Intelligent vehicle path planning method, device and equipment for electronic waste recovery and storage medium

Abstract

The invention discloses an intelligent vehicle path planning method, device and equipment for electronic waste recovery and a storage medium, and relates to the technical field of path planning, and the method comprises the steps: defining each recovery demand as a node demand and a side demand, and constructing a path optimization model based on each preset constraint condition and each demand; and determining a polar coordinate angle between the node required by each node and the target warehouse, sorting each node to obtain a sorting result, and performing path distribution on each node based on the sorting result by using a path optimization model to obtain a to-be-understood result, performing neighborhood search operation on the solution to be processed by using a tabu search algorithm and a simulated annealing algorithm to obtain a candidate solution set; and judging whether the to-be-processed profit of the current candidate solution is greater than the current profit maximum value, if not, determining a new current candidate solution, skipping to the step of judging the to-be-processed profit of the current candidate solution again, and setting the current candidate solution as a target solution so as to plan the vehicle path and improve the path planning efficiency.

Description

Technical Field

[0001] This invention relates to the field of route planning technology, and in particular to a method, apparatus, equipment and storage medium for intelligent vehicle route planning in electronic waste recycling. Background Technology

[0002] Currently, electronic waste (WEEE) is one of the fastest-growing solid wastes globally, and its standardized recycling is crucial for resource recycling and environmental protection. However, existing recycling models suffer from the following significant shortcomings:

[0003] Official agency on-call collection: Residents book door-to-door collection through online platforms. Although standardized, it is costly (high vehicle empty-run rate and high labor costs) and is subject to time window constraints (residents specify the door-to-door collection time), resulting in low dispatch flexibility.

[0004] Unofficial door-to-door recycling: Recyclers move along the streets to collect items. While flexible, this method is highly random and often results in empty runs due to uncertain demand. Furthermore, the load distribution is uneven (some vehicles are overloaded while others are idle), which affects efficiency and increases environmental risks (due to non-standard processing).

[0005] Therefore, traditional vehicle routing optimization algorithms are difficult to adapt to the above scenarios:

[0006] Poor initial solution quality: The greedy algorithm assigns nodes based on proximity, which can easily lead to path crossing and entanglement, increasing the difficulty of subsequent optimization;

[0007] Weak global search capability: Single tabu search or simulated annealing algorithms are prone to getting trapped in local optima, especially in scenarios with a large number of nodes (>100), where profit loss can reach more than 10%;

[0008] Insufficient constraint adaptability: It cannot handle soft time windows (early / late penalties) and capacity constraints (vehicle load limits) simultaneously, and does not consider the hybrid demand characteristics of "node + edge".

[0009] As can be seen from the above, improving the efficiency of intelligent vehicle path planning for electronic waste recycling is an urgent problem to be solved. Summary of the Invention

[0010] In view of this, the purpose of this invention is to provide a method, apparatus, device, and storage medium for intelligent vehicle path planning in electronic waste recycling, which improves the efficiency of path planning for intelligent vehicles used for electronic waste recycling. The specific solution is as follows:

[0011] In a first aspect, this application provides an intelligent vehicle path planning method for electronic waste recycling, including:

[0012] Official reservation recycling demand is defined as node demand, and unofficial street recycling demand is defined as edge demand. Then, the node demand and the edge demand are integrated to obtain the target recycling demand. A path optimization model is constructed based on preset soft time window constraints, preset vehicle capacity constraints, preset path closed loop conditions, preset profit maximization function and the target recycling demand.

[0013] The polar coordinate angle between the required nodes and the target warehouse is determined using the AD algorithm and the angle partitioning algorithm. Then, the nodes are sorted based on the polar coordinate angles to obtain the sorting results. The path optimization model is then used to allocate paths to the nodes based on the sorting results to obtain the solution to be processed. Then, the tabu search algorithm and the simulated annealing algorithm are used to perform a neighborhood search operation on the solution to be processed according to the preset neighborhood search rules to obtain the corresponding candidate solution set.

