathematical model of the electric-gas coupled energy system.

[0036] The identifier of the electric-gas coupled energy system is:

[0037] (11);

[0038] in, For the The estimated value of the state vector of each energy subsystem, is the estimated value of the state vector elements, for derivative with respect to time; is the activation function estimate, is the estimated value of the neural network weight, is the dimension of the vector of neural network weight estimates, for The estimated value of is the Hurwitz matrix.

[0039] The weight adaptation law of the identifier of the electric-gas coupled energy system is:

[0040] (13);

[0041] in, for The time derivative, is the designed symmetric positive definite matrix, is the learning rate, , , is a positive definite matrix, is the state error of the identifier:

[0042] (12);

[0043] in, For the The time derivative of the state error of an energy subsystem.

[0044] Step 3: Treat each energy subsystem in the electric-gas coupled energy system as an isolated subsystem, and use the judgment network to update and iteratively solve the event-triggered approximate optimal control strategy of the electric-gas coupled energy system. At the same time, based on the identifier of the electric-gas coupled energy system obtained in step 2, the weight adaptive law of the judgment network is designed.

[0045] Step 3.1: Define the performance index of the isolated subsystem and obtain the time-driven Hamiltonian function based on the performance index of the isolated subsystem.

[0046] The performance indicators of the isolated subsystem are:

[0047] (14);

[0048] in, For the The performance index of an isolated subsystem, is the utility function, and are all symmetric positive definite matrices, For time.

[0049] The time-driven Hamiltonian function is:

[0050] (16);

[0051] in, is the time-driven Hamiltonian function; , is a smooth function, yes about The partial derivative function of For the Performance indicators of an isolated subsystem.

[0052] Step 3.2: Define the optimal value function of the performance indicator of the isolated subsystem, and based on the optimal value function of the performance indicator of the isolated subsystem and the time-driven Hamiltonian function obtained in step 3.1, obtain the optimal control strategy and the corresponding time-triggered HJB equation.

[0053] The optimal value function of the performance index of the isolated subsystem is:

[0054] (17);

[0055] in, For the The optimal value function of the performance index of an isolated subsystem.

[0056] Setting the optimal value function of the performance index of the isolated subsystem Exist and exist The above is differentiable, and the optimal control strategy is:

[0057] (18);

[0058] in, is the optimal value function of the performance index of the isolated subsystem about The partial derivative of Input variables for optimal control.

[0059] The corresponding time-triggered HJB equation is:

[0060] (19);

[0061] in, is the time-triggered HJB equation, is the utility function of the optimal control strategy.

[0062] Step 3.3: Based on the optimal control strategy obtained in step 3.2, establish the event-triggered optimal control strategy and the corresponding event-triggered HJB equation.

[0063] The event triggers the optimal control strategy as follows:

[0064] (20);

[0065] in, is the event triggering optimal control vector, , is the sampling state, For the A trigger moment, For the A trigger moment.

[0066] The event-triggered HJB equation is:

[0067] (twenty one);

[0068] in, For event triggering HJB equation, is the utility function of the event-triggered optimal control strategy.

[0069] Step 3.4: Use the neural network to approximate the optimal value function of the performance index of the isolated subsystem, obtain the optimal value function approximated by the neural network, and then obtain a new expression of the event-triggered optimal control strategy.

[0070] The optimal value function approximated by the neural network is:

[0071] (twenty two);

[0072] in, is the optimal value function approximated using a neural network, is the optimal weight of the neural network, is the number of neurons in the hidden layer, is the activation function of the neural network, is the approximation error of the neural network.

[0073] The new expression of the event-triggered optimal control strategy is:

[0074] (twenty four);

[0075] in, and They are and about The partial derivative of .

[0076] Step 3.5: Use the judgment network to approximately solve the new expression of the event-triggered optimal control strategy and obtain the event-triggered approximate optimal control strategy.

[0077] Using a judgement network The output of is used to approximate the event-triggered optimal control strategy, where is the optimal weight of the neural network The estimated value of To judge the output value of the network, is the incentive function for judging the network.

[0078] The event-triggered approximate optimal control strategy is:

[0079] (25);

[0080] in, is the approximate optimal control vector, is the estimated value of the optimal weight of the neural network.

[0081] Step 3.6: Based on the identifier of the electric-gas coupled energy system obtained in step 2, the error function of the event-triggered approximate optimal control strategy is minimized by estimating the optimal weights of the neural network in the approximate event-triggered approximate optimal control strategy of the evaluation network, and the weight adaptation law of the evaluation network is designed.

[0082] According to the dynamic , the approximate Hamiltonian equation of the event-triggered approximate optimal control strategy is obtained as:

[0083] (26);

[0084] in, is the approximate Hamiltonian equation for the event-triggered approximate optimal control strategy, is the utility function of the event-triggered approximate optimal control strategy, yes about The partial derivative of For the defined parameters.

