New energy automobile power battery echelon utilization site selection and network design method based on improved genetic algorithm

Abstract

The invention discloses a new energy automobile power battery echelon utilization site selection and network optimization design method based on an improved genetic algorithm, and the method comprises the following steps: (1) constructing a new energy automobile power battery recovery network cost model, and calculating a site selection cost scheme at a first stage; (2) obtaining the network center site selection cost through the site selection scheme, and forming a mixed integer programming model in a second stage by taking the lowest network logistics freight cost as an objective function and taking the logistics route capacity and the center capacity as constraint conditions; (3) solving the two-stage mixed integer programming model by adopting an improved genetic algorithm to obtain a solving result; and (4) analyzing according to a solving result to obtain a power battery recovery site selection and network optimization design scheme. The cost can be reduced, the efficiency is increased, and the utilization rate of each center is improved.

Description

Technical Field

[0001] The present invention relates to the field of new energy vehicle power battery recycling, and in particular to a new energy vehicle power battery cascade utilization site selection and network optimization design method based on an improved genetic algorithm. Background Art

[0002] Competition among new energy vehicle companies is becoming increasingly fierce. The global shortage of raw materials for automotive power batteries, coupled with site costs, construction costs, etc., has led to a gradual increase in the total costs of new energy vehicle companies and battery manufacturers. Moreover, the green attributes of new energy vehicles depend to a large extent on the recycling of materials, especially the cascade utilization of key components such as power batteries and the harmless treatment after scrapping. If power batteries are not handled properly, they will cause significant pollution to the environment and increase carbon emissions throughout the life cycle of the vehicle. Therefore, how to reduce costs and carbon emissions has become a practical issue that new energy vehicle companies urgently need to consider.

[0003] There are two main processing modes for retired batteries: recycling and disassembly and cascade utilization. Recycling and disassembly refers to directly crushing and disassembling power batteries in poor condition, extracting metals such as nickel and lithium from them and reusing them. Cascade utilization refers to disassembling retired batteries with acceptable performance to obtain battery modules, and then reassembling battery modules with good consistency to obtain cascade utilization battery products, which are used in communication base station energy storage, low-speed electric vehicles and other fields.

[0004] The network model for recycling used batteries generally includes three main parts: collection center, recycling and recollection center, and waste treatment center. The collection center will disassemble the various used batteries it has acquired, and then the disassembled small batteries will be tested and classified according to quality. Generally, the batteries in used batteries can be roughly divided into the following three categories according to their capacity: Level 1 (L1) – capacity is higher than 80% of the initial capacity; Level 2 (L2) – capacity is between 60% and 80% of the initial capacity; Level 3 (L3) – capacity is less than 60% of the initial capacity.

[0005] Based on the test results, different downstream processing strategies will be adopted for batteries of different qualities. The specific operations are as follows: (1) L1-level batteries will be transported to remanufacturing plants and reassembled into new electric vehicle batteries; (2) L2-level batteries will be transported to recycling plants. Some batteries will be transported to remanufacturing plants and directly reused as batteries, while other L2-level batteries in recycling plants will be disassembled and transported to remanufacturing plants to manufacture new battery cells; (3) L3-level batteries and waste from remanufacturing plants and recycling plants will be directly transported to waste treatment centers.

[0006] Currently, there is a lack of site selection and network optimization technology for the cascade utilization of power batteries for new energy vehicles. Summary of the invention

[0007] The present invention aims to overcome the above-mentioned shortcomings of the prior art, and combines the power battery recycling network model to provide a new energy vehicle power battery cascade utilization site selection and network optimization design method based on an improved genetic algorithm. The method can determine the number and location of network nodes and the flow of batteries in the network, thereby minimizing the total cost of network operation, where the total cost includes transportation cost, fixed cost, procurement cost, processing cost and environmental cost related to carbon emissions.

[0008] The site selection and network optimization design method for the cascade utilization of new energy vehicle power batteries provided by the present invention is scientific and reasonable, quick to calculate, easy to use and highly versatile. It not only reduces the fixed and purchase costs of site selection for enterprises, but also saves the transportation costs and energy consumption costs required for logistics and transportation, and has practical application value.

