A method for configuring capacity of a flexible direct-current traction power supply system

Abstract

The application discloses a capacity configuration method of a flexible DC traction power supply system, and has the lowest annual equivalent investment operation cost as an optimization target to construct a capacity configuration target function; the optimization variables of the target function are photovoltaic system installed power, energy storage system rated power and rated capacity; constraint conditions are established, including power balance constraints, photovoltaic system power constraints, energy storage system charging and discharging constraints, energy storage system maximum charging and discharging power constraints, energy storage system capacity constraints, energy router power constraints and traction network voltage constraints; the photovoltaic system installed power, the energy storage system rated power and the rated capacity are taken as particle populations, an improved GA-PSO hybrid algorithm is used for optimization and solution, and the optimal photovoltaic system installed power, the energy storage system rated power and the rated capacity are obtained. The application is suitable for the energy bidirectional flow of the flexible DC traction power supply system and the new energy access characteristics, has strong universality, and can be directly applied to capacity planning of a rail transit flexible DC power supply project.

Description

Technical Field

[0001] This invention relates to the field of rail transit power supply technology, and in particular to a method for configuring the capacity of a flexible DC traction power supply system. Background Technology

[0002] With the continuous expansion of rail transit, flexible DC traction power supply systems have become the core development direction for green power supply in rail transit due to their advantages such as bidirectional energy flow, efficient absorption of regenerative braking energy, and compatibility with photovoltaic new energy access. This system integrates photovoltaic systems, energy storage systems, MMC converter stations, and energy routers. The rational configuration of its photovoltaic installed capacity, rated energy storage power, and capacity is crucial for achieving safe, stable, economical, and efficient system operation.

[0003] The existing capacity configuration methods for flexible DC traction power supply systems have significant technical flaws:

[0004] 1. The capacity configuration does not take the economic efficiency of the whole life cycle as the core objective, and does not comprehensively consider the initial investment cost, operation and maintenance cost and electricity cost. The configuration result is prone to problems such as overcapacity leading to excessive costs or undercapacity affecting the stability of power supply.

[0005] 2. The constraints are incomplete, failing to take into account multiple constraints such as power balance, energy storage state of charge, traction network voltage, and equipment power, making it difficult for the configuration scheme to meet actual operational safety requirements;

[0006] 3. The optimization algorithm has inherent defects. Traditional genetic algorithm has strong global search ability but slow convergence speed and is prone to premature convergence. Traditional particle swarm algorithm converges quickly but is prone to getting trapped in local optima. A single algorithm cannot achieve the global optimal solution for capacity configuration.

[0007] 4. No dedicated capacity configuration scheme was designed for the bidirectional energy flow and random fluctuations of new energy sources in flexible DC traction power supply systems, resulting in poor universality and unsatisfactory energy-saving and economic effects.

[0008] In summary, existing technologies lack a capacity configuration method for flexible DC traction power supply systems that takes into account the economic efficiency throughout the entire life cycle, multiple safety constraints, and global optimal solutions, and thus cannot meet the development needs of green, low-carbon, economical, and efficient rail transit. Summary of the Invention

[0009] This invention provides a method for configuring the capacity of a flexible DC traction power supply system, the technical solution of which is as follows:

[0010] A capacity configuration method for a flexible DC traction power supply system, wherein the flexible DC traction power supply system includes a traction substation and multiple stations, each station is equipped with a DC bus, and the DC bus of each station is connected to the traction substation through an energy router. The DC buses of all stations are also connected to the traction network. Each station further includes an energy storage system and / or a photovoltaic system, and the energy storage system and photovoltaic system are respectively connected to the DC bus of that station. The configuration method includes:

[0011] With the goal of minimizing the system's annual equivalent investment and operating cost, a capacity configuration objective function is constructed; the optimization variables of the objective function are the installed power of the photovoltaic system, the rated power of the energy storage system, and the rated capacity.