[0085] definition and weight error , is a function of the state estimate, the approximate optimal control vector, and the optimal weight of the neural network, and is defined as .

[0086] The error function of the event-triggered approximate optimal control strategy is: ,in , and then the adaptive law of the evaluation network is designed as:

[0087] (27);

[0088] Where, for The time derivative, is the learning rate, , , is the weight error.

[0089] Step 4: Utilize the event triggering approximate optimal control strategy of the electric-gas coupled energy system and the weight adaptive law of the evaluation network to realize the control of the electric-gas coupled energy system.

[0090] Compared with the prior art, the advantages of the present invention are:

[0091] Compared to existing control technologies, the intelligent edge control method for information energy systems based on adaptive dynamic programming can automatically adjust control strategies based on real-time changes in system status and environment. Through online learning and dynamic programming principles, it continuously iterates and optimizes control strategies, enabling the system to consistently achieve optimal performance in a changing environment. Compared to static preset strategies, it is more capable of handling complex dynamic scenarios, ensuring system stability and control effectiveness even with parameter changes. Furthermore, the adaptive dynamic programming method has demonstrated excellent performance when applied to solving optimization problems for nonlinear systems, effectively overcoming the adverse effects of the "curse of dimensionality," undoubtedly more aligned with actual industrial needs.

[0092] Fine-grained adaptive control helps avoid overloaded operation of energy equipment, reduces failure rates, and thereby increases equipment life and the economy of the overall system operation; and better coordinates and optimizes the conversion and distribution of different energy sources for the conversion and interaction of various energy forms in information energy systems, thereby improving overall energy utilization efficiency, reducing losses, and promoting energy conservation and emission reduction.

[0093] The intelligent edge control method of information energy systems based on adaptive dynamic programming breaks through the limitations of traditional control methods, achieves a higher degree of automation and intelligent control, effectively improves the overall efficiency and stability of the system, and provides a more advanced solution for the optimized operation of information energy systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0094] Figure 1 This is a flow chart of an intelligent edge control method for an information energy system based on adaptive dynamic programming in an embodiment of the present invention. DETAILED DESCRIPTION

[0095] The present invention is described in detail below with reference to the accompanying drawings and embodiments.

[0096] This invention combines key technologies such as intelligent and adaptive control algorithms with the actual needs of information energy systems. It is aimed at a series of problems such as the management, scheduling, protection and economic operation of the energy Internet. It is an energy Internet control technology based on adaptive dynamic programming methods.

[0097] Unlike the traditional centralized control mode, decentralized control breaks down the optimization problems of large systems into local problems of each subsystem, greatly reducing the computing pressure of a single controller and conforming to the trend of distributed and interconnected modern energy systems.

[0098] Unlike previous static and quasi-static control technologies, this invention builds a dynamic model of the electric-gas coupled energy system for analysis. This intelligent edge control technology for information energy systems, based on adaptive dynamic programming, fully leverages adaptive dynamic programming theory to adjust control strategies in real time based on actual operating conditions. This adapts to the ever-changing and complex nature of information energy systems and demonstrates strong self-learning and evolutionary capabilities. Whether the system's internal state changes or external environmental fluctuations, it can quickly respond to find the optimal or suboptimal control strategy.

[0099] Intelligent edge control method of information energy system based on adaptive dynamic programming, such as Figure 1 As shown, the following steps are included:

[0100] Step 1: Consider the secondary optimal control problem of the electric-gas coupled energy system, the purpose of which is to make the state deviation of the electric-gas coupled energy system tend to 0, model the gas grid system and the power grid system in the electric-gas coupled energy system respectively, and obtain the dynamic model of the electric-gas coupled energy system by comprehensive simplification. Use a neural network to approximate the function terms in the dynamic model of the electric-gas coupled energy system to obtain an approximate dynamic mathematical model of the electric-gas coupled energy system; the electric-gas coupled energy system is a comprehensive energy system in which the power grid system and the gas grid system are interconnected and interact with each other through specific technical means, the gas grid system includes several gas grid subsystems, and the power grid system includes several power grid subsystems, and each gas grid subsystem corresponds to a power grid subsystem. Such a group of corresponding gas grid subsystems and power grid subsystems is called an energy subsystem.

[0101] Step 1.1: Model the gas grid system and obtain the dynamic model of each gas grid subsystem.