[0009] In order to achieve the above object, the present invention provides a method for site selection and network optimization design of new energy vehicle power battery cascade utilization based on an improved genetic algorithm, comprising the following steps:

[0010] (1) Construct a power battery recycling network cost model and calculate the site selection cost plan in the first stage

[0011] The step (1) specifically refers to constructing a site selection cost model for the cascade utilization of power batteries for new energy vehicles, including purchase costs and fixed costs. In step (1), the objective function of the first stage is:

[0012] Z 1 =Z BC +Z FC (1)

[0013] In the formula, Z BC The total cost of purchasing batteries for recycling in the first stage collection center; Z FC The total fixed cost of site selection for each center in the first phase.

[0014]

[0015] QB k is the total amount of used batteries collected by collection center k; u is the percentage of batteries with different quality levels u (L1 / L2 / L3); r u is the unit purchase cost of batteries with different quality levels u (L1 / L2 / L3); FC k is the fixed cost of building the collection center k; FC j is the fixed cost of building the recycling and remanufacturing center j; FC h is the fixed cost of building the waste treatment center h; x k is a 0-1 variable, collecting center k if the open decision variable is 1, otherwise it is 0; xj is a 0-1 variable. If the recycling and remanufacturing center j is open, the decision variable is 1, otherwise it is 0; h It is a 0-1 variable. If the waste treatment center h is open, the decision variable is 1, otherwise it is 0.

[0016] (2) The network center location cost is obtained through the above site selection scheme. In the second stage, the minimum network logistics freight cost is taken as the objective function, and the logistics route capacity and center capacity are taken as constraints to form a mixed integer programming model.

[0017] In step (2), the second-stage objective function after determining the site selection plan is:

[0018] Z 2 =ZP C +Z TC +ZF EC +Z TEC (4)

[0019] In formula (4), Z PC is the total processing cost of each center in the battery recycling network; Z TC is the total cost of transporting batteries between centers in the battery recycling network; ZF EC The carbon emission costs generated by the processing of each center in the battery recycling network; Z TEC The carbon emission costs generated during the transportation of products between centers in the battery recycling network.

[0020] In formula (4),

[0021]

[0022] Among them, QCR kj QCM is the total amount of used batteries disassembled into small batteries by collection center k and flowing into recycling center j; kj QW is the total amount of waste batteries disassembled into small batteries by collection center k and then flowing into remanufacturing center j; jh is the total amount of waste flowing into waste treatment center h from recycling and remanufacturing center j; QCW kh is the total amount of waste batteries disassembled into small batteries by collection center k and then flowing into waste treatment center h; pcc k The processing cost of small cells per processing unit for the collection center k; pcr j The processing cost of a small battery per processing unit in a recycling center; pcm j Processing cost of small battery per unit in remanufacturing center j; pcw h The processing cost of a unit of small batteries processed by waste treatment center h; w is the ratio of the number of small batteries that flow to the remanufacturing center after waste utilization at the recycling center to the number received by the recycling center.

[0023] In formula (4),

[0024]

[0025] Among them, t cc is the transportation cost per battery; t cw is the transportation cost of unit waste; D kj is the distance between the collection center k and the recycling and remanufacturing center j; D kh is the distance between the collection center k and the waste treatment center h; D jh is the distance between the recycling and remanufacturing center j and the waste treatment center h.

[0026] In formula (4),

[0027]

[0028]

[0029] Among them, F ct fec is the carbon tax cost per unit of carbon emissions; k The carbon emissions generated by processing one unit of used batteries for collection center k; j The carbon emissions generated by processing one unit of small battery for a recycling center j; fem j The carbon emissions generated by processing one unit of small battery for a remanufacturing center j; few h The carbon emissions generated by processing one unit of waste or small battery for a waste treatment center h.

[0030] In formula (4),

[0031]

[0032] Among them, t e The carbon emissions generated for each kilometer of transporting a unit of small battery.

[0033] The entire mixed integer programming model has three main constraints: quantity, quality and capacity.

[0034] The quantity constraint means that the number of batteries flowing out of the collection center is consistent with the number of batteries flowing into the recycling and remanufacturing center and the waste treatment center. The formula is as follows:

[0035]

[0036] bc is the number of small batteries in a unit of used battery.

[0037] The quality constraint means that the quality of production waste flowing into the waste treatment center from the recycling and remanufacturing center is consistent with the quality of waste generated by the battery treatment that will flow into the recycling and remanufacturing center. The formula is as follows:

[0038]

[0039] w b Indicates the weight of small batteries after disassembling the waste batteries; It represents the ratio of the amount of waste flowing out of recycling and remanufacturing centers to the amount of waste flowing out of disposal centers.