[0012] Establish constraints, including power balance constraints, photovoltaic system power constraints, energy storage system charge and discharge constraints, energy storage system maximum charge and discharge power constraints, energy storage system capacity constraints, energy router power constraints, and traction grid voltage constraints;

[0013] Using the installed capacity of the photovoltaic system, the rated power and rated capacity of the energy storage system as the particle population, an improved GA-PSO hybrid algorithm is used for optimization to obtain the optimal installed capacity of the photovoltaic system, the rated power and rated capacity of the energy storage system.

[0014] Preferably, in the elite retention step of the improved GA-PSO hybrid algorithm, the elite particles with the best fitness in the population are selected according to the elite retention ratio and directly retained without participating in subsequent selection, crossover, or mutation operations.

[0015] Preferably, the capacity configuration objective function is: In the formula, The system's annual equivalent investment and operating cost, The annual equivalent initial investment cost of the system, The annual equivalent operating and maintenance cost of the system, This represents the system's annual electricity cost; among which,

[0016] In the formula, The initial investment cost of the photovoltaic system, The initial investment cost of the energy storage system, This is the annual value coefficient;

[0017] In the formula, For the installed capacity of photovoltaic systems, Cost per unit power of photovoltaic modules The installation cost per unit power photovoltaic module;

[0018] In the formula, The rated power of the energy storage system. The investment cost per unit power of the energy storage system. The installation cost per unit power of the energy storage system. For the capacity of the energy storage system, The investment cost per unit capacity of the energy storage system;

[0019] In the formula, For the discount rate, This refers to the system's service life.

[0020] In the formula, The annual equivalent operation and maintenance cost of a photovoltaic system. The annual equivalent operation and maintenance cost of the energy storage system;

[0021] In the formula, The annual operation and maintenance costs of photovoltaic equipment, The annual land lease cost for photovoltaic equipment; In the formula, The operation and maintenance factor of photovoltaic equipment; In the formula, The rental fee is per unit area of ​​land. The floor area occupied by photovoltaic equipment;

[0022] In the formula, The operation and maintenance factor of energy storage equipment. This refers to the amount of electricity released by the energy storage system throughout the year. To reduce the maintenance costs per unit of electricity for energy storage systems;

[0023] In the formula, This refers to the unit price of electricity. The annual energy consumption of the flexible DC traction power supply system; In the formula, The duration of daily operation of the traction network. for The output power of the traction substation after drawing power from the grid.

[0024] Preferred,

[0025] The power balance constraint is In the formula, For the real-time power of the energy storage system, This refers to the real-time power of the energy router. Real-time power of the traction substation;

[0026] The energy storage system's charge and discharge constraints are as follows: In the formula, For the real-time state of charge of the energy storage system, This is the minimum allowable charge value for the energy storage system. This represents the maximum allowable charge value for the energy storage system. This refers to the maximum discharge or charge power of the energy storage system.

[0027] The power constraint of the energy router is: In the formula, The feedback power threshold of the energy router. The maximum traction power provided by the energy router;

[0028] The traction network voltage constraint is In the formula, This is the minimum voltage required for the traction network when the train is running. This is the actual voltage of the traction network. This is the highest voltage in the traction network when the train is running.

[0029] The power constraints of the photovoltaic system, the maximum charging and discharging power constraints of the energy storage system, and the capacity constraints of the energy storage system are determined based on the lower and upper limits of the design values.