[0102] The main consideration is the slow transient processes caused by the injection or output of natural gas in the natural gas pipeline and fluctuations in natural gas load. The transmission of natural gas along the pipeline is assumed to be a constant temperature process. At this time, the energy conservation equation commonly considered in the natural gas pipeline network can be ignored. It is assumed that the flow direction of gas in the natural gas pipeline will not change when the load demand fluctuates. The flow direction is determined by the scheduling strategy and is not considered in this scope. In order to respond to the uncertainty of renewable energy, the output of the gas unit needs to be adjusted in a timely manner at the three-level control level. The output of the gas unit is affected by the flow and pressure at each node in the natural gas pipeline network. When the output changes, the flow and pressure requirements of each node in the natural gas pipeline network are bound to change. This requires adjusting the compression ratio of the compressor in the natural gas pipeline network to achieve stable pressure and flow in the pipeline, thereby ensuring the stable operation of the entire pipeline network.

[0103] Specifically: N The electric-gas coupled energy system of each energy subsystem is regarded as an energy node, and the energy node is constructed from an energy node. To another energy node Natural gas pipeline state model:

[0104] (1);

[0105] in, and is the number of the energy node, For the Energy nodes to the The gas pressure increment input by each energy node, For the Energy nodes to the The gas flow increment output by each energy node, For the Energy nodes to the The time derivative of the gas pressure increment input by the energy node, For the Energy nodes to the The time derivative of the gas flow increment output by each energy node, is the speed of sound in natural gas, is the cross-sectional area of the natural gas pipeline, For the Energy nodes and The length of the natural gas pipeline between energy nodes, Indicates the friction factor when gas flows in a natural gas pipeline. is the flow rate value at the steady-state operating point, is the pressure value at the steady-state operating point, is the diameter of the natural gas pipeline, For the Energy nodes to the The gas flow increment input by each energy node, For the Energy nodes to the The gas pressure increment output by each energy node.

[0106] make (2);

[0107] make (3);

[0108] in, For the The energy nodes connected to the first A column vector consisting of the gas pressure increments input by the energy nodes, For the The energy nodes connected to the first The column vector of gas flow increments output by energy nodes, Indicates that in the electric-gas coupled energy system The number of the energy node connected to the energy node, Indicates that The number of the energy node in the energy node set connected to the energy nodes, and , For the The number of energy nodes connected to each energy node; Indicates the Energy nodes to the The gas pressure increment input by each energy node, Indicates the Energy nodes to the The gas flow increment output by each energy node.

[0109] From formula (1), (2) and (3), we can get:

[0110] (4);

[0111] (5);

[0112] in, is a column vector The time derivative, is a column vector The time derivative, , , , , , .

[0113] For the Energy nodes , allowing gas to flow to the energy node All energy nodes are called energy nodes Upstream nodes, energy nodes The upstream node set of , Energy Node The upstream node, is the number of the upstream node in the upstream node set, is the number of upstream nodes, correspondingly, let the set of upstream nodes The upstream nodes Energy nodes The input gas pressure increment is , Represents a set of upstream nodes The The upstream node sends Energy nodes The input gas pressure increment is set from the upstream node The upstream nodes Energy nodes The output gas flow increment is , Represents a set of upstream nodes The The upstream node sends Energy nodes Output gas flow increment.

[0114] Let the gas flow from the energy node All energy nodes that flow to are called nodes Downstream nodes, energy nodes The downstream node set is , Energy Node The downstream node, is the number of the downstream node in the downstream node set, is the number of downstream nodes, correspondingly, let Energy nodes Downstream node collection The gas pressure increment inputted by all downstream nodes in is , Energy Node To the downstream node set The gas pressure increment input from the downstream node is Energy nodes Downstream node collection The gas flow increment output by all downstream nodes is , Energy Node To the downstream node set The gas flow increment output by the downstream node.

[0115] Order The gas grid state vector of energy nodes is , according to the gas network state vector, according to the gas network state vector, construct the The dynamic model of a gas grid subsystem is:

[0116] (6);

[0117] in, The first The status of each gas grid subsystem, for The time derivative, for The time derivative, for The time derivative, , , , , , For upstream nodes The gas flow increment input by each energy node, For the The gas pressure increment output by each energy node to the downstream node, , , For upstream nodes and The length of the natural gas pipeline between energy nodes, For the The length of the natural gas pipeline between the energy node and the downstream node, is the flow rate value at the steady-state operating point, is the pressure value at the steady-state operating point, where For interconnected items, interconnected items Outflow energy node The gas flow increment and the gas flow into the energy node The gas pressure increment, and none of them are energy nodes flow and pressure conditions, where .

[0118] Step 1.2: Model the power network system and obtain the dynamic model of each power network subsystem.

[0119] Specifically: For the For the power network subsystem in each energy node, the control process is formulated as follows:

[0120] (7);

[0121] in, For the i The state vector of the power network subsystem in each energy node, for The time derivative, For the The frequency increment of the generator set in the power network subsystem in the energy node, for The time derivative, For the The output power increment of the generator set in the power network subsystem in the energy node, for The time derivative, For the The increment of the speed regulator valve opening of the generator set in the power network subsystem in the energy node, for derivative with respect to time; For the The time constant of the generator set in the power network subsystem in each energy node, is the feedback adjustment gain, For the The time constant of the speed regulator in the power network subsystem in each energy node, For the The gain constant of the generator set in the power network subsystem in the energy node, For the The time constant of the steam turbine in the power network subsystem in each energy node, For the The gain constant of the steam turbine in the power network subsystem in the energy node, is the control input variable, For the The increase in electric power provided by the power network subsystem in each energy node.

