CN-119813367-B - Distributed multi-objective optimization method and device suitable for electric hydrogen coupling system

Abstract

The application relates to the technical field of energy optimization, and discloses a distributed multi-objective optimization method and device suitable for an electro-hydrogen coupling system. The method comprises the steps of constructing a double-target optimization model between a power supplier and a hydrogen supplier, so as to effectively solve the problem of competition conflict between the power supplier and the hydrogen supplier in the scheduling process in the existing electric hydrogen system, realize economic benefits between the power supplier and the hydrogen supplier, and ensure that the bidirectional energy coupling process between the electric power and the hydrogen energy system is more efficient. And designing a distributed zeta-constraint optimization algorithm based on an alternate direction multiplier method to solve the double-target optimization model. The algorithm can capture the optimal pareto front edge on the premise of protecting the privacy of each participation subject of the electric hydrogen system, limit the objective function of each participation subject to be in a reasonable range, ensure the stability and high efficiency of the scheduled distributed computing process, improve the coordination among energy suppliers while meeting the objective optimization of electric power and hydrogen energy suppliers, and realize the minimization of the running cost of the system.

Inventors

• DING TAO
• SUN JIAWEI
• HE YUANKANG
• SUN XIAOQIANG
• LIU JING
• YUAN YI
• JIA WENHAO
• Mu Chenggang
• HUANG YUHAN
• LIU YUZHENG
• YANG YUEYANG
• WANG SHUNQI
• Xie Liushuangfei

Assignees

• 西安交通大学
• 国家电网有限公司西北分部

Dates

Publication Date

20260505

Application Date

20241022

Claims (4)

