Optimal configuration method of reusability treatment equipment based on hybrid branch and bound method
By applying hybrid branches in the configuration of reuse governance equipment, combined with the optimization strategy of genetic algorithms and branches limit laws, it solves the problem of configuration of reused governance equipment in complex electrical environments, and realizes efficient and accurate optimal configuration Essence
Patent Information
- Application Number
- CN202411513610.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-10-28
- Publication Date
- 2025-01-24
AI Technical Summary
The prior art is difficult to achieve the optimal configuration of reusable governance equipment in a complex and changeable electrical environment. Traditional configuration methods rely on empirical judgment or trial and error processes, making it difficult to ensure the optimal solution.
Using the optimal configuration method based on the multi -branch boundaries method, through the definition of problem modeling and target functions, use the genetic algorithm to generate an initial solution, and combine the branch limit method for in -depth optimization. Later, the intermediate solution was selected as a new initial type of genetic algorithm, and the feedback optimization results formed a closed -loop optimization process.
It significantly improves the configuration efficiency and accuracy of reuse governance equipment, ensures the optimality of the configuration scheme, improves the adaptability and robustness of the algorithm, and provides new technologies for the efficient and accurate configuration of reuse governance equipment. support.
Smart Images
Figure CN119358945A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of reusability management equipment configuration, and in particular to an optimal configuration method for reusability management equipment based on a hybrid branch and bound method. Background Art
[0002] With the continuous advancement of technology and its increasing application, the importance of reusable governance equipment as a key component to improve system performance and reduce operation and maintenance costs has become increasingly prominent. This type of equipment is usually designed to be flexibly configured in a variety of application scenarios to meet the specific needs of different electrical systems, such as power distribution, harmonic governance, reactive power compensation, etc. However, how to achieve the optimal configuration of reusable governance equipment for complex and changing electrical environments has become a technical problem that needs to be solved urgently.
[0003] Traditional configuration methods often rely on expert experience or simple trial and error, which not only limits the innovation of configuration solutions, but also makes it difficult to find the optimal solution globally. With the rapid development of computing science, a series of advanced optimization algorithms have emerged, such as genetic algorithms and branch and bound methods, which have shown great potential in solving complex optimization problems.
[0004] Genetic algorithm, as a search algorithm that simulates the biological evolution process, is good at quickly locating the approximate optimal solution in a vast solution space, and its global search capability has been widely recognized. The branch and bound method is another effective means of solving optimization problems by systematically enumerating all possible solutions and gradually narrowing the search range until the exact optimal solution is found.
[0005] However, when faced with the specific task of optimizing the configuration of reusable governance equipment, the application of a single algorithm often has limitations. The configuration of reusable governance equipment not only involves the fine adjustment of multiple parameters, but also requires comprehensive consideration of multiple dimensions such as the overall performance, economic cost, and scalability of the system. It is a highly complex multi-objective, multi-constraint optimization problem. Summary of the invention
[0006] In order to solve the problems existing in the prior art, the present invention proposes an optimal configuration method for reusable governance equipment based on a hybrid branch and bound method. First, the use scenarios and performance requirements of the reusable governance equipment are deeply analyzed, a problem model is established, and a clear objective function is defined to evaluate the pros and cons of the configuration scheme. Next, a genetic algorithm is used to generate an initial solution set with diversity, providing a basis for subsequent optimization. Then, the branch and bound method is used to deeply optimize the initial solution set, and the optimal configuration scheme is gradually approached through iterative search and bound pruning. On this basis, a hybrid strategy is innovatively introduced: after each iteration of the branch and bound method, some intermediate solutions are selected as the new initial population of the genetic algorithm to further optimize the configuration of the reusable governance equipment, and the optimization results are fed back to the branch and bound method to form a closed-loop optimization process. This method not only improves the optimization efficiency, but also ensures the optimality of the configuration scheme, providing new technical support for the efficient and accurate configuration of reusable governance equipment.
[0007] The technical solution adopted by the present invention is:
[0008] A method for optimal configuration of reusability management equipment based on a hybrid branch and bound method, characterized by comprising the following steps:
[0009] Step S1: problem modeling and objective function definition;
[0010] First, a mathematical model is constructed to comprehensively consider the instability of distributed energy output, the dynamic changes in load demand, and the comprehensive impact of these factors on power quality. Then, an objective function is defined.