[0122] make , , , for The state vector is composed of for and The column vector formed by for , and The column vector is composed of , , , , , , and other interconnected items , which is the global state of the electric-gas coupled energy system A complex nonlinear function, where For the The state vector of the energy subsystem is Accurate mathematical modeling is extremely difficult.

[0123] In summary, The dynamic model of the power network subsystem in the energy subsystem can be simplified as follows:

[0124] (8);

[0125] Step 1.3: Integrate the dynamic model of the gas grid subsystem and the dynamic model of the power grid subsystem to construct a dynamic model of each energy subsystem in the electricity-gas coupled energy system.

[0126] (9);

[0127] in, for The time derivative, , Indicates the The state vector of the energy subsystem The influence of each element on the dynamic process within the gas grid subsystem, such as Indicates the The state vector of the energy subsystem The influence of each element on the dynamic process inside the gas grid subsystem, For the The state vector of the energy subsystem elements, ,like For the i The state vector of the energy subsystem elements, , , , ,..., , ,..., ; For the The unknown internal dynamics of each gas grid subsystem, For interconnected items No. elements, , is the input control vector, is the matrix of interconnected terms.

[0128] Step 1.4: Use a neural network to approximate the function terms of the dynamic model of each energy subsystem in the electric-gas coupled energy system to obtain an approximate dynamic mathematical model of each energy subsystem.

[0129] Introducing the stability matrix , The design size is The matrix, For size The zero matrix of .

[0130] Approximation with neural networks The approximate dynamic mathematical model of the energy subsystem is:

[0131] (10);

[0132] in, is an integer, and , , is the ideal value of the neural network weight, is the activation function, ,in is the approximation error of the neural network.

[0133] Step 2: Design an identifier for the electric-gas coupled energy system based on the approximate dynamic mathematical model of the electric-gas coupled energy system.

[0134] make is a Hurwitz matrix, then for any given positive definite matrix , there must be a unique positive definite matrix satisfy .

[0135] The identifier of the electric-gas coupled energy system is:

[0136] (11);

[0137] in, For the The estimated value of the state vector of each energy subsystem, is the estimated value of the state vector elements, for derivative with respect to time; is the activation function estimate, is the estimated value of the neural network weight, is the dimension of the vector of neural network weight estimates, for estimated value.

[0138] remember , , and the state error of the identifier of the electric-gas coupled energy system is The elements are in For the The first elements, the simplified expression combining the two formulas is:

[0139] (12);

[0140] For the The time derivative of the state error of an energy subsystem.

[0141] The weight adaptation law of the identifier of the electric-gas coupled energy system is:

[0142] (13);

[0143] in, for The time derivative, is the designed symmetric positive definite matrix, is the learning rate, is the state error of the identifier.

[0144] Assume any and , in a tight set Above, the ideal value of the neural network weights , activation function and neural network approximation error are all norm-bounded, that is, , , , , and They are the bounds of the ideal value norm of the neural network weights, the bounds of the activation function norm, and the bounds of the approximation error norm, all of which are positive constants.

[0145] In the electric-gas coupled energy system under study, the above assumptions are met and the designed identifier is used. When the corresponding identifier weight adaptive law is adopted, the state estimation error and weight estimation error are all uniformly eventually bounded.

[0146] The identifier can reconstruct the unknown dynamics of the electric-gas coupled energy system online. The reconstructed energy subsystem is isolated, that is, there is no interconnection term. The dynamics of the isolated subsystem corresponding to the energy subsystem of the unknown electric-gas coupled energy system is: ,along with Approaching , the energy subsystem dynamics identified by the identifier can be regarded as an approximate form of the isolated subsystem dynamics. It can be reduced by increasing the number of hidden layer nodes in the neural network.

[0147] Step 3: Treat each energy subsystem in the electric-gas coupled energy system as an isolated subsystem, and use the judgment network to update and iteratively solve the event-triggered approximate optimal control strategy of the electric-gas coupled energy system. At the same time, based on the identifier of the electric-gas coupled energy system obtained in step 2, the weight adaptive law of the judgment network is designed; the isolated subsystem has no interconnection relationship with other subsystems, that is, there is no interconnection term in the dynamic model of the isolated subsystem, which is opposite to the interconnected subsystem.

[0148] Step 3.1: Define the performance index of the isolated subsystem and obtain the time-driven Hamiltonian function based on the performance index of the isolated subsystem.

