Energy storage architecture optimization method and system

Abstract

The invention relates to the technical field of power distribution network architecture optimization design, and discloses an energy storage architecture optimization method and system, and the method comprises the steps: initializing the particle position and speed of a particle swarm according to the optimization target of a power distribution network architecture, and setting the initial parameter of each particle; constructing a fitness function according to the key indexes of the power distribution network, calculating a fitness value for the current state of each particle, and recording a global optimal position; dynamically adjusting the speed and position of each particle by introducing a linear decreasing inertia weight and an acceleration factor updating mechanism for balancing global and local search capabilities, so that the particle is close to a global optimal solution; and when a preset termination condition is met, outputting an optimal design scheme of the power distribution network architecture. The method is improved in the aspects of improving the optimization efficiency, ensuring clear target orientation, constructing a precise fitness function, realizing dynamic adjustment, ensuring global optimum, setting a reasonable termination condition, outputting a practical result, ensuring algorithm robustness and the like.

Description

Technical Field

[0001] The present invention relates to the technical field of distribution network architecture optimization design, and specifically to an energy storage architecture optimization method and system. Background Art

[0002] Reasonable distribution network architecture optimization design can significantly reduce power network losses and construction costs, reduce capital investment in the construction phase and maintenance costs during operation. How to determine high-quality distribution network design and optimization solutions, and determine the development direction and key indicators of distribution network design projects have become a hot topic in the industry.

[0003] The traditional particle swarm optimization algorithm has the following problems in the optimization of distribution network architecture: (1) The inertia weight is fixed, which leads to insufficient global search capability in the early stage and easy to fall into local optimality; (2) The acceleration factor lacks dynamic adjustment, the convergence speed is slow, and it is difficult to adapt to complex distribution network models. Summary of the invention

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

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions: a method for optimizing energy storage architecture, comprising: According to the optimization goal of the distribution network architecture, the particle position and velocity of the particle swarm are initialized, and the initial parameters of each particle are set; According to the key indicators of the distribution network, a fitness function is constructed, and the fitness value of each particle is calculated for its current state, and the global optimal position is recorded; By introducing a linearly decreasing inertia weight and an acceleration factor update mechanism that balances global and local search capabilities, the speed and position of each particle are dynamically adjusted to make it move closer to the global optimal solution; When the preset termination conditions are met, the optimal design solution of the distribution network architecture is output.

[0006] As a preferred solution of the energy storage architecture optimization method described in the present invention, the optimization objectives of the distribution network architecture include minimizing the annual comprehensive cost, minimizing the network loss, and minimizing the equipment depreciation cost; The initial parameters of each particle include position, speed, inertia weight, acceleration factor, random number, individual optimal position and global optimal position.

[0007] As a preferred solution of the energy storage architecture optimization method described in the present invention, wherein: the fitness function includes constructing a fitness function according to key indicators and constraints of the distribution network; The constraints include node voltage value constraints, line current constraints, load capacity constraints and radiation network structure constraints; The key indicators of the distribution network include annual comprehensive costs, network losses, equipment depreciation costs, node voltage, line current and load capacity.

[0008] As a preferred solution of the energy storage architecture optimization method described in the present invention, the linear decreasing inertia weight includes using a linear decreasing inertia weight function to gradually reduce the inertia weight, enhance the global search in the early stage, and enhance the local search in the later stage. The formula is expressed as: , in, represents the weight at iteration step t, represents the initial weight, represents the final weight, k represents the index of the current iteration step, Indicates the maximum number of iterations.

[0009] As a preferred solution of the energy storage architecture optimization method described in the present invention, a dynamic acceleration factor is adopted, a larger cognitive acceleration factor and a smaller social acceleration factor are initially set to avoid particles falling into the local optimum too early, and as the iteration proceeds, the acceleration factor is adjusted by positive feedback to gradually balance the global and local search capabilities; The formula is: , in, represents the change with time t value, represents the change with time t value, express The value at the initial time, express The value at the initial time, express The target value at the end time, express The target value at the end time, t represents the time variable, and T represents the total time. In the initial stage of optimization, set To enhance the ability of particles to explore the individual optimal solution and avoid premature convergence; as the iteration proceeds, gradually increase This can enhance the ability of particles to utilize the global optimal solution and achieve a balance between global and local searches.