[0040] Capacity constraint means that the processing capacity of each processing center is limited. The formula is as follows:

[0041]

[0042]

[0043] In the formula, K k is the maximum processing capacity of the collection center k; J j The maximum processing capacity of the recycling and remanufacturing center; h is the maximum processing capacity of waste treatment center h. Among them, the decision variables mainly include the flow of each center and the site selection decision point:

[0044] QB k , QCR kj , QCM kj , QW jh ≥0 (14)

[0045] x k , x j , x h ∈(0, 1)(15)

[0046] (3) The improved genetic algorithm is used to solve the two-stage mixed integer programming model and the solution is obtained.

[0047] The step (3) specifically includes the following steps: (a) initializing the site selection plan, initializing the number of battery requirements, the total number of iterations, and the number of populations; (b) using the dual structure rule to initialize the encoding and decoding chromosome pairs, i.e., the additional code and the 0-1 variable code, the site selection plan corresponding to the randomly generated chromosome needs to meet the capacity constraint, i.e., the capacity of the selected center must accommodate all the requirements, thereby generating an initial population; (c) according to the objective function of the first stage of step 2, calculating the fixed cost and purchase cost of the initial plan, according to the objective function of the second stage of step 2, using the linear programming tool linprog function to optimize the network design, accurately calculating the service and transportation costs of each center, and converting each The inverse of the sum of costs is used as the fitness value formula to calculate the fitness of the first generation population, and selection, crossover, and mutation operations are performed to form a new generation population; (d) A selection probability is randomly generated, and rule discrimination is used. Individuals are proportionally selected according to the given parameters to produce offspring using a tournament or roulette method to ensure population diversity and excellent gene retention; (e) A crossover probability is randomly generated, and a partial mapping crossover (PMX) operator is used to produce effective offspring; (f) A mutation probability is randomly generated, and an inverse genetic operator is used to produce effective offspring. Two mutation points are randomly selected for the parent individual, and the uplink additional code between the two points is rearranged in reverse order, while the downlink variable code order remains unchanged. (g) The most preferred addressing scheme, transportation route, number of batteries, and minimum cost in the current generation population are output; (h) Finally, after reaching the number of iterations, the optimal result after convergence is obtained.

[0048] (4) Based on the solution results, the site selection and network optimization design scheme for power battery recycling is obtained

[0049] Through the above technical scheme, the present invention develops a mixed integer programming model for the power battery recycling network cost considering carbon tax. The proposed method can select an effective and reasonable site selection scheme under a variety of known alternative schemes, and then give the optimal network optimization scheme based on the optimal site selection scheme.

[0050] It can be seen from the above technical solution that the beneficial features of the present invention are as follows:

[0051] First, the present invention can reduce costs and increase efficiency, thereby improving the utilization rate of each center;

[0052] Second, the present invention can balance the relationship between the site selection plan and the network optimization plan

[0053] Third, the present invention is beneficial for battery recycling companies to conduct market research in the initial stage and provide a reference for the implementation stage. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 This is a power battery recycling network diagram of the present invention.

[0055] Figure 2 It is a flow chart of the power battery recycling method of the present invention.

[0056] Figure 3 It is a framework diagram of the method of the present invention.

[0057] Figure 4 It is a scatter plot of each center position of the present invention.

[0058] Figure 5 It is the transportation route map of each central position after adjustment of the present invention.

[0059] Figure 6 It is a diagram of the operation results of the method of the present invention. DETAILED DESCRIPTION

[0060] In order to make the purpose, technical solution and advantages of the present invention clearer, the present invention is further described in detail below in conjunction with Example 1. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0061] Relevant information was found in Zhejiang Province, and a total of 23 treatment centers, 4 recycling and reuse centers, and 2 waste utilization centers were set as alternative sites. Due to lack of information, the battery pack demand and the percentage of graded batteries processed in each treatment center follow a normal distribution; in this study, hydrometallurgical recycling was considered. Relevant information was found, and the reuse quality index was set to 0.8, and the carbon emissions per ton of small battery transportation was 0.00249 kg / (ton·km).