[0149] Under certain conditions, the optimal control strategy of an isolated system is also the decentralized control strategy of the original interconnected system. Using the adaptive dynamic programming method to solve the optimal control strategy of the corresponding isolated subsystem can achieve the purpose of stabilizing the system.

[0150] The performance indicators of the isolated subsystem are:

[0151] (14);

[0152] in, For the The performance index of an isolated subsystem, is the utility function, and are all symmetric positive definite matrices, For time.

[0153] set up exist If the above is continuously differentiable, we can get the infinitesimal form of the above formula:

[0154] (15);

[0155] in, yes about The partial derivative function of For the Performance indicators of an isolated subsystem.

[0156] For the convenience of expression, , is a smooth function, so the time-driven Hamiltonian function can be expressed as:

[0157] (16);

[0158] in, is the time-driven Hamiltonian function.

[0159] Step 3.2: Define the optimal value function of the performance indicator of the isolated subsystem, and based on the optimal value function of the performance indicator of the isolated subsystem and the time-driven Hamiltonian function obtained in step 3.1, obtain the optimal control strategy and the corresponding time-triggered HJB equation.

[0160] First define the The optimal value function of the performance index of an isolated subsystem:

[0161] (17);

[0162] in, For the The optimal value function of the performance index of an isolated subsystem.

[0163] Setting the optimal value function of the performance index of the isolated subsystem Exist and exist The above is differentiable, and the optimal control strategy is:

[0164] (18);

[0165] in, is the optimal value function of the performance index of the isolated subsystem about The partial derivative of Input variables for optimal control.

[0166] The corresponding time-triggered HJB equation is:

[0167] (19);

[0168] in, is the time-triggered HJB equation, is the utility function of the optimal control strategy, .

[0169] Step 3.3: In order to reduce the communication and computing burden of the entire electric-gas coupled energy system, consider using an event-triggered control mechanism. Based on the optimal control strategy obtained in step 3.2, an event-triggered optimal control strategy and the corresponding event-triggered HJB equation are established.

[0170] The event triggers the optimal control strategy as follows:

[0171] (20);

[0172] in, is the event triggering optimal control vector, , is the sampling state, For the A trigger moment, For the At the same time, the time-triggered HJB equation changes to the event-triggered form, that is, the event-triggered HJB equation is:

[0173] (twenty one);

[0174] in, For event triggering HJB equation, is the utility function of the event-triggered optimal control strategy.

[0175] It should be pointed out that the event triggering error is introduced , the event triggering HJB equation (21) is not equal to 0.

[0176] Step 3.4: Use the neural network to approximate the optimal value function of the performance index of the isolated subsystem, obtain the optimal value function approximated by the neural network, and then obtain a new expression of the event-triggered optimal control strategy.

[0177] Continue to derive and set: optimal control strategy About event triggering error is Lipschitz continuous, that is, for any , there exists a positive constant Make , is the optimal control strategy derived from the actual state of the isolated subsystem, is the optimal control strategy derived from the sampled state, It is the sampling state.

[0178] Since the time-triggered HJB equation is a nonlinear partial differential equation, generally speaking, its optimal value function It is difficult to obtain analytically, so the adaptive dynamic programming method is used here to approximate the optimal value function.

[0179] With the help of neural networks, the optimal value function can be expressed as:

[0180] (twenty two);

[0181] in, is the optimal value function approximated using a neural network, is the optimal weight of the neural network, is the number of neurons in the hidden layer, and is the activation function of the neural network, is the approximation error of the neural network.

[0182] According to the above derivation, the new expression of the optimal control strategy is:

[0183] (twenty three);

[0184] and They are and about The partial derivative of .

[0185] Similarly, the event-triggered optimal control strategy can be expressed as:

[0186] (twenty four);

[0187] in, and They are and about The partial derivative of .

[0188] Step 3.5: Use the judgment network to approximately solve the new expression of the event-triggered optimal control strategy and obtain the event-triggered approximate optimal control strategy.

[0189] because is unknown, using the judgment network The output of is used to approximate the event-triggered optimal control strategy, where is the optimal weight of the neural network The estimated value of To judge the output value of the network, , the event-triggered approximate optimal control strategy finally designed is:

[0190] (25);

[0191] in, is the approximate optimal control vector, is the estimated value of the optimal weight of the neural network.

[0192] It has been proved that the lower bound of the event triggering time interval exists. When the designed event triggering control mechanism is used, the Zeno phenomenon will not occur. At the same time, the cutoff trigger operation is introduced to avoid triggering unnecessary events. That is, when the system state has converged to the trigger threshold size range, no more events will be triggered.

[0193] Step 3.6: Based on the identifier of the electric-gas coupled energy system obtained in step 2, the error function of the event-triggered approximate optimal control strategy is minimized by estimating the optimal weights of the neural network in the approximate event-triggered approximate optimal control strategy of the evaluation network. At the same time, a weight adaptive law of the evaluation network is designed to adjust the approximation effect of the neural network.