[0010] As a preferred solution of the energy storage architecture optimization method described in the present invention, wherein: according to the current optimal position and the global optimal position of each particle, the speed and position of the particle are adjusted in real time using the updated inertia weight and acceleration factor; The formula for updating particle velocity is expressed as: , The formula for updating the particle position is expressed as: , in, Indicates Particles in The speed of time, represents the inertia weight, Indicates Particles in time The speed of time, represents the attraction of the particle to its historical optimal position, Represents a random number, Indicates The best historical position experienced by a particle, Indicates Particles in time The location at the moment, represents the attraction of the particle to the global optimal position in the group, Represents a random number.

[0011] Said A linear decreasing function is used, working together with a dynamic acceleration factor, to ensure the coordinated optimization of the initial global search and the later local convergence.

[0012] As a preferred solution of the energy storage architecture optimization method described in the present invention, wherein: the termination condition includes that the number of iterations reaches the maximum limit and the fitness convergence standard; The number of iterations reaching the maximum limit includes, when the number of iterations reaches the upper limit set by the algorithm, stopping the optimization and outputting the current global optimal solution; The fitness convergence criteria include: if the change of the fitness function tends to be small after multiple iterations, it means that the algorithm has approached the global optimal solution, then the iteration is stopped and the optimal solution is output; When the termination condition is not met, the iteration continues, the fitness value of each particle is recalculated, and the position and speed are updated again until the termination condition is met; When the preset termination conditions are met, the global optimal design of the distribution network architecture is obtained through particle swarm optimization; When the number of iterations reaches the upper limit set by the algorithm, the optimization is stopped and the current global optimal solution is output; if the change in the fitness function in 10 consecutive iterations is less than or equal to 1×10 −51 ×10 −5 , the algorithm is judged to have converged, the iteration is stopped and the optimal solution is output; when the termination condition is not met, the iteration continues until the condition is met.

[0013] When the preset termination conditions are met, the output distribution network architecture is designed to obtain the global optimal solution through particle swarm optimization.

[0014] An energy storage architecture optimization system, wherein: Particle initialization module, fitness calculation module, particle update module, and termination judgment module; The particle initialization module initializes the particle position and velocity of the particle swarm according to the optimization goal of the distribution network architecture, and sets the initial parameters of each particle; The fitness calculation module constructs a fitness function based on the key indicators of the distribution network, calculates the fitness value of each particle's current state, and records the global optimal position; The particle update module dynamically adjusts the speed and position of each particle to move closer to the global optimal solution by introducing a linearly decreasing inertia weight and an acceleration factor update mechanism that balances global and local search capabilities; The termination judgment module outputs the optimal design solution of the distribution network architecture when the preset termination conditions are met.

[0015] A computer device comprises: a memory and a processor; the memory stores a computer program, wherein the processor implements the steps of any one of the methods of the present invention when executing the computer program.

[0016] A computer-readable storage medium stores a computer program, wherein the computer program, when executed by a processor, implements the steps of any one of the methods of the present invention.

[0017] Beneficial effects of the present invention: By initializing the particle position and speed of the particle swarm and setting the initial parameters of each particle, the present method provides a good starting point for subsequent optimization search, reduces the time of blind search, and thus significantly improves the efficiency of the optimization process. The present method initializes the particle swarm according to the optimization target of the distribution network architecture, ensuring that the entire optimization process is always carried out around the actual needs, so that the final optimization design scheme is more in line with the actual application scenario. By constructing a fitness function based on the key indicators of the distribution network and calculating the fitness value for the current state of each particle, the present method can accurately evaluate the performance of different energy storage architectures and provide a reliable basis for finding the global optimal solution. The introduction of a linearly decreasing inertia weight and an acceleration factor update mechanism that balances global and local search capabilities enables particles to dynamically adjust speed and position during the search process, effectively balancing the relationship between exploration and utilization, and increasing the probability of finding the global optimal solution. By recording the global optimal position and updating it in each iteration, the present method ensures that the optimization process always moves closer to the global optimal solution and avoids the trap of local optimality. Setting preset termination conditions, such as the number of iterations reaching the maximum limit or the fitness convergence standard, ensures that the optimization process stops at the appropriate time, avoiding over-calculation and ensuring the stability of the optimization results. When the preset termination conditions are met, this method outputs the optimal design solution for the distribution network architecture, providing direct and operational guidance for practical engineering applications. The optimization mechanism of this method has strong robustness and can adapt to distribution network architecture optimization problems of different scales and complexities, providing wide applicability.