1. 1. A distributed multi-objective optimization method suitable for an electro-hydrogen coupling system, the method comprising: based on a bidirectional energy coupling relation between the electric power system and the hydrogen energy system and a competition relation between the power supplier and the hydrogen supplier, a double-target optimization model between the power supplier and the hydrogen supplier is constructed, wherein the double-target optimization model aims at minimizing the running cost of the power supplier and the hydrogen supplier, and a target function of the power supplier and a target function of the hydrogen supplier can be expressed as follows: wherein: And The operation cost of the power supplier and the operation cost of the hydrogen supplier are respectively, And The wholesale electricity price of the upper power grid and the wholesale hydrogen price of the hydrogen source factory in the t period are respectively, And The wholesale electricity and hydrogen in t period, And The unit maintenance costs of the electrolyzer and the hydrogen fuel cell respectively, And The electric quantity is input to the electrolyzer and the hydrogen quantity is input to the hydrogen fuel cell at the t period, And The electricity purchase price from the hydrogen fuel cell and the hydrogen purchase price from the electrolysis cell in the period t respectively, And Respectively represents the output electric quantity of the hydrogen fuel cell and the output hydrogen quantity of the electrolytic tank of the node j in the period t, And The electricity selling price given to the user by the electricity supplier and the hydrogen selling price given to the user by the hydrogen supplier are respectively in the t period, And Respectively the electricity purchase quantity and the hydrogen purchase quantity of the node j in the period t, B is a set formed by all the nodes j in the system, Scheduling time periods for units; The constraint conditions include: (4) (5) (6) (7) (8) (9) wherein j, u and v are all indexes of the system node, And Representing the leg from node j to node u and the leg from node v to node j respectively, 、 Respectively represent t time period branches The active power and the reactive power flowing through the power supply, 、 、 Respectively represent t time period branches Active power, reactive power and current flowing through, 、 、 、 Respectively represent branches Resistance, reactance and branch of (c) Is used for the resistance, the reactance of the capacitor, And Representing the voltages of nodes j and u of the t-period respectively, Representing the electrical load of node j for the period t, And Respectively forming a set of all nodes and a set of all branches in the power distribution network; (10) (11) (12) (13) (14) (15) (16) Wherein, the 、 And Respectively representing the flow rate of the pipe p in the period t, the minimum flow rate and the maximum flow rate allowed to pass, And The air pressures of the inode and inode of the t period are respectively represented, And Representing the minimum and maximum pressures of node n, sgn is a sign function, 、 And Respectively representing the flow rates of the pipeline k, the hydrogen source node s and the hydrogen load node l in the period t, 、 And Respectively represent a corresponding coefficient matrix of the matrix, For the conversion coefficient of the pipe p corresponding to the period t, And Respectively representing the air pressure squares of the nodes i and j in the t period; And Respectively representing a pipeline set and a node set; (17) (18) (19) (20) Wherein, the And The conversion efficiencies of the electrolyzer and the hydrogen fuel cell respectively, And Maximum output power of the electrolyzer and the hydrogen fuel cell respectively; Solving the double-objective optimization model by using a distributed zeta-constraint optimization algorithm based on an alternate direction multiplier method to obtain the running cost of a power supplier and a hydrogen supplier and a corresponding decision variable optimization value, wherein the method comprises the following steps of: The replication variable is added to the four variables involved in formulas (17) and (19), respectively: (23) Wherein, the And Replication variables respectively representing the consumed electricity quantity and the output hydrogen energy of the electrolytic cell; And Duplicate variables representing the consumed hydrogen energy and the output electric energy of the hydrogen fuel cell, respectively; Relaxing the equation constraint of the above equation (23) into the original objective function by means of an augmented Lagrange form, resulting in an augmented Lagrange form of the objective function for the power supplier and the hydrogen supplier: (24) (25) Wherein, the A Lagrangian penalty factor; optimizing the hydrogen supplier objective function to obtain the corresponding maximum running cost And minimum running cost And establishing the expected non-dominant optimal solution set number N; The operation cost acceptable under the current step of the hydrogen supplier is established as follows: (26) Where n represents the n-th set of non-dominant optimal solutions; Adding the acceptable cost of the hydrogen supplier at the moment as constraint into a supplier optimization model, and converting the double-objective optimization problem into a single-objective form optimization problem which is easy to solve: (27) Initializing related parameters, maximum iteration times and acceptable convergence threshold of original residual errors, carrying out optimization iteration on the formula (27), calculating the original residual errors after each iteration, and judging whether the original residual errors are smaller than the convergence threshold or not, if so, stopping iteration, and outputting objective function values and optimized variable values; And finally, optimizing and calculating the objective function of the power supplier for a plurality of times according to the number of the non-dominant optimal solution sets which are expected to be obtained, obtaining the objective function value of the power supplier and the objective function value of the hydrogen supplier after each optimization is finished, and corresponding decision variables, and recording the residual error convergence condition of each solution iteration process until N solutions are obtained.
2. 2. A distributed multi-objective optimization apparatus for use in an electro-hydrogen coupling system, the apparatus comprising: The model construction module is used for constructing a double-target optimization model between the power supplier and the hydrogen supplier based on a bidirectional energy coupling relation between the power system and the hydrogen energy system and a competition relation between the power supplier and the hydrogen supplier, wherein the double-target optimization model aims at minimizing the running cost of the power supplier and the hydrogen supplier, and the power supplier objective function and the hydrogen supplier objective function can be expressed as follows: wherein: And The operation cost of the power supplier and the operation cost of the hydrogen supplier are respectively, And The wholesale electricity price of the upper power grid and the wholesale hydrogen price of the hydrogen source factory in the t period are respectively, And The wholesale electricity and hydrogen in t period, And The unit maintenance costs of the electrolyzer and the hydrogen fuel cell respectively, And The electric quantity is input to the electrolyzer and the hydrogen quantity is input to the hydrogen fuel cell at the t period, And The electricity purchase price from the hydrogen fuel cell and the hydrogen purchase price from the electrolysis cell in the period t respectively, And Respectively represents the output electric quantity of the hydrogen fuel cell and the output hydrogen quantity of the electrolytic tank of the node j in the period t, And The electricity selling price given to the user by the electricity supplier and the hydrogen selling price given to the user by the hydrogen supplier are respectively in the t period, And Respectively the electricity purchase quantity and the hydrogen purchase quantity of the node j in the period t, B is a set formed by all the nodes j in the system, Scheduling time periods for units; The constraint conditions include: (4) (5) (6) (7) (8) (9) wherein j, u and v are all indexes of the system node, And Representing the leg from node j to node u and the leg from node v to node j respectively, 、 Respectively represent t time period branches The active power and the reactive power flowing through the power supply, 、 、 Respectively represent t time period branches Active power, reactive power and current flowing through, 、 、 、 Respectively represent branches Resistance, reactance and branch of (c) Is used for the resistance, the reactance of the capacitor, And Representing the voltages of nodes j and u of the t-period respectively, Representing the electrical load of node j for the period t, And Respectively forming a set of all nodes and a set of all branches in the power distribution network; (10) (11) (12) (13) (14) (15) (16) Wherein, the 、 And Respectively representing the flow rate of the pipe p in the period t, the minimum flow rate and the maximum flow rate allowed to pass, And The air pressures of the inode and inode of the t period are respectively represented, And Representing the minimum and maximum pressures of node n, sgn is a sign function, 、 And Respectively representing the flow rates of the pipeline k, the hydrogen source node s and the hydrogen load node l in the period t, 、 And Respectively represent a corresponding coefficient matrix of the matrix, For the conversion coefficient of the pipe p corresponding to the period t, And Respectively representing the air pressure squares of the nodes i and j in the t period; And Respectively representing a pipeline set and a node set; (17) (18) (19) (20) Wherein, the And The conversion efficiencies of the electrolyzer and the hydrogen fuel cell respectively, And Maximum output power of the electrolyzer and the hydrogen fuel cell respectively; the optimization solving module is used for solving the double-objective optimization model by using a distributed zeta-constraint optimization algorithm based on an alternate direction multiplier method to obtain the running cost of the power supplier and the hydrogen supplier and the corresponding decision variable optimization value, and the method comprises the following steps of: The replication variable is added to the four variables involved in formulas (17) and (19), respectively: (23) Wherein, the And Replication variables respectively representing the consumed electricity quantity and the output hydrogen energy of the electrolytic cell; And Duplicate variables representing the consumed hydrogen energy and the output electric energy of the hydrogen fuel cell, respectively; Relaxing the equation constraint of the above equation (23) into the original objective function by means of an augmented Lagrange form, resulting in an augmented Lagrange form of the objective function for the power supplier and the hydrogen supplier: (24) (25) Wherein, the A Lagrangian penalty factor; optimizing the hydrogen supplier objective function to obtain the corresponding maximum running cost And minimum running cost And establishing the expected non-dominant optimal solution set number N; The operation cost acceptable under the current step of the hydrogen supplier is established as follows: (26) Where n represents the n-th set of non-dominant optimal solutions; Adding the acceptable cost of the hydrogen supplier at the moment as constraint into a supplier optimization model, and converting the double-objective optimization problem into a single-objective form optimization problem which is easy to solve: (27) Initializing related parameters, maximum iteration times and acceptable convergence threshold of original residual errors, carrying out optimization iteration on the formula (27), calculating the original residual errors after each iteration, and judging whether the original residual errors are smaller than the convergence threshold or not, if so, stopping iteration, and outputting objective function values and optimized variable values; And finally, optimizing and calculating the objective function of the power supplier for a plurality of times according to the number of the non-dominant optimal solution sets which are expected to be obtained, obtaining the objective function value of the power supplier and the objective function value of the hydrogen supplier after each optimization is finished, and corresponding decision variables, and recording the residual error convergence condition of each solution iteration process until N solutions are obtained.
3. 3. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the method of claim 1 when executing the program.
4. 4. A computer readable storage medium, characterized in that the computer readable storage medium has stored thereon a computer program which, when executed by a processor, implements the method as claimed in claim 1.