[0030] Compared with existing technologies, the advantages of this invention are as follows: it constructs a full life-cycle economic capacity configuration optimization model, comprehensively considering multiple costs such as investment, operation and maintenance, and electricity costs; it establishes a complete safety constraint system, covering all dimensions of power, energy storage, equipment, and voltage, ensuring that the configuration scheme meets the requirements for safe operation of the project; it uses an improved GA-PSO hybrid algorithm to overcome the shortcomings of slow convergence of traditional genetic algorithms and easy local optima of particle swarm algorithms, significantly improving the solution accuracy and convergence speed; it is adapted to the bidirectional energy flow and new energy access characteristics of flexible DC traction power supply systems, has strong universality, and can be directly applied to the capacity planning of flexible DC power supply projects for rail transit. Attached Figure Description

[0031] Figure 1 The diagram illustrates the structure of a flexible DC traction power supply system as an example. The components are labeled as follows: 1. MMC converter; 2. DC-DC energy router; 3. AC-DC energy router; 4. Energy storage system; 5. Photovoltaic system; 6. DC bus; 7. Traction network; 8. DC circuit breaker.

[0032] Figure 2 This is a diagram illustrating the system cost structure of an example.

[0033] Figure 3 The flowchart of the improved GA-PSO algorithm is shown in the example.

[0034] Figure 4 A flowchart illustrating capacity configuration using the improved GA-PSO hybrid algorithm. Detailed Implementation

[0035] The purpose of this invention is to address the problems of unreasonable capacity configuration, easy susceptibility of optimization algorithms to local optima, and difficulty in balancing economy and safety in existing flexible DC traction power supply systems. This invention provides a capacity configuration optimization method for flexible DC traction power supply systems, constructing a capacity configuration optimization model with the goal of minimizing annual equivalent investment and operating costs. Multiple safety constraints are established, and an improved parallel interactive GA-PSO hybrid algorithm is used to solve for the optimal capacity configuration scheme, achieving optimal economy and stable power supply throughout the system's lifecycle.

[0036] The technical solution for achieving the objective of this invention is as follows:

[0037] A method for optimizing the capacity configuration of a flexible DC traction power supply system includes:

[0038] 1. Construction of Capacity Configuration Optimization Model

[0039] With the goal of minimizing the system's annual equivalent investment and operating costs, a capacity configuration objective function is constructed. The objective function includes three parts: annual equivalent initial investment cost, annual equivalent operation and maintenance cost, and annual electricity cost. The optimization variables are the installed power of the photovoltaic system and the rated power and rated capacity of the energy storage system.

[0040] 2. Establishing Constraints

[0041] Establish power balance constraints, energy storage system charge and discharge constraints, photovoltaic and energy storage power capacity constraints, energy router power constraints, and traction grid voltage constraints to ensure that the capacity configuration scheme meets the system's safe operation requirements.

[0042] 3. Improved GA-PSO hybrid algorithm for solving

[0043] By integrating the global search advantages of genetic algorithms and the local optimization advantages of particle swarm optimization, an improved GA-PSO hybrid algorithm with parallel interaction is designed. It adopts elite retention, tournament selection, single-point crossover, and Gaussian mutation operations, and interacts with the optimal and worst solutions of the two populations during iteration, overcoming the shortcomings of single algorithms in local optima and slow convergence, and solving for the optimal solution of capacity configuration.

[0044] 4. Configuration Scheme Output and Verification

[0045] The system outputs the optimal configuration results for the installed power of the photovoltaic system and the rated power and capacity of the energy storage system, and verifies the rationality of the configuration scheme through economic indicators and operational constraints.

[0046] The capacity configuration optimization method of the present invention is applicable to flexible DC traction power supply systems with photovoltaic access and supercapacitor energy storage, and is compatible with single power supply and hybrid power supply traction substation topologies.

[0047] Example:

[0048] by Figure 1 Taking the urban rail transit flexible DC traction power supply system shown as an example, the system includes traction substations (main and auxiliary substations) and multiple stations.

[0049] Figure 1 In the diagram, the numbers 1-8 within the boxes are the station numbers. For example... Figure 1 As shown, each station is equipped with a DC bus, and all stations' DC buses are connected to the traction network. The station's DC bus is connected to the traction substation via an energy router. If it is a DC-DC energy router, it is connected to the traction substation via an MMC converter; if it is an AC-DC energy router, it is directly connected to the traction substation. Each station is equipped with an energy storage system (supercapacitor energy storage system), and stations 2 and 6 are also equipped with photovoltaic systems.