[0062] The coordinate positions of different centers and the small battery capacities that can be processed are shown in Table 1:

[0063] Table 1 Latitude and longitude coordinates and capacity of each recycling center

[0064]

[0065]

[0066] The total specifications of the recycled battery packs for new energy vehicles are shown in Table 2 below:

[0067] Table 2 Battery pack specifications

[0068]

[0069] The battery pack requirements of each processing center and the percentage of graded batteries processed are shown in Table 3:

[0070] Table 3 Processing center requirements and battery classification information

[0071]

[0072] The costs required for each center are shown in Table 4 below:

[0073] Table 4 Costs of the processing center

[0074]

[0075] The carbon emissions data in the logistics network are shown in Table 5 below:

[0076] Table 5 Carbon emissions during processing at each center

[0077]

[0078]

[0079] like Figure 3 As shown, a method for optimizing economic dispatch of a power system containing renewable energy based on an improved genetic algorithm comprises the following steps in order:

[0080] (1) Obtain the above basic data information, encode the site selection plan with dual structure rules, and initialize the population;

[0081] (2) Determine whether the cut-off condition is met. If so, output the result; otherwise, enter the loop;

[0082] (3a) Based on the basic data information obtained from the initial site selection plan, the first stage objective function is calculated as follows:

[0083] Z 1 =Z BC +Z FC (1)

[0084] In the formula, Z BC The total cost of purchasing batteries for recycling in the first stage collection center; Z FC The total cost of constructing the sites for each center in the first phase.

[0085] (3b) Combined with the objective function of the first stage, the linear programming solver is used to solve the minimization problem of the objective function of the second stage under the premise of satisfying the constraints. The formula is as follows:

[0086] Z 2 =ZP C +Z TC +ZF EC +Z TEC (4)

[0087] In the formula, Z PC is the total processing cost of each center in the battery recycling network; Z TC is the total cost of transporting batteries between centers in the battery recycling network; ZF ECThe carbon emission costs generated by the processing of each center in the battery recycling network; Z TEC The carbon emission cost generated by the product transportation process between centers in the battery recycling network. The above formula indicators are all explained in detail in the invention content and will not be repeated here.

[0088] The entire mixed integer programming model has two main constraints: quantity, quality and capacity.

[0089] Quantity constraint: The number of batteries flowing out of the collection centers is consistent with the number of batteries flowing into the recycling and remanufacturing centers and waste treatment centers, with the following formula:

[0090]

[0091] b c The number of small batteries in a unit of used batteries.

[0092] Quality constraint: The quality of production waste flowing into the waste treatment center from the recycling and remanufacturing center is consistent with the quality of waste generated by the battery treatment that will flow into the recycling and remanufacturing center, and the formula is as follows:

[0093]

[0094] w b Indicates the weight of small batteries after disassembling the waste batteries; It represents the ratio of the amount of waste flowing out of recycling and remanufacturing centers to the amount of waste flowing out of disposal centers.

[0095] Capacity constraints: The processing capacity of each center is limited

[0096]

[0097] In the formula, K k is the maximum processing capacity of the collection center k; J j The maximum processing capacity of the recycling and remanufacturing center; h The maximum processing capacity of the waste treatment center.

[0098] Among them, the decision variables mainly include the flow of each center and the location decision point:

[0099] QB k , QCR kj , QCM kj , QW jh ≥0 (14)

[0100] x k , x j , x h ∈(0, 1)(15)

[0101] (3c) The two-stage objective functions are summed and the inverse is taken as the fitness value, which is applied to the iteration of the heuristic method, hoping to obtain the population with the maximum fitness value.

[0102] (4) Randomly generate a selection probability, use rule discrimination, and select individuals proportionally according to established parameters to produce offspring using a tournament or roulette method to ensure population diversity and retention of excellent genes;

[0103] (5) Randomly generate a crossover probability and use the partial mapping crossover (PMX) operator to generate valid offspring;

[0104] (6) A mutation probability is randomly generated, and the inverse genetic operator is used to generate effective offspring. Two mutation points are randomly selected from the parent individual, and the upstream additional codes between the two points are rearranged in reverse order, while the downstream variable code order remains unchanged.

[0105] (7) Output the most optimal location solution, transportation route, number of batteries and minimum cost in the current generation population;

[0106] (8) Repeat steps 3-7 until the cutoff condition is met, and obtain the optimal location solution and logistics network optimization solution after convergence.

[0107] In order to reflect the optimization and reproducibility of the method, under the condition that the maximum number of iterations is set to 150 and the normal distribution input is consistent (the random number seed is 50), an experimental design is carried out for the combination of initial population size (50, 100, 150, 200), crossover probability (0.6, 0.7, 0.8, 0.9), selection probability (championship: roulette) (0.2, 0.4, 0.6, 0.8) and mutation probability (0.005, 0.01, 0.05, 0.1). The experiments are repeated many times in the hope of obtaining the parameter combination with the optimal results.