[0014] The first candidate solution in the candidate solution set is set as the current candidate solution. Then, it is determined whether the unprocessed profit of the current candidate solution is greater than the current maximum profit. If it is greater, the unprocessed profit is set as the current maximum profit. If it is not greater, the next candidate solution in the candidate solution set is set as the new current candidate solution, and the process jumps back to the step of determining whether the unprocessed profit of the current candidate solution is greater than the current maximum profit. This process continues until all candidate solutions in the candidate solution set have been determined. Then, the current candidate solution is set as the target solution, and the path of the electronic waste recycling vehicle is planned based on the target solution.

[0015] Optionally, the step of defining official reservation recycling demand as node demand and unofficial street-side recycling demand as edge demand, and then integrating the node demand and the edge demand to obtain the target recycling demand, and constructing a path optimization model based on preset soft time window constraints, preset vehicle capacity constraints, preset path closed-loop conditions, preset profit maximization function, and the target recycling demand, includes:

[0016] Official scheduled recycling needs are defined as node needs, and unofficial street-side recycling needs are defined as edge needs. Then, the node needs and edge needs are integrated to obtain the target recycling needs, which are graph-structured. The target recycling needs include nodes and edges. Each node corresponds to a node need, and each edge corresponds to an edge need.

[0017] A path optimization model is trained and constructed based on preset soft time window constraints, preset vehicle capacity constraints, preset path closure conditions, a preset profit maximization function, and the target recycling demand. The preset soft time window constraint allows vehicles used for electronic waste recycling to arrive and wait before the specified time window corresponding to the node, and incurs a penalty if the vehicle arrives after the specified time window. The preset vehicle capacity constraint ensures that the loading capacity of each vehicle does not exceed its corresponding rated load. The preset path closure condition ensures that the starting and ending points of the path for each vehicle are both the target warehouse. The preset profit maximization function is a total profit maximization function determined based on total recycling revenue, fuel costs, labor costs, and penalties. The total recycling revenue includes the total revenue obtained by each vehicle in fulfilling node and edge demands. The labor costs include driver wages and overtime pay, and the penalty is the penalty imposed when each vehicle fails to meet the preset soft time window constraints.

[0018] Optionally, the step of using the AD algorithm and the angle partitioning algorithm to determine the polar coordinate angle between each node's required node and the target warehouse, and then sorting each node based on each polar coordinate angle to obtain a sorting result, is used to apply the path optimization model and allocate paths to each node based on the sorting result to obtain the solution to be processed, including:

[0019] The target warehouse is set as the origin of the coordinate system, and the node coordinates corresponding to each node are determined based on the origin of the coordinate system. Then, the corresponding polar coordinate angle is determined by using the angle division algorithm and the AD algorithm based on the origin of the coordinate system and the coordinates of each node.

[0020] The nodes are sorted according to a preset direction and based on the polar coordinate angles to obtain the corresponding sorting results. The target service area is divided into several sector partitions using the path optimization model. Then, the nodes located in the same sector partition are assigned to the electronic waste recycling service path corresponding to the same vehicle to obtain the corresponding solution to be judged.

[0021] Determine whether the solution to be determined satisfies the preset vehicle capacity constraint after being assigned to the service path. If the solution to be determined satisfies the preset vehicle capacity constraint after being assigned to the service path, update the path load corresponding to the service path to obtain the solution to be processed.

[0022] If the solution to be judged, after being assigned to the service path, does not meet the preset vehicle capacity constraint, the service path is closed and a new service path is created to assign the solution to be judged to the new service path, thus obtaining the solution to be processed; the solution to be processed is the initial path planning scheme corresponding to each vehicle.

[0023] Optionally, the step of using the tabu search algorithm and simulated annealing algorithm and performing a neighborhood search operation on the solution to be processed according to a preset neighborhood search rule to obtain the corresponding candidate solution set includes:

[0024] The tabu search algorithm and simulated annealing algorithm are used to determine the preset neighborhood search rule corresponding to the solution to be processed. If the preset neighborhood search rule is a node swapping rule, then the two nodes in the single path of the solution to be processed are swapped, and the nodes in the multi-path of the solution to be processed are swapped to obtain the corresponding candidate solutions.