[0011] Step S2: Generate an initial solution set based on a genetic algorithm;
[0012] Step S3: Branch and bound optimization.
[0013] Furthermore, the step S1 specifically includes:
[0014] S11: define input data;
[0015] Distributed energy output is P DE (t), represents the energy output at different time points t; the load fluctuation data set is P L (t), represents the load demand at different time points t; the power quality index requirement is Q req , represents the standard that the power quality needs to meet;
[0016] S12: Conduct problem modeling;
[0017] Define n management devices to form a configuration plan x = (x1, x2, ..., x n), x represents the configuration status of the governance equipment, specifically 0 or 1, 0 means no configuration, 1 means configuration, each x i They are all binary variables, and the cost of governance equipment is defined as C i , represents the cost of the i-th control equipment, and defines the power quality improvement effect as Q imp (x,P DE ,P L ), reflecting the degree of improvement of power quality under a given configuration;
[0018] S13: Define the objective function to balance the economic efficiency of the control equipment configuration and the improvement of power quality. The objective function is as follows:
[0019]
[0020] Among them, st represents constraint, To minimize the total cost of treatment equipment;
[0021] Furthermore, the step S2 specifically includes:
[0022] S21: Randomly generate individuals;
[0023] For every individual x in the population j (j=1,2,...,m),x j There are n configuration variables x i A specific example of a configuration scheme composed of a management equipment matching scheme, randomly selecting an access point p i (i=1,2,...,n), for the selected access point, in its corresponding capacity range [c min ,c max ] randomly generates a capacity value c ij , repeat the above process until individual x j All or specified access points and corresponding capacity are allocated, individual x j It can be expressed as a vector (p j1 ,c j1 ,p j2 ,c j2 ,...,p jk ,c jk ), where k is the number of access points contained in the individual;
[0024] S22: Fitness evaluation, for each individual x j , calculate its corresponding fitness g(x j ); The fitness g is based on the comprehensive management results of economy and power quality, specifically:
[0025]
[0026] Among them, α and β are weight coefficients used to balance the proportion of economy and power quality in the fitness function; cost(p ji ) is the access point p ji Unit governance cost; q(p ji ,c ji ) is at access point p ji With capacity c ij Power quality indicators during treatment;
[0027] S23: Form the initial population, and take all generated individuals x1, x2, ..., x m Form an initial population P pop ={x1,x2,...,x m}.
[0028] Furthermore, the step S3 specifically includes:
[0029] S31: branch;
[0030] For each individual x in the initial population j , generating a set of child nodes {x j1 ,x j2 ,...,x jnj}, where n j is x j The number of child nodes generated, each child node represents x j A configuration change of
[0031] S32: Bounds; for each child node x jh (h=1,2,...,n j ), calculate its objective function value f(x jh ),
[0032]
[0033] Among them, C i is the cost of the i-th governance equipment, x i is a binary variable, under the constraint Q imp (x,P DE ,P L )≥Q req Next proceed;
[0034] If f(x jh )>f(x best ), then prune, that is, no longer expand to x jh is the subtree of the root;
[0035] S33: Iteration and update; repeat steps S31 and S32 for the unpruned child nodes. If a better solution is found in a certain iteration, that is, there is x jh So that f(x jh )<f(x best ), then update x best =x jh and f(x best )=f(x jh );
[0036] S34: Termination condition: When all nodes have been processed (i.e., no more child nodes can be generated or all child nodes have been pruned), or when the preset number of iterations and time limit conditions are reached, the algorithm terminates and outputs the current optimal solution, which is recorded as x best , that is, the optimal configuration solution is obtained.
[0037] Preferably, a hybrid strategy of step S4 is also included; the hybrid strategy means that after each iteration of step S33, some intermediate solutions are selected as new starting points of the genetic algorithm, and then a new candidate solution set is obtained through selection, crossover and mutation operations of the genetic algorithm, which is fed back to step S3 to continuously promote the optimization process and obtain a better configuration solution.