[0108] like Figure 4 , 5 As shown in Figure 6, when the initial population size is 150, the crossover probability is 0.7, the mutation probability is 0.1, and the selection probability is 0.6, the number of running iterations is increased to 200 times, and the optimal addressing scheme for this instance is solved as [L1, L2, L3, L4, L5, L6, L7, L10, L11, L12, L13, L14, L15, L16, L19, Z24, Z26, Z27, W28, W29]. A total of 15 processing centers, 3 recycling and reuse centers and 2 waste utilization centers are selected. The selected points have a very obvious focusing phenomenon. L1 and L2 small batteries realize stratified logistics. The optimal solution appears in the 133rd generation, and the minimum total cost is 359116383.81 yuan.

[0109] Tables 7-10 show that based on the current basic site selection plan, the optimal logistics network battery circulation plan has been further formulated. As shown in Table 11, compared with the original battery recycling enterprise's new energy vehicle power battery cascade utilization site selection and network optimization design plan, the total cost is reduced by 6.11%. The application of this method in the cascade utilization of power batteries can effectively reduce carbon emissions in the recycling process, reducing carbon emission costs by 13.01%.

[0110] Table 7 L1 battery quantity matrix

[0111]

[0112]

[0113] Table 8 L2 battery quantity matrix

[0114]

[0115] Table 9 Matrix of waste quantity received from recycling and reuse centers

[0116]

[0117] Table 10L3 level battery receiving quantity matrix

[0118]

[0119] Table 11 Comparison of cost results between the GA-INT algorithm optimization solution and the original solution

[0120]

[0121]

[0122] It can be seen from the above embodiments that the embodiments of the present invention adopt the above specific method to effectively reduce the operating costs of each center and improve the utilization rate of each center, which is beneficial for battery recycling companies to balance the relationship between site selection plans and network optimization plans, and provide a reference in the initial market research and implementation stage.

Claims

1. A site selection and network optimization design method for cascade utilization of new energy vehicle power batteries based on an improved genetic algorithm includes the following steps: (1) Construct a cost model for the new energy vehicle power battery recycling network and calculate the site selection cost plan in the first stage; (2) The network center location cost is obtained through the above location selection scheme. In the second stage, the minimum network logistics freight cost is taken as the objective function, and the logistics route capacity and center capacity are taken as constraints to form a mixed integer programming model; (3) An improved genetic algorithm is used to solve the two-stage mixed integer programming model and obtain the solution; (4) Based on the solution analysis, the power battery recycling site selection and network optimization design plan are obtained.

2. The method for site selection and network optimization design for cascade utilization of power batteries for new energy vehicles based on an improved genetic algorithm as claimed in claim 1, characterized in that: The site selection cost model for the cascade utilization of new energy vehicle power batteries described in step (1) includes purchase cost and fixed cost. In step (1), the first-stage objective function is: Z1=Z BC +Z FC (1) In the formula, Z BC The total cost of purchasing batteries for recycling in the first stage collection center; Z FC The total fixed cost of site selection for each center in the first phase; QB k is the total amount of used batteries collected by collection center k; u is the percentage of batteries with different quality levels u (L1 / L2 / L3); r u is the unit purchase cost of batteries with different quality levels u (L1 / L2 / L3); FC k is the fixed cost of building the collection center k; FC j is the fixed cost of building the recycling and remanufacturing center j; FC h is the fixed cost of building the waste treatment center h; x k is a 0-1 variable, collecting center k if the open decision variable is 1, otherwise it is 0; x j is a 0-1 variable. If the recycling and remanufacturing center j is open, the decision variable is 1, otherwise it is 0; h It is a 0-1 variable. If the waste treatment center h is open, the decision variable is 1, otherwise it is 0.