[0194] The weighted adaptive law predicts the value of an action based on the input state information or directly outputs the optimal action. By continuously updating the weights, the system's performance indicators are gradually optimized, ultimately approaching the optimal solution. When solving the optimal control problem for the corresponding isolated subsystem, the optimal control strategy solved using adaptive dynamic programming can stabilize the system.

[0195] According to the dynamic , the approximate Hamiltonian equation of the event-triggered approximate optimal control strategy is obtained as:

[0196] (26);

[0197] in, is the approximate Hamiltonian equation for the event-triggered approximate optimal control strategy, which represents a function of the estimated values of the state, the approximate optimal control vector, and the optimal weights of the neural network. is the utility function of the event-triggered approximate optimal control strategy, yes about The partial derivative of For the defined parameters.

[0198] In addition, the definition and weight error , is a function of the state estimate, the approximate optimal control vector, and the optimal weights of the neural network.

[0199] Subtracting the two equations gives .

[0200] The goal of adaptive criterion design is to find an estimate of the optimal weights To minimize the error function ,in , using the normalized gradient descent algorithm, design an estimate of the optimal weights The adaptive law is:

[0201] (27);

[0202] Where, for The time derivative, is the learning rate, , , is the weight error.

[0203] Due to the weight is a fixed value, The dynamics can be expressed as ,in for The derivative with respect to time.

[0204] For the energy subsystem under study, the continuous excitation condition is maintained and the designed adaptive law of the judgment network weight is used to ,exist , making When the designed event-based decentralized control strategy is adopted, the energy subsystem state , sampling status And judge the network weight estimation error They are all uniformly bounded. By choosing an appropriate Lyapunov function, we prove that the designed closed-loop control system is stable, and using adaptive dynamic programming to solve the optimal control strategy for the corresponding isolated subsystem can achieve the goal of stabilizing the system.

[0205] Step 4: Utilize the event triggering approximate optimal control strategy of the electric-gas coupled energy system and the weight adaptive law of the evaluation network to realize the control of the electric-gas coupled energy system.

[0206] Considering the high coupling and complex nonlinear characteristics of the electric-gas coupled energy system, a neural network model is used to reconstruct the unknown internal dynamics of the electric-gas coupled energy system. Then, an adaptive dynamic programming method is used to learn the event-triggered approximate optimal control strategy. At the same time, the event-triggered mechanism is adopted to reduce the communication burden between different regions.

[0207] The aforementioned steps derive a near-optimal control strategy. This strategy is a set of action selection or decision rules that, under constraints, optimizes the performance metrics (such as cost and efficiency) of the event-triggered near-optimal control strategy system, achieving the desired goal. In an actual event-triggered near-optimal control strategy system, sensors or other measuring devices are required to obtain the system's current state. These sensors collect key data (such as gas pressure, temperature, and flow) from the event-triggered near-optimal control strategy system to obtain the system's sampled state. A preliminary analysis of the collected data determines whether the system has deviated from its ideal operating state. To optimize performance metrics, the optimal control action to be taken under the current state is calculated based on the derived near-optimal control strategy. Specific control instructions are output through digital-to-analog conversion. These instructions typically include operating commands for various actuators in the steam turbine (such as valves, pumps, and motors). These commands can be used to adjust valve openings to alter gas pressure and flow, adjust generator output, and so on, to alter the system's operating state.

[0208] A key feature of adaptive dynamic programming is its ability to learn online, meaning it continuously improves its decisions as new data arrives. For a coupled power-gas energy system, this means it can flexibly adjust control parameters based on real-time load demand, weather forecasts, market prices, and other factors, ensuring the system always operates optimally.