[0038] Furthermore, the step S4 specifically includes:
[0039] Suppose the intermediate solution set generated by the lth iteration of the branch and bound method is S l ={x1,x2,...,x n}, calculate the fitness of each solution and arrange them in descending order, and select z solutions as the new initial population P0 = {x i1 ,x i2 ,...,x iz}, where x iz ∈S l , the first character of the subscript "i" is a general identifier, indicating that these paths belong to the first version or iteration of the initial population P0, and the second character of the subscript "z" is the specific index or number of these paths in P0. The calculation formula of z is:
[0040]
[0041] Among them, Size_ratio is a proportional coefficient, which indicates the ratio of the expected initial population size to the intermediate solution set size n; Diversity_factor is a diversity factor used to adjust the value of z to increase the diversity of the population; Min_z is a minimum z value;
[0042] Perform genetic algorithm operations on P0:
[0043] (1) Selection: Select excellent individuals to form the parent population P′ according to the fitness function g(x);
[0044] (2) Crossover: Perform a crossover operation on the individuals in P′ to generate a descendant population C;
[0045] (3) Mutation: Perform mutation operations on the individuals in C to obtain a new candidate solution set N = {y1, y2, ..., y y}, and feed the new candidate solution set N back to step S3, i.e. S l +1 = S l ∪N, continue the optimization iteration.
[0046] The beneficial effects of the present invention are:
[0047] 1. Improve configuration efficiency and accuracy: This invention significantly improves the configuration efficiency and accuracy of reusable governance equipment by integrating the hybrid optimization strategy of genetic algorithm and branch and bound method. The global search capability of genetic algorithm ensures the diversity of the initial solution set, while the precise solution of branch and bound method ensures the optimality of the final configuration solution. The combination of the two makes the configuration process more efficient and the results more accurate.
[0048] 2. Enhance algorithm adaptability and robustness: Faced with the complexity and diversity of reusability management equipment configuration in the field of electrical engineering, the hybrid optimization strategy of the present invention demonstrates strong adaptability and robustness. The global search characteristics of the genetic algorithm enable the algorithm to flexibly respond to configuration requirements in different scenarios, while the precise solution of the branch and bound method ensures the stability of the algorithm in the face of complex constraints and goals. In addition, through continuous iteration and optimization of the hybrid strategy, the algorithm can also gradually learn and adapt to new configuration environments and requirements, thereby further improving its performance in practical applications.
[0049] 3. The invention not only opens up a new path for the configuration optimization of reusable governance equipment, but also provides useful inspiration for complex optimization problems in other fields. With the continuous advancement of technology and the in-depth expansion of applications, this hybrid optimization strategy is expected to play an important role in a wider range of fields, promoting the continuous improvement of system performance and the continuous optimization of resource allocation. BRIEF DESCRIPTION OF THE DRAWINGS
[0050] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0051] Figure 1It is an implementation flow chart of the present invention;
[0052] Figure 2 This is a topology diagram of a 13-node power distribution system. DETAILED DESCRIPTION
[0053] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0054] like Figure 1 As shown, a method for optimal configuration of reusability management equipment based on a hybrid branch and bound method includes the following steps:
[0055] Step S1: problem modeling and objective function definition;
[0056] First, a comprehensive mathematical model is constructed, which needs to comprehensively consider the instability of distributed energy output, the dynamic changes in load demand, and the combined impact of these factors on power quality. To this end, three core data need to be collected and analyzed: real-time output data of distributed energy, load fluctuations, and standards that power quality must meet. Subsequently, an objective function is defined, the core of which is to seek a balance between two key objectives: on the one hand, to minimize the configuration cost of the governance equipment, and on the other hand, to ensure a significant improvement in power quality. Through this comprehensive modeling and objective function setting, not only an initial mathematical model containing all necessary variables is formed, but also a clear direction is provided for the subsequent application of the hybrid branch and bound method for optimization, ensuring the efficiency and practicality of the entire configuration method.
[0057] (1) Define input data;
[0058] Distributed energy output is P DE (t) (represents the energy output at different time points t), the load fluctuation data set is P L (t) (representing the load demand at different time points t), and the power quality index requirement is Q req (represents the standard that the power quality needs to meet);
[0059] (2) Conduct problem modeling;
[0060] Define n management devices to form a configuration plan x = (x1, x2, ..., x n ), indicating the configuration status of the governance equipment (0 or 1, 0 means not configured, 1 means configured, each x iare both binary variables), define the cost of governance equipment C i , represents the cost of the i-th control equipment, and defines the power quality improvement effect Q imp (x,P DE ,P L ), which reflects the degree of improvement in power quality under a given configuration.