[0018] In summary, this method has significant beneficial effects in improving optimization efficiency, ensuring clear goal orientation, constructing accurate fitness function, realizing dynamic adjustment, ensuring global optimality, setting reasonable termination conditions, outputting practical results, and ensuring algorithm robustness, etc., providing a scientific and effective technical means for the optimization design of distribution network architecture. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0020] Figure 1 An overall flow chart of an energy storage architecture optimization method provided for the first embodiment of the present invention.

[0021] Figure 2 The algorithm test results corresponding to the test function Sphere of the energy storage architecture optimization method provided in the first embodiment of the present invention.

[0022] Figure 3 The algorithm test results corresponding to the test function Rosenbrock of the energy storage architecture optimization method provided in the first embodiment of the present invention.

[0023] Figure 4 The algorithm test results corresponding to the test function Rastrigrin of the energy storage architecture optimization method provided in the first embodiment of the present invention.

[0024] Figure 5 The algorithm test results corresponding to the test function Griewangk of the energy storage architecture optimization method provided in the first embodiment of the present invention.

[0025] Figure 6 A distribution network structure diagram before optimization of an energy storage architecture optimization method provided in the fourth embodiment of the present invention.

[0026] Figure 7 A schematic diagram of a distribution network structure after optimization according to a method for optimizing an energy storage architecture provided in accordance with a fourth embodiment of the present invention.

[0027] Figure 8 A comparison chart of the number of PSO algorithm iterations of an energy storage architecture optimization method provided in the fourth embodiment of the present invention.

[0028] Fig. 9 A comparison chart of key indicators of a PSO algorithm for an energy storage architecture optimization method provided in the fourth embodiment of the present invention. DETAILED DESCRIPTION

[0029] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0030] Example 1, reference Figure 1-Figure 5 , is an embodiment of the present invention, and provides a method for optimizing energy storage architecture, such as Figure 1 As shown, including: S1: According to the optimization goal of the distribution network architecture, the particle position and velocity of the particle swarm are initialized, and the initial parameters of each particle are set.

[0031] It should be noted that the optimization objectives of the distribution network architecture include minimizing the annual comprehensive cost, minimizing the network loss, and minimizing the equipment depreciation cost; It should be noted that the initial parameters of each particle include position, velocity, inertia weight, acceleration factor, random number, individual optimal position and global optimal position.

[0032] The formula of the optimization objective function is expressed as: , in, represents the objective function, represents the weight coefficient of investment cost and construction cost, represents the weight coefficient of power loss, represents the investment recovery rate, represents the equipment depreciation expense ratio, Indicates The length of the transmission line, Indicates The unit construction cost of a transmission line, Indicates Do the transmission lines need to be optimized? Represents the unit electricity price, Indicates the maximum load utilization hours, Indicates The energy loss of a transmission line is represented by m, and m represents the mth line.

[0033] Furthermore, in the optimization problem, the particle swarm represents multiple solution candidates. Initializing the position and velocity of the particle swarm is the first step of the algorithm, the purpose of which is to provide different initial solutions so that the algorithm has a variety of exploration starting points. Step S1 can ensure that the algorithm starts searching for the global optimal solution from multiple directions, thereby avoiding falling into the local optimal solution too early.

[0034] S2: According to the key indicators of the distribution network, a fitness function is constructed, and the fitness value of each particle is calculated for its current state, and the global optimal position is recorded.