Claims

1. An intelligent edge control method for an information energy system based on adaptive dynamic programming, characterized in that: The following steps are involved: Step 1: Model the gas grid system and power grid system in the electricity-gas coupled energy system separately to obtain a dynamic model of the electricity-gas coupled energy system. Use a neural network to approximate the function terms in the dynamic model of the electricity-gas coupled energy system to obtain an approximate dynamic mathematical model of the electricity-gas coupled energy system. The electric-gas coupled energy system includes an electric power network system and a gas network system. The gas network system includes a plurality of gas network subsystems, and the electric power network system includes a plurality of electric power network subsystems. Each gas network subsystem corresponds to a power network subsystem. Such a group of corresponding gas network subsystems and power network subsystems is called an energy subsystem. Step 2: Design an identifier for the electric-gas coupled energy system based on the approximate dynamic mathematical model of the electric-gas coupled energy system; Step 3: Treat each energy subsystem in the electric-gas coupled energy system as an isolated subsystem and use the judgment network to iteratively update and solve the event-triggered approximate optimal control strategy of the electric-gas coupled energy system. At the same time, based on the identifier of the electric-gas coupled energy system obtained in step 2, the weight adaptive law of the judgment network is designed; Step 4: Utilize the event-triggered approximate optimal control strategy of the electric-gas coupled energy system and the weight adaptive law of the evaluation network to realize the control of the electric-gas coupled energy system; The step 1 specifically includes: Step 1.1: Model the gas grid system and obtain the dynamic model of each gas grid subsystem; For containing N The electric-gas coupled energy system of each energy subsystem is considered as an energy node. Energy nodes , allowing gas to flow to the energy node All energy nodes are called energy nodes Upstream nodes, energy nodes The upstream node set of , Energy Node The upstream node, is the number of the upstream node in the upstream node set, is the number of upstream nodes, let The upstream nodes Energy nodes The input gas pressure increment is , Represents a set of upstream nodes The The upstream node sends Energy nodes The input gas pressure increment is set from the upstream node The upstream nodes Energy nodes The output gas flow increment is , Represents a set of upstream nodes The The upstream node sends Energy nodes Output gas flow increment; Let the gas flow from the energy node All energy nodes that flow to are called nodes Downstream nodes, energy nodes The downstream node set is , Energy Node The downstream node, is the number of the downstream node in the downstream node set, is the number of downstream nodes, let Energy nodes Downstream node collection The gas pressure increment inputted by all downstream nodes in is , Energy Node To the downstream node set The gas pressure increment input from the downstream node is Energy nodes Downstream node collection The gas flow increment output by all downstream nodes is , Energy Node To the downstream node set The gas flow increment output by the downstream node; Order The gas grid state vector of energy nodes is , according to the gas network state vector, construct the The dynamic model of a gas grid subsystem is: （6）； in, The first The status of each gas grid subsystem, for The time derivative, for The time derivative, for The time derivative, , , , , , For upstream nodes The gas flow increment input by each energy node, For the The gas pressure increment output by each energy node to the downstream node, , , For upstream nodes and The length of the natural gas pipeline between energy nodes, For the The length of the natural gas pipeline between the energy node and the downstream node, is the flow rate value at the steady-state operating point, is the pressure value at the steady-state operating point, is the speed of sound in natural gas, is the cross-sectional area of the natural gas pipeline, Indicates the friction factor when gas flows in a natural gas pipeline. is the diameter of the natural gas pipeline, where is the interconnected item, ; Step 1.2: Model the power network system and obtain the dynamic model of each power network subsystem; No. The dynamic model of the power network subsystem in the energy subsystem is: （8）； in, For the The frequency increment of the generator set in the power network subsystem in the energy node, for The time derivative, For the The output power increment of the generator set in the power network subsystem in the energy node, for The time derivative, For the The increment of the speed regulator valve opening of the generator set in the power network subsystem in the energy node, for derivative with respect to time; , , , is the control input variable, , For the The state vector of the power network subsystem in each energy node, , , , is the interconnected item, is the global state of the electric-gas coupled energy system, For the The state vector of the energy subsystem, For the The time constant of the generator set in the power network subsystem in each energy node, is the feedback adjustment gain, For the The time constant of the speed regulator in the power network subsystem in each energy node, For the The gain constant of the generator set in the power network subsystem in the energy node, For the The time constant of the steam turbine in the power network subsystem in each energy node, For the The gain constant of the steam turbine in the power network subsystem in the energy node, For the The electric power increment provided by the power network subsystem in each energy node; Step 1.3: Integrate the dynamic model of the gas grid subsystem and the dynamic model of the power grid subsystem to construct a dynamic model of each energy subsystem in the electricity-gas coupled energy system; The dynamic model of each energy subsystem in the electric-gas coupled energy system is: （9）； in, for The time derivative, , For the The state vector of the energy subsystem elements, , Indicates the The state vector of the energy subsystem The influence of each element on the dynamic process inside the gas grid subsystem, For interconnected items No. elements, , is the input control vector, is the matrix of interconnected items; Step 1.4: Use a neural network to approximate the function terms of the dynamic model of each energy subsystem in the electric-gas coupled energy system to obtain an approximate dynamic mathematical model of each energy subsystem; The approximate dynamic mathematical model of each energy subsystem is: （10）； in, is a stable matrix, The design size is The matrix, For size The zero matrix of , is the ideal value of the neural network weights and , is the activation function, ,in is the approximation error of the neural network.

2. The intelligent edge control method of information energy system based on adaptive dynamic programming according to claim 1 is characterized in that: The identifier of the electric-gas coupled energy system in step 2 is: （11）； in, For the The estimated value of the state vector of each energy subsystem, is the estimated value of the state vector elements, for derivative with respect to time; is the activation function estimate, is the estimated value of the neural network weight, is the dimension of the vector of neural network weight estimates, for The estimated value of is the Hurwitz matrix; The weight adaptation law of the identifier of the electric-gas coupled energy system is: （13）； in, for The time derivative, is the designed symmetric positive definite matrix, is the learning rate, , , is a positive definite matrix, is the state error of the identifier: （12）； in, For the The time derivative of the state error of an energy subsystem.