[0061] (3) Define the objective function to balance the economic efficiency of the control equipment configuration and the improvement of power quality:
[0062]
[0063] Among them, st represents constraint, To minimize the total cost of the treatment equipment, that is, to minimize the total cost of the treatment equipment, while ensuring that the improvement effect of power quality is not lower than the required standard.
[0064] Step S2: Generate an initial solution set based on a genetic algorithm;
[0065] First, a number of individuals are randomly generated to form the initial population. Each individual represents a governance equipment configuration scheme, which specifically includes the selection of each candidate access point and the corresponding capacity setting. Then, the fitness of each individual is evaluated, that is, according to the preset objective function (such as considering equipment cost, power quality improvement, etc.), the specific value under the configuration scheme is calculated, that is, the initial set is generated, which serves as the basis for subsequent operations to promote the generation of better solutions.
[0066] Define the candidate access point set P of distributed management equipment = {p1, p2, ..., p n}, define the management equipment capacity range set C corresponding to each access point = {c1,c2,...,c n}, where c i =[c min ,c max ] represents the maximum and minimum values of the management equipment capacity of the i-th access point.
[0067] (1) Randomly generate individuals. For each individual x in the population j (j=1,2,...,m),x j There are n configuration variables x i A specific example of a configuration scheme composed of a management equipment matching scheme, randomly selecting an access point p i (i=1,2,...,n), for the selected access point, in its corresponding capacity range [c min ,c max ] randomly generates a capacity value c ij Repeat the above process until the individual x jAll or specified access points and corresponding capacities are allocated. j It can be expressed as a vector (p j1 ,c j1 ,p j2 ,c j2 ,...,p jk ,c jk ), where k is the number of access points contained in the individual.
[0068] (2) Fitness evaluation. For each individual x j , calculate its corresponding fitness g(x j ), the fitness g is usually based on the comprehensive management results of economic type and power quality, which can be expressed as:
[0069]
[0070] Among them, α and β are weight coefficients used to balance the proportion of economy and power quality in the fitness function; cost(p ji ) is the access point p ji Unit governance cost; q(p ji ,c ji ) is at access point p ji With capacity c ij Power quality indicators during governance.
[0071] (3) Form the initial population. All generated individuals x1, x2, ..., x m Form an initial population P pop ={x1,x2,...,x m}.
[0072] Step S3: Branch and bound optimization;
[0073] Based on the initial population generated by the genetic algorithm, a deep search is performed on each individual (i.e., configuration scheme). Through branching operations, starting from the current individual, all possible configuration changes are explored to generate a series of child nodes. Next, each child node is quickly screened using the limit conditions, its objective function value is calculated, and compared with the currently known optimal solution. If the solution quality of the child node is poor (such as the objective function value is greater than the current optimal solution or exceeds the preset threshold), pruning is performed to avoid invalid calculations. This process is iterated until all unpruned nodes are traversed. Finally, we record and update the optimal solution of the current stage, i.e., the optimal configuration scheme.
[0074] Let the current optimal solution be x best , and its corresponding objective function value is Under the constraint Q imp (x,P DE ,PL )≥Q req Next proceed.
[0075] (1) Branch. For each individual x in the initial population j , generating a set of child nodes where n j is x j The number of child nodes that may be generated, each child node represents x j A configuration change (such as adjusting access points or capacity).
[0076] (2) Bounds. For each byte x jh (h=1,2,...,n j ), calculate its objective function value f(x jh ),
[0077]
[0078] If f(x jh )>f(x best ), then prune, that is, no longer expand to x jh The subtree with root.
[0079] (3) Iteration and update. Repeat the branching and bounding process for the unpruned child nodes. If a better solution is found in a certain iteration, then there exists x jh So that f(x jh )<f(x best ), then update x best =x jh and f(x best )=f(x jh ).