[0050] The capacity configuration parameters of the flexible DC traction power supply system are optimized. The optimization targets include: the installed capacity of the photovoltaic systems at sites 2 and 6, and the rated power and capacity of the energy storage system at each site. The optimization objective is to achieve optimal system economy, equivalent to minimizing the system's annual investment and operating costs. The cost structure of the flexible DC traction power supply system is as follows: Figure 2 As shown.

[0051] Let the annual investment and operating cost of the system be C. Therefore, the overall objective function for capacity allocation is as follows:

[0052]

[0053] In the formula, This represents the system's annual equivalent initial investment cost; The annual equivalent operating and maintenance cost of the system; This represents the system's annual electricity cost.

[0054] (1) Annual equivalent initial investment cost

[0055] The annual equivalent initial investment cost of a system refers to the costs incurred before the system is built. It is mainly divided into the annual equivalent initial investment cost of a photovoltaic system and the annual equivalent investment cost of an energy storage system, as shown in the following formula:

[0056]

[0057] In the formula, This refers to the initial investment cost of the photovoltaic system. This represents the initial investment cost of the energy storage system.

[0058] Since the system uses annual equivalent cost calculations, parameters need to be introduced. Convert the total cost into an annual equivalent cost. It is expressed as the isochronous value coefficient, as shown in the following formula:

[0059]

[0060] In the formula, For the discount rate, This refers to the system's lifespan.

[0061] The initial investment cost of a photovoltaic system is a function that has a linear relationship with the installed capacity of the photovoltaic system, as shown in the following formula:

[0062]

[0063] In the formula, This refers to the installed capacity of the photovoltaic system. Cost per unit power of photovoltaic modules; The installation cost per unit power photovoltaic module mainly includes the integration cost of the unit power photovoltaic system and the cost of the transformer for matching the grid connection system.

[0064] The initial investment cost of an energy storage system is a function of the system's rated power and design capacity, as shown in the following formula:

[0065]

[0066] In the formula, This refers to the rated power of the energy storage system. The investment cost per unit power energy storage system; The installation cost per unit power energy storage system mainly consists of the cost of the DC-DC converter that matches the connected system to the unit power energy storage system. The design capacity of the energy storage system; This refers to the investment cost per unit capacity of an energy storage system.

[0067] (2) Annual equivalent operating and maintenance costs

[0068] Once the system is established, its various components require operation and maintenance to ensure normal and stable operation. The annual equivalent operation and maintenance cost of the system is mainly divided into the annual equivalent operation and maintenance cost of the photovoltaic system and the annual equivalent operation and maintenance cost of the energy storage system, as shown in the following formula:

[0069]

[0070] In the formula, The annual equivalent operation and maintenance cost of a photovoltaic system; This refers to the annual equivalent operation and maintenance cost of the energy storage system.

[0071] The annual equivalent operation and maintenance cost of a photovoltaic system mainly includes the annual operation and maintenance cost of the photovoltaic equipment and the annual land lease cost of the photovoltaic equipment, as shown in the following formula:

[0072]

[0073] In the formula, The annual operating and maintenance costs of photovoltaic equipment; This refers to the annual land lease cost for photovoltaic equipment.

[0074] The annual operation and maintenance costs of photovoltaic (PV) equipment are primarily proportional to the initial investment cost of the PV system. The annual operation and maintenance costs of PV equipment can be expressed as a percentage of the initial investment cost of the PV system, as shown in the following formula:

[0075]

[0076] In the formula, This refers to the operation and maintenance coefficient of photovoltaic equipment.

[0077] The annual land lease cost for photovoltaic equipment is:

[0078]

[0079] In the formula, The rental fee is calculated per unit area of ​​land. This refers to the floor space occupied by photovoltaic (PV) equipment. Approximately 150W of PV panels can be installed on 1 square meter of land.