[0035] It should be explained that the fitness function is constructed based on the key indicators and constraints of the distribution network; It should be noted that the constraints include node voltage value constraints, line current constraints, load capacity constraints and radiation network structure constraints; It should be noted that the key indicators of the distribution network include annual comprehensive costs, network losses, equipment depreciation costs, node voltage, line current and load capacity.

[0036] Furthermore, the node voltage value constraint includes a voltage penalty factor, which can be expressed as: , in, Represents the penalty factor corresponding to each node voltage; is the node voltage value of the distribution network system; , are the lower and upper limits of the node voltage value respectively.

[0037] Furthermore, the line current constraint, the line current of each transmission line in the distribution network must satisfy: , in, represents the penalty factor for the current of each transmission line in the distribution network; It indicates the current transmitted by the transmission line; Expressed as peak current, Expressed as valley current.

[0038] Furthermore, the load capacity constraint can be expressed as: , in, , They respectively represent the power flow power and the maximum allowable load capacity power of the distribution network branch.

[0039] Furthermore, a penalty function is used to implement radial network constraints in the process of optimizing the distribution network architecture. The penalty function can be expressed as: , in, Indicates the overload state of the distribution network. , They represent the penalty coefficient when the distribution network is overloaded and the non-radiative penalty coefficient when it is non-radiative.

[0040] Furthermore, the fitness function is used to evaluate the quality of each particle's solution, which is the core of the entire optimization process. By constructing a fitness function, the pros and cons of each solution can be quantified, and the current optimal solution can be determined by comparing the fitness values ​​in each iteration. The fitness function integrates multiple key indicators in the distribution network, such as annual comprehensive costs, equipment depreciation costs, etc., to ensure that the optimization results meet actual needs.

[0041] S3: By introducing a linearly decreasing inertia weight and an acceleration factor update mechanism that balances global and local search capabilities, the speed and position of each particle are dynamically adjusted to move closer to the global optimal solution.

[0042] It should be noted that a linear decreasing inertia weight function is used to balance the global search and local search capabilities of the traditional particle swarm optimization algorithm, improve the credibility of the optimization results, and thus realize the structural improvement method of the traditional particle swarm optimization algorithm. The linear decreasing inertia weight function can be expressed as: , in, represents the weight at iteration step t, represents the initial weight, represents the final weight, k represents the index of the current iteration step, Indicates the maximum number of iterations.

[0043] Furthermore, in the improved particle swarm optimization distribution network architecture design algorithm, in order to better utilize particle cognition, quickly determine the global optimal result, and quickly reach the optimal solution, an acceleration factor is used. and In the initial optimization stage of the algorithm, Set to a larger value, Set it to a smaller value to prevent particles from falling into local optimization too early in the group search process and determine the optimal solution for the current optimization stage. After accumulating a certain amount of optimization experience through the group optimization process, the acceleration factor and Perform positive feedback adjustment and continue to find the global optimal result based on the existing solution. and The positive feedback regulation process can be expressed as: , in, and represents the acceleration factor , The corresponding initial value; and represents the acceleration factor , The corresponding final value; Indicates the current iteration number of the optimization process; Indicates the maximum number of iterations set by the algorithm. In the initial stage of the optimization solution process, the particle swarm algorithm has the disadvantage that the individual cognitive ability is too strong and the social cognitive ability is relatively weak, which leads to the slow optimization speed of the algorithm. In the final stage of the optimization solution, the weak cognitive ability of the particles leads to insufficient diversity, which easily leads to the optimization process entering the local optimum. In view of the problem of the dynamic adjustment mechanism of the acceleration factor, the acceleration factor update method is proposed, which can be expressed as: , Among them, the acceleration factor , The symmetrical linear gradient range is Substitute the acceleration factor update method into the acceleration factor and The positive feedback regulation process is carried out, and the optimization performance of the proposed particle swarm optimization algorithm is tested and compared using the test functions Sphere, Rosenbrock, Rastrigrin and Griewangk. The test function can be expressed as: , in, represents the input vector, where represents the dimension of the input vector, Indicates variables; represents the Sphere function, represents the Rosenbrock function, represents the Rastrigin function, represents the Griewank function, Represented as the i+1th variable.