[0080] (4) Termination condition: When all nodes have been processed (i.e., no more child nodes can be generated or all child nodes have been pruned), or when the preset number of iterations, time limit, and other conditions are reached, the algorithm terminates and outputs the current optimal solution x best , that is, to obtain the optimal.
[0081] Step S4: Mixed strategy;
[0082] After each iteration, some intermediate solutions are selected as the new starting point of the genetic algorithm. Then, through the selection, crossover, and mutation operations of the genetic algorithm, a new set of candidate solutions is obtained and fed back to the branch and bound method to continuously promote the optimization process and obtain a better configuration solution.
[0083] Suppose the intermediate solution set generated by the lth iteration of the branch and bound method is S l ={x1,x2,...,x n}, calculate the fitness of each solution and arrange them in descending order, and select z solutions as the new initial population P0 = {x i1 ,x i2 ,...,x iz}, where x iz ∈S l , the first character of the subscript "i" is a general identifier, indicating that these paths belong to the first version or iteration of the initial population P0, and the second character of the subscript "z" is the specific index or number of these paths in P0. The calculation formula of z is:
[0084]
[0085] Among them, Size_ratio is a proportional coefficient, which represents the ratio of the expected initial population size to the size of the intermediate solution set n; Diversity_factor is a diversity factor, which is used to adjust the value of z to increase the diversity of the population, and can be adjusted according to the complexity of the problem and the dimension of the search space; higher Diversity_factor values lead to smaller z values, thereby increasing the diversity of the population; Min_z is a minimum z value, which is used to ensure that at least some solutions are selected as the new initial population even when the intermediate solution set is small or the Diversity_factor is high.
[0086] Perform genetic algorithm operations on P0:
[0087] (1) Select the parent population; select excellent individuals to form the parent population P′ according to the fitness function g(x).
[0088] For each individual in the current population, the number is z, so it can be expressed as calculating its fitness value g(x u ), calculate the total fitness of the population:
[0089]
[0090] For each individual, calculate the probability of being selected:
[0091]
[0092] Next, the procedure shown in Table 1 is performed to select the parent population P′.
[0093] Table 1. The selection process of the parent population P′
[0094]
[0095] (2) Crossover: Perform a crossover operation on the individuals in P′, randomly select a crossover point in the gene sequence of the parent individual, and then exchange the gene fragments after this point. Generate two offspring individuals. Suppose x1' and x'2 are selected for the crossover operation, x1'=(p 11 ,c 11 ,p 12 ,c 12 ,...,p 1k ,c 1k ), x'2=(p 21 ,c 21 ,p 22 ,c 22 ,...,p 2k ,c 2k ), k is the number of access points contained in the individual, the randomly selected crossover point is a, 1≤a≤k, and the two offspring individuals after the crossover are c1=(p 11 ,c 11 ,...,p 2a ,c 2a ,...,p 2k ,c 2k ) and c2=(p 21 ,c 21 ,...,p 1a ,c 1a ,...,p 1k ,c 1k ), the other parents that have not undergone crossover are automatically transferred to the offspring, forming the offspring population C.
[0096] (3) Mutation: Randomly select an individual c from C b Perform mutation operation, randomly select an access point, and then mutate the access point and its corresponding capacity value at the same time. Suppose we choose p b2 and c b2 The new access point and capacity values are p' b2 and c' b2 , the individual after mutation is y b =(p b1 ,c b1 ,p b '2,c b '2,...,p bk ,c bk ), the other offspring that have not undergone mutation operation are automatically transferred to the new candidate solution set. For example, if y1=c1, the new candidate solution set is N={y1,y2,...,y y}, finally, the new candidate solution set N is fed back to step S3, that is, S l +1 = S l ∪N, continue the optimization iteration.
[0097] In order to enable those skilled in the art to better understand the present invention, the example analysis includes the following components:
[0098] 1. Case description and simulation results analysis
[0099] The present invention adopts Figure 2 The system shown is used as an example analysis to verify the effectiveness of the optimal configuration method of the reusable power quality management equipment proposed in the present invention. The system reference voltage is 6.6kV, the basic capacity is 10MVA, and the parameters and loads of each branch are shown in Table 2.