[0080] For the annual equivalent operation and maintenance cost of energy storage systems, since the energy storage equipment used in urban rail transit occupies a relatively small area, land lease costs for energy storage systems are not considered; only the operation and maintenance costs of the energy storage equipment are considered. The operation and maintenance costs of energy storage equipment are related not only to the initial investment cost of the energy storage system but also to the amount of electricity released by the energy storage system, as shown in the following formula:

[0081]

[0082] In the formula, This refers to the operation and maintenance coefficient of energy storage equipment. This refers to the amount of electricity released by the energy storage system throughout the year. The maintenance cost per unit of electricity released for the energy storage system.

[0083] (3) Annual electricity cost of the system

[0084] To evaluate the economic efficiency of a flexible DC traction power supply system, it is necessary to calculate the system's annual energy consumption. The system's annual energy consumption is:

[0085]

[0086] In the formula, The annual energy consumption of the flexible DC traction power supply system; The daily operating hours of urban rail transit; For traction substation The power value at any given time is taken, and only the positive part is considered (i.e., the power output from the traction substation after drawing power from the grid; the negative part is the power transmitted from the traction network through the traction substation). Based on the system's annual energy consumption, the annual electricity cost of the system can be calculated, as shown in the following formula:

[0087]

[0088] In the formula, This refers to the unit price of electricity.

[0089] This system satisfies the following constraints:

[0090] (1) Power constraint

[0091] During capacity configuration, it is essential to ensure that the system maintains power balance at every moment. During system operation, the power balance between the energy storage system, energy router, and traction substation should be maintained at every moment, as shown in the following equation:

[0092]

[0093] In the formula, This represents the real-time power value of the traction station. For the real-time power of the energy storage system, This refers to the real-time power output of the energy router.

[0094] (2) Charge and discharge constraints of energy storage system

[0095] To ensure the safe and stable operation of the energy storage system, the charging and discharging power of the system should not exceed its set power level. Simultaneously, the system's SoC (System-on-a-Chip) should be in normal operating condition to prevent overcharging or over-discharging, which could affect the lifespan of the energy storage system. As shown in the following formula:

[0096]

[0097] In the formula, Real-time state of charge of the energy storage system; This is the minimum allowable charge value for the energy storage system; This represents the maximum allowable charge value for the energy storage system. This refers to the maximum discharge or charging power of the energy storage system.

[0098] (3) Power and capacity design constraints of photovoltaic systems and energy storage systems

[0099] The rated power and capacity of photovoltaic (PV) and energy storage systems cannot be infinite. Furthermore, PV and energy storage systems are only necessary when a certain design scale is reached. Due to limitations imposed by objective factors such as system design technology and site conditions, the power and capacity design of PV and energy storage systems should be within specified limits.

[0100] In this embodiment, the power constraints of the photovoltaic system and the energy storage system are shown in Table 1.

[0101] Table 1 Power Constraints for Photovoltaic Systems and Energy Storage Systems

[0102]

[0103] (4) Power constraints of energy routers

[0104] To ensure the normal operation of the energy router, its power output should be within a certain range, as shown in the following formula:

[0105]

[0106] In the formula, The feedback power threshold of the energy router. The maximum traction power provided to the energy router.

[0107] (5) Traction network voltage constraints

[0108] During system operation, the real-time status of the train is closely related to the voltage of the traction network. When the traction network voltage is too high, the train's regenerative braking function fails, requiring other braking methods to cooperate; when the traction network voltage is too low, the train's current is too high, triggering the train's overcurrent protection and affecting the train's traction capacity. Therefore, it is necessary to limit the traction network voltage within a certain range to ensure the safe and stable operation of the train. This is illustrated in the following formula:

[0109]

[0110] In the formula, This is the minimum voltage required for the traction network when the train is running. This represents the actual voltage value of the traction network at a certain moment. This is the highest voltage of the traction network when the train is running.