[0025] If the preset neighborhood search rule is a node movement rule, then a node movement operation is performed on the single path in the solution to be processed, and the node in the first path in the solution to be processed is moved to the specified position of the second path to obtain the corresponding candidate solution.

[0026] If the preset neighborhood search rule is the A-Opt transformation rule, then based on the value of A, the edges in the path of the solution to be processed are deleted and added to obtain the corresponding candidate solution;

[0027] If the preset neighborhood search rule is a node reversal rule, then the sequence of each child node in the path of the solution to be processed is reversed; if the preset neighborhood search rule is a multi-path merging rule, then several paths in the solution to be processed that satisfy the preset vehicle capacity constraint are merged to obtain the corresponding candidate solutions.

[0028] Construct a corresponding set of candidate solutions based on each of the aforementioned candidate solutions.

[0029] Optionally, after obtaining the corresponding candidate solution set by using the tabu search algorithm and simulated annealing algorithm to perform a neighborhood search operation on the solution to be processed according to a preset neighborhood search rule, the method further includes:

[0030] The tabu list length is determined based on the number of candidate solutions in the candidate solution set. A tabu list is constructed based on the tabu list length. Neighborhood search operations performed on the solution to be processed are monitored, and the neighborhood search operations are recorded in the tabu list. The tabu list is used to record the neighborhood search operations performed within a preset time interval and the corresponding related attribute information.

[0031] When performing a neighborhood search operation, check whether the neighborhood search operation is recorded in the taboo table. If the neighborhood search operation is not recorded in the taboo table, then record the neighborhood search operation in the taboo table.

[0032] The simulated annealing algorithm is used to determine the current temperature parameter and the current temperature decrease rate, and the solution to be processed is subjected to a neighborhood search iterative operation according to a preset neighborhood search rule; wherein, the value of the current temperature parameter corresponding to each iteration decreases sequentially according to the current temperature decrease rate; the value of the current temperature parameter is negatively correlated with the fineness of the search.

[0033] Optionally, the step of setting the first candidate solution in the candidate solution set as the current candidate solution, and then determining whether the unprocessed profit of the current candidate solution is greater than the current maximum profit, if it is greater, then setting the unprocessed profit as the current maximum profit; if it is not greater, then setting the next candidate solution in the candidate solution set as the new current candidate solution, and jumping back to the step of determining whether the unprocessed profit of the current candidate solution is greater than the current maximum profit, until all candidate solutions in the candidate solution set have been determined, includes:

[0034] The first candidate solution in the candidate solution set is set as the current candidate solution. Then, it is determined whether the unprocessed profit of the current candidate solution is greater than the current maximum profit. If the unprocessed profit of the current candidate solution is greater than the current maximum profit, the desire criterion is triggered to unban the tabu list using the desire criterion. Then, the unprocessed profit is set as the current maximum profit.

[0035] If the unprocessed profit of the current candidate solution is not greater than the current maximum profit, then the current temperature and the profit difference between the unprocessed profit and the current maximum profit are determined. Then, the Metropolis criterion is used and the acceptance probability is determined based on the current temperature and the profit difference. Based on the acceptance probability and a preset acceptance threshold, it is determined whether to accept the current candidate solution. Then, the process jumps back to the step of determining whether the unprocessed profit of the current candidate solution is greater than the current maximum profit, until all candidate solutions in the candidate solution set have been judged.

[0036] Optionally, after setting the current candidate solution as the target solution to plan the path of the electronic waste recycling vehicle based on the target solution, the method further includes:

[0037] Adaptive adjustment rules are determined based on actual application scenarios to adaptively adjust the model parameters corresponding to the path optimization model, thereby obtaining an adjusted path optimization model. These adaptive adjustment rules include adjusting the model parameters based on newly added or canceled scheduled recycling tasks, adjusting the model parameters based on real-time identified opportunities for recycling along the street, and adjusting the model parameters based on changes in travel time caused by changes in road traffic conditions.