3. The intelligent edge control method of information energy system based on adaptive dynamic programming according to claim 2 is characterized in that: Step 3 specifically includes: Step 3.1: Define the performance index of the isolated subsystem and obtain the time-driven Hamiltonian function based on the performance index of the isolated subsystem; Step 3.2: Define the optimal value function of the performance indicator of the isolated subsystem, and based on the optimal value function of the performance indicator of the isolated subsystem and the time-driven Hamiltonian function obtained in step 3.1, obtain the optimal control strategy and the corresponding time-triggered HJB equation; Step 3.3: Based on the optimal control strategy obtained in step 3.2, establish the event-triggered optimal control strategy and the corresponding event-triggered HJB equation; Step 3.4: Use the neural network to approximate the optimal value function of the performance index of the isolated subsystem, obtain the optimal value function approximated by the neural network, and then obtain a new expression of the event-triggered optimal control strategy; Step 3.5: Use the judgment network to approximate the new expression of the event-triggered optimal control strategy and obtain the event-triggered approximate optimal control strategy; Step 3.6: Based on the identifier of the electric-gas coupled energy system obtained in step 2, the error function of the event-triggered approximate optimal control strategy is minimized by estimating the optimal weights of the neural network in the approximate event-triggered approximate optimal control strategy of the evaluation network, and the weight adaptation law of the evaluation network is designed.

4. The intelligent edge control method of information energy system based on adaptive dynamic programming according to claim 3 is characterized in that: The performance indicators of the isolated subsystem described in step 3.1 are: （14）； in, For the The performance index of an isolated subsystem, is the utility function, and are all symmetric positive definite matrices, For time; The time-driven Hamiltonian function is: （16）； in, is the time-driven Hamiltonian function; , is a smooth function, yes about The partial derivative function of For the Performance indicators of an isolated subsystem.

5. The intelligent edge control method of information energy system based on adaptive dynamic programming according to claim 4 is characterized in that: The optimal value function of the performance index of the isolated subsystem described in step 3.2 is: （17）； in, For the The optimal value function of the performance index of an isolated subsystem; Setting the optimal value function of the performance index of the isolated subsystem Exist and exist The above is differentiable, and the optimal control strategy is: （18）； in, is the optimal value function of the performance index of the isolated subsystem about The partial derivative of Input variables for optimal control; The corresponding time-triggered HJB equation is: （19）； in, is the time-triggered HJB equation, is the utility function of the optimal control strategy.

6. The intelligent edge control method of information energy system based on adaptive dynamic programming according to claim 5 is characterized in that: The event-triggered optimal control strategy described in step 3.3 is: （20）； in, is the event triggering optimal control vector, , is the sampling state, For the A trigger moment, For the A trigger moment; The event-triggered HJB equation is: （21）； in, For event triggering HJB equation, is the utility function of the event-triggered optimal control strategy.

7. The method for intelligent edge control of an information energy system based on adaptive dynamic programming according to claim 6, characterized in that: The optimal value function of the neural network approximation in step 3.4 is: （22）； in, is the optimal value function approximated using a neural network, is the optimal weight of the neural network, is the number of neurons in the hidden layer, is the activation function of the neural network, is the approximation error of the neural network; The new expression of the event-triggered optimal control strategy is: （24）； in, and They are and about The partial derivative of .

8. The intelligent edge control method of information energy system based on adaptive dynamic programming according to claim 7 is characterized in that: Step 3.5: Use the judgement network The output of is used to approximate the event-triggered optimal control strategy, where is the optimal weight of the neural network The estimated value of To judge the output value of the network, To judge the network incentive function; The event-triggered approximate optimal control strategy is: （25）； in, is the approximate optimal control vector, is the estimated value of the optimal weight of the neural network.

9. The intelligent edge control method of information energy system based on adaptive dynamic programming according to claim 8 is characterized in that: Step 3.6 Based on the dynamics of the identifier , the approximate Hamiltonian equation of the event-triggered approximate optimal control strategy is obtained as: （26）； in, is the approximate Hamiltonian equation for the event-triggered approximate optimal control strategy, is the utility function of the event-triggered approximate optimal control strategy, yes about The partial derivative of For the defined parameters; definition and weight error , is a function of the state estimate, the approximate optimal control vector, and the optimal weight of the neural network, and is defined as ; The error function of the event-triggered approximate optimal control strategy is: ,in , and then the adaptive law of the evaluation network is designed as: （27）； Where, for The time derivative, is the learning rate, , , is the weight error.

Images