[0111] To ensure the proper functioning of the energy management strategy, the voltage of the train should fall within different ranges when it is in traction and braking states, as shown in the following formula:

[0112]

[0113] In the formula, The voltage at which the train is braking. This refers to the voltage at which the train is in traction mode. This refers to the voltage of the traction network under no-load conditions.

[0114] Finally, an improved GA-PSO hybrid algorithm is used to solve the problem.

[0115] Based on the characteristics of the system capacity configuration optimization model, the GA-PSO hybrid algorithm designed in this invention has the following core idea: Using the particle swarm optimization algorithm as its basic framework, leveraging its advantages of fast convergence and strong local mining capabilities, it quickly locates the region where the optimal solution is located in the search space; it introduces selection, crossover, and mutation evolution operations of the genetic algorithm to periodically optimize the particle population of the particle swarm optimization algorithm, enhancing population diversity, breaking local optima, and guiding the population to approach the global optimum; simultaneously, based on the large number of input parameters in the capacity configuration model, it designs a high-dimensional adaptation strategy to solve problems such as insufficient population diversity, easy destruction of superior genes, and the contradiction between convergence speed and optimization accuracy, thereby improving algorithm performance.

[0116] The GA-PSO hybrid optimization algorithm designed in this invention has a fusion framework consisting of four core modules: an initialization module, a PSO local optimization module, a GA global optimization module, and an iteration termination module. The functions and collaborative relationships of each module are as follows:

[0117] (1) Initialization module: This module is responsible for generating the initial positions and velocities of the particle swarm, calculating the fitness value of each particle, and initializing individual optimal and global optimal information to provide a basis for subsequent iterative optimization. This module adopts a random initialization strategy while ensuring that the positions and velocities of the particles are within the constraints, thereby improving the diversity of the initial population.

[0118] (2) PSO Local Optimization Module: Responsible for updating particle velocity and position, initial updates of individual and global optima, and realizing local mining. This module guides particles toward individual and global optima by using the velocity-position update formula of PSO, quickly locates the region where the optimal solution is located in the search space, and improves the convergence speed of the algorithm.

[0119] (3) GA Global Optimization Module: This module is responsible for the evolutionary optimization of the particle population after PSO update, realizing global exploration. Through GA selection, crossover, and mutation operations, this module enhances population diversity, breaks the local optima that PSO may get stuck in, optimizes the population structure, and at the same time retains superior genes through elite retention operations, guiding the population to approach the global optimum and improving the optimization accuracy of the algorithm.

[0120] (4) Iteration Termination Module: This module is responsible for determining whether the algorithm has reached the termination condition. If it has, it outputs the global optimal solution and the optimal fitness value; otherwise, it returns to the PSO local optimization module to continue iterating. The termination condition uses a dual judgment of "maximum number of iterations + fitness accuracy" to ensure that the algorithm will not fail to converge due to insufficient iterations, nor will it increase the computational cost due to excessive iterations.

[0121] Specific steps are as follows: Figure 3 As shown. Detailed steps are as follows:

[0122] (1) Initialization module:

[0123] Particle swarm initialization: Within the constraints, the initial capacity of the photovoltaic system and the energy storage system is randomly generated, and the photovoltaic and energy storage data are used as the initial particle swarm data.

[0124] Fitness calculation: Calculate the fitness value of each particle; initialize the individual's optimal position and optimal fitness value; determine the global optimal position and global optimal fitness value.

[0125] (2) Local optimization module of particle swarm optimization algorithm:

[0126] Update particle velocity and position: Update the velocity and position of each particle, while constraining the updated velocity and position to ensure they are within a reasonable range; the expression for updating particle velocity and position is:

[0127]

[0128]

[0129] In the formula, k is the number of iterations; Inertial weight; , These are learning factors, which control the intensity of the particle's learning towards the individual optimum and the global optimum, respectively. , The random number between [0,1] is used to increase the randomness of particle updates.