[0038] The vehicle routes of vehicles used for electronic waste recycling are recalculated using a preset hybrid optimization algorithm and based on the adjusted path optimization model to obtain target vehicle routes, so as to dynamically and intelligently schedule each vehicle based on the target vehicle routes.

[0039] Secondly, this application provides an intelligent vehicle path planning device for electronic waste recycling, comprising:

[0040] The path optimization model construction module is used to define the official reservation recycling demand as node demand and the unofficial street recycling demand as edge demand. Then, the node demand and the edge demand are integrated to obtain the target recycling demand. The path optimization model is constructed based on preset soft time window constraints, preset vehicle capacity constraints, preset path closed loop conditions, preset profit maximization function and the target recycling demand.

[0041] The candidate solution set determination module is used to determine the polar coordinate angle between the node requiring each node and the target warehouse using the AD algorithm and the angle partitioning algorithm. Then, the nodes are sorted based on the polar coordinate angle to obtain the sorting result. The path optimization model is used to allocate paths to each node based on the sorting result to obtain the solution to be processed. Then, the tabu search algorithm and the simulated annealing algorithm are used to perform a neighborhood search operation on the solution to be processed according to the preset neighborhood search rules to obtain the corresponding candidate solution set.

[0042] The path planning module is used to set the first candidate solution in the candidate solution set as the current candidate solution, and then determine whether the unprocessed profit of the current candidate solution is greater than the current maximum profit. If it is greater, the unprocessed profit is set as the current maximum profit. If it is not greater, the next candidate solution in the candidate solution set is set as the new current candidate solution, and the process jumps back to the step of determining whether the unprocessed profit of the current candidate solution is greater than the current maximum profit, until all candidate solutions in the candidate solution set have been determined. Then, the current candidate solution is set as the target solution, so as to plan the path of the electronic waste recycling vehicle based on the target solution.

[0043] Thirdly, this application provides an electronic device, comprising:

[0044] Memory, used to store computer programs;

[0045] A processor is used to execute the computer program to implement the aforementioned intelligent vehicle path planning method for electronic waste recycling.

[0046] Fourthly, this application provides a computer-readable storage medium for storing a computer program, wherein the computer program, when executed by a processor, implements the aforementioned intelligent vehicle path planning method for electronic waste recycling.

[0047] As can be seen from the above, before performing intelligent vehicle route planning for electronic waste recycling, this application needs to define official scheduled recycling demand as node demand and unofficial street-side recycling demand as edge demand. Then, it integrates node and edge demands to obtain the target recycling demand. A route optimization model is then constructed based on preset soft time window constraints, preset vehicle capacity constraints, preset path closure conditions, preset profit maximization functions, and the target recycling demand. The AD algorithm and angle partitioning algorithm are used to determine the polar coordinate angles between each node's demand and the target warehouse. Then, each node is sorted based on its polar coordinate angle to obtain the sorting result. Finally, the route optimization model is used, and based on the sorting result, paths are allocated to each node to obtain the waste disposal results. The process involves understanding the concepts, then using the tabu search algorithm and simulated annealing algorithm, and performing a neighborhood search operation on the solution to be processed according to the preset neighborhood search rules to obtain the corresponding candidate solution set. The first candidate solution in the candidate solution set is set as the current candidate solution. Then, it is determined whether the unprocessed profit of the current candidate solution is greater than the current maximum profit. If it is greater, the unprocessed profit is set as the current maximum profit. If it is not greater, the next candidate solution in the candidate solution set is set as the new current candidate solution, and the process jumps back to the step of determining whether the unprocessed profit of the current candidate solution is greater than the current maximum profit. This process continues until all candidate solutions in the candidate solution set have been determined. Finally, the current candidate solution is set as the target solution, and the path of the electronic waste recycling vehicle is planned based on the target solution.