[0044] When testing the algorithm based on the configuration and verification method of the algorithm test, the dimension is selected as 30, the number of particles is 50, and the maximum number of iterations is 1500. The linear decrease range of the inertia weight is 0.9 to 0.4. The test results of the performance comparison of the basic particle swarm (PSO) algorithm and the improved PSO algorithm proposed in the present invention using four test functions are as follows: Figure 2 , Figure 3 , Figure 4 and Figure 5 shown.

[0045] Based on the analysis of the configuration and verification method of the algorithm test, when the improved particle swarm optimization algorithm is used to optimize the design of the distribution network architecture, the algorithm steps are as follows: Initialize the particle population and set the initial position and velocity of each particle; Calculate the fitness function value of each particle at the current moment according to the objective function of the optimization problem; Compare the fitness function value corresponding to each particle at the current moment with the optimal position it has experienced, determine and record the current global optimal position and the corresponding particle.

[0046] Modify the acceleration factor c according to the rules of the implementation design 1 、c 2 , so that the particle swarm can quickly reach the optimal solution and shorten the duration of the optimization calculation process.

[0047] Compare the fitness function value of each particle with the global optimal position, determine and record the global optimal position and the corresponding particle number at the current moment.

[0048] The position and velocity of each particle are updated in real time according to the following formula.

[0049] The formula for updating particle velocity is expressed as: , The formula for updating the particle position is expressed as: , in, Indicates Particles in The speed of time, represents the inertia weight, Indicates Particles in time The speed of time, represents the attraction of the particle to its historical optimal position, Represents a random number, Indicates The best historical position experienced by a particle, Indicates Particles in time The location at the moment, represents the attraction of the particle to the global optimal position in the group, Represents a random number.

[0050] Furthermore, by introducing a linearly decreasing inertia weight, the global and local search capabilities of the particle swarm can be effectively controlled. In the early stages of the search, a larger inertia weight helps the global search; in the later stages of the search, a smaller inertia weight promotes the local search and helps converge to the optimal solution. At the same time, the acceleration factor is used to further adjust the particle's dependence on the historical optimal solution and the global optimal solution, thereby improving the convergence efficiency and accuracy of the algorithm.

[0051] S4: When the preset termination conditions are met, the optimal design solution of the output distribution network architecture is output.

[0052] The number of iterations reaches the maximum limit and fitness convergence criteria; The number of iterations reaching the maximum limit includes, when the number of iterations reaches the upper limit set by the algorithm, stopping the optimization and outputting the current global optimal solution; The fitness convergence standard includes: if the change of the fitness function tends to be small after multiple iterations, it means that the algorithm has approached the global optimal solution. At this time, the iteration is stopped and the optimal solution is output.

[0053] When the termination condition is not met, the iteration continues, the fitness value of each particle is recalculated, and the position and velocity are updated again until the termination condition is met.

[0054] When the preset termination conditions are met, the output distribution network architecture is designed to obtain the global optimal solution through particle swarm optimization.

[0055] Furthermore, by setting the number of iterations and fitness convergence criteria as termination conditions, the algorithm can be prevented from falling into an infinite loop and the optimal solution can be output at the right time. The termination condition ensures that when the quality of the solution reaches a certain standard, the optimization is stopped in time and the optimal design solution is output, thereby improving the computational efficiency.

[0056] When the number of iterations reaches the upper limit set by the algorithm, the optimization is stopped and the current global optimal solution is output; if the change in the fitness function in 10 consecutive iterations is less than or equal to 1×10 −51 ×10 −5 , the algorithm is judged to have converged, the iteration is stopped and the optimal solution is output; when the termination condition is not met, the iteration continues until the condition is met.