[0130] Individual optimal and global optimal update: Calculate the updated fitness value of each particle and compare it with the result of the previous iteration. If the fitness value is better than the individual optimal fitness value of the particle, then update the individual optimal position and the individual optimal fitness value. Traverse the individual optimal fitness values ​​of all particles. If there is an individual with a better fitness value than the global optimal fitness value, then update the global optimal position and the global optimal fitness value.

[0131] (3) Genetic Algorithm Global Optimization Module:

[0132] Elite retention: Based on the elite retention ratio, the elite particles with the best fitness in the population are selected and directly retained without participating in subsequent selection, crossover, or mutation operations, so as to avoid the destruction of excellent genes.

[0133] Tournament selection: The tournament selection method has the advantages of good stability and is not easily affected by extreme individuals. It is suitable for high-dimensional optimization problems in the capacity allocation model of this invention. Its core logic is: randomly select several individuals from the population, select the individual with the best fitness as the parent individual, and repeat this process until a sufficient number of parent individuals are selected. Therefore, the tournament selection method is used to select parent particles with better fitness from the remaining particles. That is, randomly select several individuals from the population to "compete", select the individual with the best performance according to the fitness value as the winner, and the final winning particle is the parent particle.

[0134] Single-point crossover: For the optimization problem of the model in this invention, the single-point crossover method is simple and efficient. Its core logic is: randomly select a crossover point, exchange all gene segments after the crossover point between the two parent individuals, and generate two new offspring individuals. Therefore, this invention performs a single-point crossover operation on the selected parent particles according to a set crossover probability to generate offspring particles, thereby increasing population diversity. That is, randomly select a crossover point, exchange all gene segments after the crossover point between the two parent individuals, and generate two new offspring individuals;

[0135] Gaussian mutation: Gaussian mutation has the advantages of small perturbation and minimal disruption of superior genes, making it suitable for the capacity allocation model of this invention. This invention performs Gaussian mutation on offspring particles according to a set mutation probability, perturbing the particle's variables dimension by dimension to further enhance population diversity and avoid getting trapped in local optima. Specifically, it involves randomly selecting a variable of an individual and subjecting it to a Gaussian perturbation. (in To follow a Gaussian distribution random numbers, (For disturbance intensity), after the mutation operation, it is necessary to ensure that the value of the variable is within its constraint range.

[0136] Merging populations: Merging elite particles with their offspring particles after crossover and mutation operations to form a new population, which serves as the initial population for the next iteration.

[0137] (4) Iteration Termination Judgment Module

[0138] Determine if the following two iteration termination conditions are met: ① The number of iterations reaches the maximum number of iterations; ② The global optimal fitness value is less than the preset precision. If satisfied, output the global optimal position and the global optimal fitness value; otherwise, return to the local optimization module of the particle swarm optimization algorithm and continue iterating.

[0139] The process of configuring the capacity of a flexible DC traction power supply system using the GA-PSO algorithm is as follows: Figure 4 As shown in Table 2, the algorithm parameters are set appropriately for the flexible DC traction power supply system.

[0140] Table 2 GA-PSO Hybrid Algorithm Setting Parameters

[0141]

[0142] The optimization results are shown in Table 3.

[0143] Table 3 Capacity Configuration Results

[0144]

[0145] Calculations show that the system's annual equivalent operating cost decreased from 30.821 million yuan to 27.535 million yuan, a reduction of 10.7%. The system's annual energy consumption decreased from 41,967.7 MWh before capacity configuration to 32,338.2 MWh, a reduction of 22.9%. The results demonstrate that the method of this invention can achieve optimal capacity configuration for flexible DC traction power supply systems, balancing economy, energy efficiency, and operational safety.