[0048] Therefore, this application first needs to define official appointment recycling demand as node demand and unofficial street-side recycling demand as edge demand. Then, it integrates node demand and edge demand to obtain target recycling demand. Based on preset soft time window constraints, preset vehicle capacity constraints, preset path closure conditions, preset profit maximization function, and target recycling demand, a path optimization model is constructed. Second, the AD algorithm and angle partitioning algorithm are used to determine the polar coordinate angle between each node demand node and the target warehouse. Then, each node is sorted based on each polar coordinate angle to obtain the sorting result. The path optimization model is then used to allocate paths to each node based on the sorting result to obtain the solution to be processed. Finally, tabu search is used. The algorithm combines the search algorithm and simulated annealing algorithm to perform a neighborhood search operation on the solution to be processed according to a preset neighborhood search rule, obtaining the corresponding candidate solution set. Then, the first candidate solution in the candidate solution set is set as the current candidate solution. Next, it is determined whether the unprocessed profit of the current candidate solution is greater than the current maximum profit. If it is, the unprocessed profit is set as the current maximum profit; otherwise, the next candidate solution in the candidate solution set is set as the new current candidate solution, and the process returns to the step of determining whether the unprocessed profit of the current candidate solution is greater than the current maximum profit. This process continues until all candidate solutions in the candidate solution set have been evaluated. Finally, the current candidate solution is set as the target solution, and the path of the e-waste recycling vehicle is planned based on the target solution. This improves the efficiency of path planning for intelligent vehicles used for e-waste recycling, thereby enhancing the user experience. Attached Figure Description

[0049] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0050] Figure 1 This application discloses a flowchart of a method for intelligent vehicle path planning in electronic waste recycling.

[0051] Figure 2 This diagram illustrates an initial solution generated by a greedy algorithm and an initial solution generated by an Algebraic Divergence (AD) algorithm. Figure 2 (a) is a schematic diagram of the initial solution generated using a greedy algorithm. Figure 2 (b) is a schematic diagram of the initial solution generated using the AD algorithm;

[0052] Figure 3 This application discloses a flowchart of a specific intelligent vehicle path planning process for electronic waste recycling.

[0053] Figure 4 This is a schematic diagram of a specific node exchange disclosed in this application;

[0054] Figure 5 This is a schematic diagram illustrating a specific node movement disclosed in this application, wherein, Figure 5 (a) is a schematic diagram of a single node moving on a single path; Figure 5 (b) is a schematic diagram of multi-node movement on a single path; Figure 5 (c) is a schematic diagram of single-node movement on a multi-path path. Figure 5 (d) is a schematic diagram of multi-node movement on a multi-path;

[0055] Figure 6 This is a schematic diagram of a specific A-Opt transform disclosed in this application; wherein, Figure 6 (a) is a schematic diagram of the cross edge generation of the A-Opt transform when A=2; Figure 6 (b) is a schematic diagram of the independent edge generation of the A-Opt transformation when A=2; Figure 6 (c) is a schematic diagram of the A-Opt transform when A=3;

[0056] Schematic diagram;

[0057] Figure 7 This is a specific node reversal diagram disclosed in this application;

[0058] Figure 8 This is a schematic diagram illustrating a specific multipath merging method disclosed in this application;

[0059] Figure 9 This is a schematic diagram of the intelligent vehicle path planning process device for electronic waste recycling disclosed in this application;

[0060] Figure 10 This is a structural diagram of an electronic device disclosed in this application. Detailed Implementation

[0061] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0062] Currently, e-waste is one of the fastest-growing solid wastes globally, and its standardized recycling is crucial for resource recycling and environmental protection. Official agency-arranged collection is costly (high vehicle empty-running rates and high labor costs) and subject to time window constraints, resulting in low scheduling flexibility. Unofficial street-side collection often results in empty runs due to uncertain demand and uneven load distribution (some vehicles overloaded, some idle), impacting efficiency and increasing environmental risks (non-standard handling). Therefore, traditional vehicle routing optimization algorithms are ill-suited to these scenarios. To address this, this application provides an intelligent vehicle routing process method for e-waste recycling, which improves the efficiency of routing planning for intelligent vehicles used for e-waste recycling.