[0100] Table 2 Distribution network branch parameters and loads
[0101] Branch Road Impedance / Ω node Load / kVA <![CDATA[L1]]> 0.34+j0.16 1 100+j35 <![CDATA[L2]]> 0.30+j0.29 2 90+j40 <![CDATA[L3]]> 1.06+j0.71 3 120+j50 <![CDATA[L4]]> 0.45+j0.18 4 60+j30 <![CDATA[L5]]> 0.24+j0.13 5 200+j80 <![CDATA[L6]]> 0.24+j0.10 6 60+j20 <![CDATA[L7]]> 0.70+j0.34 7 60+j35 <![CDATA[L8]]> 0.59+j0.21 8 60+j35 <![CDATA[L9]]> 1.10+0.75 9 60+j45 <![CDATA[L 10 ]]> 0.70+j0.41 10 120+j10 <![CDATA[L 11 ]]> 0.71+j0.24 11 100+j30 <![CDATA[L 12 ]]> 0.82+j0.58 12 60+j10 13 90+j35
[0102] The power quality problems existing in the system are as follows: ① The loads at nodes 3, 9, and 13 will input high-content voltage and current harmonics into the system; ② Voltage fluctuations will occur at nodes 5, 6, and 10, with the maximum fluctuation rate exceeding 30%. The power quality problems existing in the distribution network are shown in Table 3. Therefore, it is necessary to install the reusable power management device active power filter APF (Active Power Filter) and distribution network static VAR compensator DSVC (Distribution Static Var Compensator) on the distribution network to solve the problems of harmonics and voltage fluctuations respectively. The improvement requirements for each node with problems are shown in Table 4.
[0103] Table 3 Power quality problems in distribution network
[0104]
[0105]
[0106] Table 4 Requirements for improving power quality in distribution networks
[0107]
[0108] Table 5 Optimization configuration results
[0109]
[0110] Table 6 Comparison of power quality optimization results
[0111]
[0112]
[0113] When the present invention is used for optimization configuration, each individual of the branch and bound method generates 5 child nodes, the number of iterations is 100, Size_ratio=0.2, Diversity_factor=1.5, Min_z=5, the genetic algorithm population size is set to 100, the number of iterations is 100, the selection probability is 0.8, the crossover probability is 0.7, and the mutation probability is 0.1. The results of the optimization configuration using the present invention and the ordinary optimization configuration are shown in Table 5, and the power quality optimization results after the configuration device are compared with those before optimization, and the comparison results are shown in Table 6. From the comparison results, it can be seen that the current and voltage harmonic distortion rates are smaller and the voltage fluctuation is smaller when the optimization configuration method of the present invention is used.
[0114] The above description is only a preferred embodiment of the present invention and is not intended to limit the protection scope of the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention are included in the protection scope of the present invention.
Claims
1. A method for optimal configuration of reusability management equipment based on a hybrid branch and bound method, characterized in that: The following steps are involved: Step S1: problem modeling and objective function definition; First, a mathematical model is constructed to comprehensively consider the instability of distributed energy output, the dynamic changes in load demand, and the comprehensive impact of these factors on power quality. Then, an objective function is defined. Step S2: Generate an initial solution set based on a genetic algorithm; Step S3: Branch and bound optimization.
2. The optimal configuration method for reusability management equipment based on the hybrid branch and bound method as claimed in claim 1, characterized in that: The step S1 specifically includes: S11: define input data; Distributed energy output is P DE (t), represents the energy output at different time points t; the load fluctuation data set is P L (t), represents the load demand at different time points t; the power quality index requirement is Q req , represents the standard that the power quality needs to meet; S12: Conduct problem modeling; Define n management devices to form a configuration plan x = (x1, x2, ..., x n ), x represents the configuration status of the governance equipment, specifically 0 or 1, 0 means no configuration, 1 means configuration, each x i are both binary variables; define the cost of governance equipment as C i , represents the cost of the i-th control equipment; the power quality improvement effect is defined as Q imp (x,P DE ,P L ), reflecting the degree of improvement of power quality under a given configuration; S13: Define the objective function to balance the economic efficiency of the control equipment configuration and the improvement of power quality. The objective function is as follows: Among them, st represents constraint, To minimize the total cost of governance equipment.