Claims

1. A capacity configuration method for a flexible DC traction power supply system, wherein the flexible DC traction power supply system includes a traction substation and multiple stations, each station is equipped with a DC bus, and the DC bus of each station is connected to the traction substation through an energy router; the DC buses of all stations are also connected to the traction network; each station further includes an energy storage system and / or a photovoltaic system, the energy storage system and the photovoltaic system being respectively connected to the DC bus of that station; characterized in that, With the goal of minimizing the system's annual equivalent investment and operating cost, a capacity configuration objective function is constructed; the optimization variables of the objective function are the installed power of the photovoltaic system, the rated power of the energy storage system, and the rated capacity. Establish constraints, including power balance constraints, photovoltaic system power constraints, energy storage system charge and discharge constraints, energy storage system maximum charge and discharge power constraints, energy storage system capacity constraints, energy router power constraints, and traction grid voltage constraints; Using the installed capacity of the photovoltaic system, the rated power and rated capacity of the energy storage system as the particle population, an improved GA-PSO hybrid algorithm is used for optimization to obtain the optimal installed capacity of the photovoltaic system, the rated power and rated capacity of the energy storage system.

2. The capacity configuration method for a flexible DC traction power supply system as described in claim 1, characterized in that, In the elite retention step of the improved GA-PSO hybrid algorithm, the elite particles with the best fitness in the population are selected according to the elite retention ratio and directly retained without participating in subsequent selection, crossover, or mutation operations.

3. The capacity configuration method for a flexible DC traction power supply system as described in claim 1, characterized in that, The capacity allocation objective function is: In the formula, The system's annual equivalent investment and operating cost, The annual equivalent initial investment cost of the system, The annual equivalent operating and maintenance cost of the system, This represents the system's annual electricity cost; among which, In the formula, The initial investment cost of the photovoltaic system, The initial investment cost of the energy storage system, This is the annual value coefficient; In the formula, For the installed capacity of photovoltaic systems, Cost per unit power of photovoltaic modules The installation cost per unit power photovoltaic module; In the formula, The rated power of the energy storage system. The investment cost per unit power of the energy storage system. The installation cost per unit power of the energy storage system. For the capacity of the energy storage system, The investment cost per unit capacity of the energy storage system; In the formula, For the discount rate, This refers to the system's service life. In the formula, The annual equivalent operation and maintenance cost of a photovoltaic system. The annual equivalent operation and maintenance cost of the energy storage system; In the formula, The annual operation and maintenance costs of photovoltaic equipment, The annual land lease cost for photovoltaic equipment; In the formula, The operation and maintenance factor of photovoltaic equipment; In the formula, The rental fee is per unit area of ​​land. The floor area occupied by photovoltaic equipment; In the formula, The operation and maintenance factor of energy storage equipment. This refers to the amount of electricity released by the energy storage system throughout the year. To reduce the maintenance costs per unit of electricity for energy storage systems; In the formula, This refers to the unit price of electricity. The annual energy consumption of the flexible DC traction power supply system; In the formula, The duration of daily operation of the traction network. for The output power of the traction substation after drawing power from the grid.

4. The capacity configuration method for a flexible DC traction power supply system as described in claim 1, characterized in that, The power balance constraint is In the formula, For the real-time power of the energy storage system, This refers to the real-time power of the energy router. Real-time power of the traction substation; The energy storage system's charge and discharge constraints are as follows: In the formula, For the real-time state of charge of the energy storage system, This is the minimum allowable charge value for the energy storage system. This represents the maximum allowable charge value for the energy storage system. This refers to the maximum discharge or charge power of the energy storage system. The power constraint of the energy router is: In the formula, The feedback power threshold of the energy router. The maximum traction power provided by the energy router; The traction network voltage constraint is In the formula, This is the minimum voltage required for the traction network when the train is running. This is the actual voltage of the traction network. This is the highest voltage in the traction network when the train is running. The power constraints of the photovoltaic system, the maximum charging and discharging power constraints of the energy storage system, and the capacity constraints of the energy storage system are determined based on the lower and upper limits of the design values.

Images