[0057] Embodiment 2 is an embodiment of the present invention, which provides an energy storage architecture optimization system, including: Particle initialization module, fitness calculation module, particle update module, termination judgment module; The particle initialization module initializes the particle position and velocity of the particle swarm according to the optimization goal of the distribution network architecture, and sets the initial parameters of each particle; The fitness calculation module constructs a fitness function based on the key indicators of the distribution network, calculates the fitness value of each particle's current state, and records the global optimal position; The particle update module dynamically adjusts the speed and position of each particle to move closer to the global optimal solution by introducing a linearly decreasing inertia weight and an acceleration factor update mechanism that balances global and local search capabilities; The termination judgment module outputs the optimal design solution of the distribution network architecture when the preset termination conditions are met.

[0058] Embodiment 3, an embodiment of the present invention, is different from the first two embodiments in that: If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., which can store program codes.

[0059] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0060] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0061] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one or a combination of the following technologies known in the art: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0062] Example 4, reference Figure 6-Figure 9 In order to verify the feasibility and effectiveness of the key indicator optimization ideas of the distribution network architecture optimization design based on the improved particle swarm optimization algorithm proposed in this invention, a case analysis was carried out based on a double-ring network architecture distribution network project model in Hebei, and simulation tests were carried out in the simulation software (MATLAB).

[0063] The goal of optimizing the distribution network architecture is to determine whether transmission lines need to be set up between nodes in the distribution network. The present invention refers to binary theory, and uses binary "0" to indicate that the node does not need to set up a transmission line, and binary "1" to indicate that the node needs to set up a transmission line. During optimization, particles are used to represent the locations of different transmission lines in the distribution network, and particle combinations are used to represent the set of transmission line locations used for optimization. The distribution network project example studied in the present invention contains 10 nodes, including 1 power source point and 9 load points. The distribution network architecture contains a total of 16 transmission lines, 2 of which have been built and 14 are to be expanded. The distribution network structure is as follows: Figure 6 The node load data of the distribution network architecture are shown in Table 1.

[0064] Table 1. Distribution network architecture node load data table , according to Figure 6 It can be seen that the distribution network grid structure optimization design planning path is: 1-2, 2-4, 1-7, 1-3, 3-5, 5-9, 7-6, 7-8, 7-10. Under this path, the total network loss is 80.53kW, the total network loss cost is 120,800 yuan, the equipment depreciation cost is 153,280 yuan, and the annual comprehensive operation cost of the distribution network construction is 165,360 yuan.

[0065] The improved PSO algorithm is used to optimize the distribution network architecture. The optimized distribution network structure is as follows: Figure 7 shown.

[0066] In order to verify that the improved PSO algorithm proposed in this invention has advantages over the basic PSO algorithm in terms of fast convergence and result optimization, the number of iterations of the basic PSO algorithm and the improved PSO algorithm in the distribution network architecture optimization design are compared. The comparison results are as follows: Figure 8 shown.

[0067] according to Figure 8 It can be seen that the improved PSO algorithm converges after 53 iterations, which is about 40% less than the basic PSO algorithm (92 times). According to the meaning of the parameters represented by the fitness function, the improved PSO algorithm has better performance, and the improved PSO algorithm has obvious advantages over the basic PSO algorithm in terms of convergence and fitness. The improved PSO algorithm has much fewer iterations and convergence speed than the basic PSO algorithm, and consumes a shorter optimization calculation time to obtain the global optimal result.

[0068] After the basic particle swarm algorithm and the improved particle swarm algorithm optimize the distribution network architecture design, the key indicators are compared. Fig. 9 The specific indicators are shown in Table 2.

[0069] Table 2. PSO algorithm key value comparison table , according to Fig. 9 As shown in Table 2, compared with the basic PSO algorithm, the improved PSO algorithm further reduces the network loss and corresponding costs and equipment depreciation in the operation of the distribution network. The comparison shows that the annual comprehensive cost of the improved PSO algorithm is reduced by 10.3%, which ensures the economic benefits of the distribution network structure optimization design and effectively reduces the distribution network operation cost. The improved PSO algorithm is used to optimize the key indicators in the distribution network architecture design, which has significant economic benefits.