Description

Distributed multi-objective optimization method and device suitable for electric hydrogen coupling system Technical Field The application relates to the technical field of energy optimization, in particular to a distributed multi-objective optimization method and device suitable for an electro-hydrogen coupling system. Background With the rapid development of hydrogen energy technology and the access of large-scale renewable energy sources, the integration of an electric power system and a hydrogen energy system is gradually becoming an important direction in the energy field. The electric hydrogen coupling system not only improves the flexibility and the utilization efficiency of energy sources through bidirectional energy conversion, but also provides possibility for solving the fluctuation problem caused by high-proportion renewable energy source access in the power system. However, in such integrated systems, there is often a conflict in the economic interests of the power and hydrogen energy suppliers, and it is difficult to achieve a uniform scheduling scheme. In addition, in the electro-hydrogen coupling system, not only energy conversion between electric power and hydrogen needs to be considered, but also various physical constraints of an electric power network and a hydrogen pipe network need to be met. These complex physical couplings increase the difficulty of scheduling, particularly in the case where the power and hydrogen energy suppliers are operated by different entities, respectively, how to achieve efficient optimal scheduling without revealing the privacy of the parties becomes a big problem. The existing centralized scheduling method can uniformly consider the overall optimization of the system, but the private data of all the participants need to be acquired, so that the problems of information safety and privacy protection are faced, and the practical application requirements are difficult to adapt. Therefore, how to design a distributed scheduling method, not only can coordinate the scheduling demands of power and hydrogen energy suppliers without revealing privacy, but also can ensure the minimum running cost of the system, and becomes a key challenge of optimizing the scheduling of the current electric hydrogen coupling system. Particularly in multi-objective optimization, how to balance the conflict of interests of different participants and efficiently capture the pareto optimal solution through distributed computation has become an important research direction in an electric hydrogen coupling system. Disclosure of Invention The application provides a distributed multi-objective optimization method suitable for an electric hydrogen coupling system, which aims to solve the problems that in the prior art, economic benefits of an electric power supplier and a hydrogen energy supplier often conflict, a unified scheduling scheme is difficult to achieve, and information security and privacy protection are faced. Correspondingly, the application also provides a distributed multi-objective optimization device, an electronic device and a computer readable storage medium which are suitable for the electro-hydrogen coupling system and are used for guaranteeing the implementation and application of the method. In order to solve the technical problems, the application discloses a distributed multi-objective optimization method suitable for an electro-hydrogen coupling system, which comprises the following steps: Based on a bidirectional energy coupling relation between the electric power system and the hydrogen energy system and a competition relation between the power supplier and the hydrogen supplier, constructing a double-target optimization model between the power supplier and the hydrogen supplier, wherein the double-target optimization model aims at minimizing the running cost of the power supplier and the hydrogen supplier; And solving a double-target optimization model by using a distributed zeta-constraint optimization algorithm based on an alternate direction multiplier method to obtain the running cost of the power supplier and the hydrogen supplier and the corresponding decision variable optimization value. Preferably, based on the bi-directional energy coupling relationship between the electric power system and the hydrogen energy system and the competition relationship between the power supplier and the hydrogen supplier, constructing a dual-objective optimization model between the power supplier and the hydrogen supplier includes: Determining an objective function of an objective optimization model, wherein the objective function comprises an objective function of a hydrogen supplier and an objective function of a hydrogen supplier; Determining constraint conditions of a target optimization model, including a supplier constraint condition, a hydrogen supplier constraint condition and coupling constraint between an electric power system and a hydrogen energy system; Constructin