3. The optimal configuration method for reusability management equipment based on the hybrid branch and bound method as claimed in claim 1, characterized in that: The step S2 specifically includes: S21: Randomly generate individuals; For every individual x in the population j (j=1,2,...,m),x j There are n configuration variables x i A specific example of a configuration scheme composed of a management equipment matching scheme, randomly selecting an access point p i (i=1,2,...,n), for the selected access point, in its corresponding capacity range [c min ,c max ] randomly generates a capacity value c ij , repeat the above process until individual x j All or specified access points and corresponding capacity are allocated, individual x j It can be expressed as a vector (p j1 ,c j1 ,p j2 ,c j2 ,...,p jk ,c jk ), where k is the number of access points contained in the individual; S22: Fitness evaluation, for each individual x j , calculate its corresponding fitness g(x j ); The fitness g is based on the comprehensive management results of economy and power quality, specifically: Among them, α and β are weight coefficients used to balance the proportion of economy and power quality in the fitness function; cost(p ji ) is the access point p ji Unit governance cost; q(p ji ,c ji ) is at access point p ji With capacity c ij Power quality indicators during treatment; S23: Form the initial population, and take all generated individuals x1, x2, ..., x m Form an initial population P pop ={x1,x2,...,x m }.
4. The optimal configuration method for reusability management equipment based on the hybrid branch and bound method as claimed in claim 1, characterized in that: The step S3 specifically includes: S31: branch; For each individual x in the initial population j , generating a set of child nodes {x j1 ,x j2 ,...,x jnj }, where n j is x j The number of child nodes generated, each child node represents x j A configuration change of S32: Bounds; for each child node x jh (h=1,2,...,n j ), calculate its objective function value f(x jh ), Among them, C i is the cost of the i-th governance equipment, x i is a binary variable, under the constraint Q imp (x,P DE ,P L )≥Q req Next proceed; If f(x jh )>f(x best ), then prune, that is, no longer expand to x jh is the subtree of the root; S33: Iteration and update; repeat steps S31 and S32 for the unpruned child nodes. If a better solution is found in a certain iteration, that is, there is x jh So that f(x jh )<f(x best ), then update x best =x jh and f(x best )=f(x jh ); S34: Termination condition: When all nodes have been processed (i.e., no more child nodes can be generated or all child nodes have been pruned), or when the preset number of iterations and time limit conditions are reached, the algorithm terminates and outputs the current optimal solution, which is recorded as x best , that is, the optimal configuration solution is obtained.
5. The optimal configuration method for reusability management equipment based on the hybrid branch and bound method as claimed in claim 4, characterized in that: It also includes a hybrid strategy in step S4; the hybrid strategy means that after each iteration of step S33, some intermediate solutions are selected as new starting points of the genetic algorithm, and then a new set of candidate solutions is obtained through the selection, crossover and mutation operations of the genetic algorithm, which are fed back to step S3 to continuously promote the optimization process and obtain a better configuration solution.
6. The optimal configuration method for reusability management equipment based on the hybrid branch and bound method as claimed in claim 5, characterized in that: The step S4 specifically includes: Suppose the intermediate solution set generated by the lth iteration of the branch and bound method is S l ={x1,x2,...,x n }, calculate the fitness of each solution and arrange them in descending order, and select z solutions as the new initial population P0 = {x i1 ,x i2 ,...,x iz }, where x iz ∈S l , the first character "i" of the subscript is a general identifier, indicating that these paths belong to the first version or iteration of the initial population P0, and the second character "z" of the subscript is the specific index or number of these paths in P0. The calculation formula of z is: Among them, Size_ratio is a proportional coefficient, which indicates the ratio of the expected initial population size to the intermediate solution set size n; Diversity_factor is a diversity factor used to adjust the value of z to increase the diversity of the population; Min_z is a minimum z value; Perform genetic algorithm operations on P0: (1) Selection: Select excellent individuals to form the parent population P′ according to the fitness function g(x); (2) Crossover: Perform a crossover operation on the individuals in P′ to generate a descendant population C; (3) Mutation: Perform mutation operations on the individuals in C to obtain a new candidate solution set N = {y1, y2, ..., y y }, and feed the new candidate solution set N back to step S3, i.e. S l +1 = S l ∪N, continue the optimization iteration.