[0070] Optimizing the design of the distribution network architecture is an important basis for improving the power supply quality and reliability of the distribution network system, and is an indispensable part of the power transmission and use process. The present invention takes the mid-term comprehensive cost of the distribution network design as the optimization target, optimizes the distribution network design process and key indicators using the particle swarm optimization algorithm, and improves the basic PSO algorithm framework by introducing a linear decreasing inertia weight and an improved acceleration factor update method.

[0071] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for optimizing energy storage architecture, characterized in that: include: According to the optimization goal of the distribution network architecture, the particle position and velocity of the particle swarm are initialized, and the initial parameters of each particle are set; Construct a fitness function, calculate the fitness value for the current state of each particle, and record the global optimal position; The speed and position of each particle are dynamically adjusted by introducing a linearly decreasing inertia weight and an acceleration factor update mechanism that balances global and local search capabilities; When the preset termination conditions are met, the optimal design solution of the distribution network architecture is output.

2. The energy storage architecture optimization method according to claim 1, characterized in that: The optimization objectives of the distribution network architecture include minimizing the annual comprehensive cost, minimizing the network loss, and minimizing the equipment depreciation cost; The initial parameters of each particle include position, speed, inertia weight, acceleration factor, random number, individual optimal position and global optimal position.

3. The energy storage architecture optimization method according to claim 2, characterized in that: Construct a fitness function based on the key indicators and constraints of the distribution network; The constraints include node voltage value constraints, line current constraints, load capacity constraints and radiation network structure constraints; The key indicators of the distribution network include annual comprehensive costs, network losses, equipment depreciation costs, node voltage, line current and load capacity.

4. The energy storage architecture optimization method according to claim 3, characterized in that: The linearly decreasing inertia weight is expressed as: ， in, represents the weight at iteration step t, represents the initial weight, represents the final weight, k represents the index of the current iteration step, Indicates the maximum number of iterations.

5. The energy storage architecture optimization method according to claim 4, characterized in that: The balanced global and local search capabilities include: The improved particle swarm optimization distribution network architecture design algorithm is adopted, and the formula is expressed as: ， in, represents the change with time t value, represents the change with time t value, express The value at the initial time, express The value at the initial time, express The target value at the end time, express The target value at the end time, t represents the time, and T represents the total time.

6. The energy storage architecture optimization method according to claim 5, characterized in that: according to and , , adjust particle speed and position in real time; The particle velocity update formula is: ， The particle position update formula is: ， in, represent the velocity and position of the ith particle at time t, respectively. is the best historical position of the individual particle, g is the global best position, and is a random number, is represented by the position of the ith particle at time t, is represented by the position of the ith particle at time t+1, Expressed as the velocity of the ith particle at time t+1.

7. The energy storage architecture optimization method according to claim 6, characterized in that: The termination condition includes, when the number of iterations reaches the upper limit set by the algorithm, stopping the optimization and outputting the current global optimal solution; If the change of the fitness function in 10 consecutive iterations is less than or equal to 1×10 −51 ×10 −5 , then the algorithm is judged to have converged, the iteration is stopped and the optimal solution is output; When the termination condition is not met, continue iterating until the condition is met.

8. A system for optimizing energy storage architecture, using a method for optimizing energy storage architecture as claimed in any one of claims 1 to 7, characterized in that: include: Particle initialization module, fitness calculation module, particle update module, and termination judgment module; The particle initialization module initializes the particle position and velocity of the particle swarm according to the optimization goal of the distribution network architecture, and sets the initial parameters of each particle; The fitness calculation module constructs a fitness function based on the key indicators of the distribution network, calculates the fitness value of each particle's current state, and records the global optimal position; The particle update module dynamically adjusts the speed and position of each particle to move closer to the global optimal solution by introducing a linearly decreasing inertia weight and an acceleration factor update mechanism that balances global and local search capabilities; The termination judgment module outputs the optimal design solution of the distribution network architecture when the preset termination conditions are met.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the energy storage architecture optimization method described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of an energy storage architecture optimization method described in any one of claims 1 to 7 are implemented.

Images