3. The method for site selection and network optimization design for cascade utilization of power batteries for new energy vehicles based on an improved genetic algorithm as claimed in claim 1, characterized in that: In step (2), the second-stage objective function after determining the site selection plan is: Z2=Z PC +Z TC +Z FEC +Z TEC (4) In formula (4), Z PC is the total processing cost of each center in the battery recycling network; Z TC is the total cost of battery transportation between centers in the battery recycling network; Z FEC The carbon emission costs generated by the processing of each center in the battery recycling network; Z TEC The carbon emission costs incurred during the transportation of products between centers in the battery recycling network; In formula (4), Among them, QCR kj QCM is the total amount of used batteries disassembled into small batteries by collection center k and flowing into recycling center j; kj QW is the total amount of waste batteries disassembled into small batteries by collection center k and then flowing into remanufacturing center j; jh is the total amount of waste flowing into waste treatment center h from recycling and remanufacturing center j; QCW kh is the total amount of waste batteries disassembled into small batteries by collection center k and then flowing into waste treatment center h; pcc k The processing cost of small cells per processing unit for the collection center k; pcr j The processing cost of a small battery per processing unit in a recycling center; pcm j Processing cost of small battery per unit in remanufacturing center j; pcw h The processing cost of a small battery per unit of waste treatment center h; w is the ratio of the number of small batteries that flow to the remanufacturing center after waste utilization at the recycling center to the number received by the recycling center; In formula (4), Among them, t cc is the transportation cost per battery; t cw is the transportation cost of unit waste; D kj is the distance between the collection center k and the recycling and remanufacturing center j; D kh is the distance between the collection center k and the waste treatment center h; D jh is the distance between the recycling and remanufacturing center j and the waste treatment center h; In formula (4), Among them, P ct fec is the carbon tax cost per unit of carbon emissions; k The carbon emissions generated by processing one unit of used batteries for collection center k; j The carbon emissions generated by processing one unit of small battery for recycling center j;dem j The carbon emissions generated by processing one unit of small battery for a remanufacturing center j; few h The carbon emissions generated by processing one unit of waste or small battery for a waste treatment center h; In formula (4), Among them, t e The carbon emissions generated per kilometer for transporting a unit of small battery; The entire mixed integer programming model has three main constraints: quantity, quality and capacity; The quantity constraint means that the number of batteries flowing out of the collection center is consistent with the number of batteries flowing into the recycling and remanufacturing center and the waste treatment center. The formula is as follows: bc is the number of small batteries in the unit waste battery; The quality constraint means that the quality of production waste flowing into the waste treatment center from the recycling and remanufacturing center is consistent with the quality of waste generated by the battery treatment that will flow into the recycling and remanufacturing center. The formula is as follows: w b Indicates the weight of small batteries after disassembling the waste batteries; It indicates the ratio of the amount of waste flowing out of the recycling and remanufacturing centers to the amount of waste flowing out of the disposal centers; Capacity constraint means that the processing capacity of each processing center is limited. The formula is as follows: In the formula, K k is the maximum processing capacity of the collection center k; J j The maximum processing capacity of the recycling and remanufacturing center; h is the maximum processing capacity of waste treatment center h; the decision variables mainly include the flow of each center and the site selection decision point: QB k ,QCR kj ,QCM kj ,QW jh ≥0 (14) x k ,x j ,x h ∈(0,1) (15)。 4. The method for site selection and network optimization design for cascade utilization of power batteries for new energy vehicles based on an improved genetic algorithm as claimed in claim 1, characterized in that: Step (3) specifically includes the following steps: (a) Initialize the site selection plan, initialize the number of battery requirements, total number of iterations, and population size; (b) Use the dual structure rule to initialize the encoding and decoding chromosome pairs, namely the additional code and the 0-1 variable code. The location scheme corresponding to the randomly generated chromosome needs to meet the capacity constraint, namely the capacity of the selected center must accommodate all demands, thereby generating the initial population; (c) According to the objective function of the first stage of step 2, the fixed cost and the purchase cost of the initial solution are calculated. According to the objective function of the second stage of step 2, the linear programming tool linprog function is used to optimize the network design, accurately calculate the service and transportation costs of each center, sum up the inverse of each cost as the fitness value formula, calculate the fitness of the first generation population, and perform selection, crossover, and mutation operations to form a new generation population; (d) Randomly generate a selection probability, use rule discrimination, and select individuals proportionally according to established parameters to produce offspring using a tournament or roulette method to ensure population diversity and retention of excellent genes; (e) Randomly generate a crossover probability and use the partial mapping crossover (PMX) operator to generate valid offspring; (f) Randomly generate a mutation probability, use the reverse genetic operator to generate effective offspring, randomly select two mutation points for the parent individual, rearrange the upper additional code between the two points in reverse order, and keep the order of the lower variable code unchanged; (g) Output the most optimal location solution, transportation route, number of batteries and minimum cost in the current generation population; (h) Finally, after reaching the number of iterations, the optimal result after convergence is obtained.

Images