[0063] See Figure 1 As shown in the figure, an embodiment of the present invention discloses a method for intelligent vehicle path planning in electronic waste recycling, including:

[0064] Step S11: Define the official reservation recycling demand as node demand and the unofficial street recycling demand as edge demand. Then integrate the node demand and the edge demand to obtain the target recycling demand. Construct a path optimization model based on preset soft time window constraints, preset vehicle capacity constraints, preset path closed loop conditions, preset profit maximization function and the target recycling demand.

[0065] In this embodiment, the initial solution generated by the greedy algorithm and the initial solution generated by the AD algorithm are as follows: Figure 2 As shown, where, Figure 2 (a) is the initial solution generated using a greedy algorithm. Figure 2 (b) is the initial solution generated using the AD algorithm. This initial solution consists of several paths, each representing the trajectory of a vehicle. Blue dots represent customer points with collection needs, solid lines represent demand edges (collection needs encountered by the vehicle while serving demand nodes), and gray dots and dashed edges do not have collection needs; they are only used as markers on the paths to calculate path lengths. The problem requires that the needs at each demand node must be satisfied, while the needs on the edges are optional.

[0066] It is worth mentioning that, Figure 3This is a schematic diagram of the algorithm flow proposed in this application, used to intuitively illustrate how the algorithm iteratively optimizes from the initial solution to the optimal solution: First, a hybrid vehicle routing problem (CGRPTW) model integrating official on-call and unofficial door-to-door recycling needs is established. This model includes soft time window constraints (allowing early / late arrivals but incurring penalties) and vehicle capacity constraints (load not exceeding the maximum volume Q). The objective function is to maximize total profit (recycling revenue minus fuel costs, labor costs, and penalties). Then, the TSSA-TS algorithm is used to solve the problem: the first stage constructs an initial solution using an angle partitioning algorithm; the second stage integrates tabu search and simulated annealing algorithms, and optimizes the path through five neighborhood search operations. Finally, an integrated electronic waste recycling path optimization scheme is obtained.

[0067] In this embodiment, the CGRPTW model needs to be constructed first: First, demand integration is performed: officially scheduled recycling is defined as "node demand" (mandatory service, such as residential addresses), and unofficial street-side recycling is defined as "edge demand" (optional service, such as street-side recycling along a certain road segment). Then, constraints are generated: soft time window: vehicles must provide service within the node demand period; early arrival requires waiting, and late arrival incurs a penalty; capacity constraint: each vehicle has a fixed load; closed-loop path: all vehicles depart from and return from the warehouse, forming a 0-ij-0 closed-loop path.

[0068] Furthermore, the expression for the objective function is as follows:

[0069] ;

[0070] in, Profits for door-to-door service units For vehicles, This indicates the vehicle assembly at the dispatch center. , As an edge, To represent the nodes that have a demand, i.e., the collection requests received by the online platform, , For distance, It is a binary number; if the value is 1, it indicates a car. Served the elements Otherwise, it is 0. It is a directed edge set. All edges in the array are assigned weights, using Indicates distance, ; Fuel cost per unit time For driving side The time required For vehicles On the side Number of times running in the air For the driver's daily wage, This refers to the number of vehicles in an informal convoy. The cost of overtime work for drivers per unit of time. For vehicles Overtime hours, for , represents a set of nodes. Represents a warehouse, and its value is... , This represents the set of customer points (demand nodes) and non-demand nodes. ; For demand points, To reach the demand point The waiting time To reach the demand point The delay time.

[0071] Specifically, official appointment-based recycling demand is defined as node demand, and unofficial street-side recycling demand is defined as edge demand. Then, node and edge demands are integrated to obtain the target recycling demand. A path optimization model is constructed based on preset soft time window constraints, preset vehicle capacity constraints, preset path closure conditions, preset profit maximization functions, and the target recycling demand. This can include: defining official appointment-based recycling demand as node demand and unofficial street-side recycling demand as edge demand; then integrating node and edge demands to obtain the target recycling demand with a graph structure; the target recycling demand includes nodes and edges; there is a one-to-one correspondence between node and node demands; there is a one-to-one correspondence between edge and edge demands; based on preset soft time window constraints, preset vehicle capacity constraints, preset path closure conditions, and preset profit maximization functions, a path optimization model is constructed. A path optimization model is trained and constructed based on the maximization function and the target recycling requirements. The model includes: a preset soft time window constraint allowing vehicles for e-waste recycling to arrive and wait before the specified time window corresponding to a node, with a penalty for arriving after the specified time window; a preset vehicle capacity constraint ensuring that each vehicle's load does not exceed its rated capacity; a preset path closure condition ensuring that both the starting and ending points of each vehicle's path are the target warehouse; and a preset profit maximization function determined based on total recycling revenue, fuel costs, labor costs, and penalties. Total recycling revenue includes the total revenue earned by each vehicle for fulfilling node and edge requirements; labor costs include driver wages and overtime pay; and penalties are the penalties imposed when a vehicle fails to meet the preset soft time window constraints.

[0072] Step S12: Determine the polar coordinate angle between the node requiring each node and the target warehouse using the AD algorithm and the angle partitioning algorithm. Then, sort the nodes based on the polar coordinate angles to obtain the sorting results. Use the path optimization model and the sorting results to allocate paths to the nodes to obtain the solution to be processed. Then, use the tabu search algorithm and the simulated annealing algorithm and the preset neighborhood search rules to perform a neighborhood search operation on the solution to be processed to obtain the corresponding candidate solution set.

[0073] In this embodiment, the angle partitioning algorithm includes the following steps: First, taking the warehouse as the origin, calculate the polar angle (the angle between the polar angle and the positive coordinate axis) of all demand nodes, and sort them clockwise to obtain the sorting result. Second, initialize empty paths, and scan the nodes according to the sorting result. Then, determine whether the node meets the capacity constraint (total load ≤ Q) after being added to the current path. If it does, add the node to the current path and update the path load. If it does not meet the constraint, close the current path, create a new path and add the node. Repeat the above steps until all demand nodes are allocated, and generate the initial path set.

[0074] In one specific implementation, the angle partitioning algorithm is based on the polar angle partitioning concept. The warehouse is considered the origin, and the polar angles (such as the angle with due east) of all demand nodes are calculated and sorted clockwise. For example, nodes with polar angles between 0° and 60° are assigned to path 1, nodes between 60° and 120° are assigned to path 2, and so on. This method reduces cross-regional travel, and the profit corresponding to the initial solution is on average 5%-10% higher than that of the greedy algorithm.

[0075] Specifically, the polar coordinate angles between each node's required nodes and the target warehouse are determined using the AD algorithm and angle partitioning algorithm. Then, the nodes are sorted based on these polar coordinate angles to obtain a sorting result. A path optimization model is then used to allocate paths to each node based on the sorting result, yielding the solution to be processed. This process may include: setting the target warehouse as the origin, determining the node coordinates based on the origin, and then using the angle partitioning algorithm and the AD algorithm to determine the corresponding polar coordinate angles based on the origin and the node coordinates; sorting the nodes according to a preset direction based on these polar coordinate angles to obtain the corresponding sorting result; and then using the path optimization model to allocate paths to the target service area. The domain is divided into several sector partitions. Nodes within the same sector partition are then assigned to the electronic waste recycling service path corresponding to the same vehicle, resulting in a solution to be determined. It is then determined whether the solution meets the preset vehicle capacity constraints after being assigned to the service path. If the solution meets the preset vehicle capacity constraints, the path load corresponding to the service path is updated, resulting in a solution to be processed. If the solution does not meet the preset vehicle capacity constraints, the service path is closed, and a new service path is created to assign the solution to the new service path, resulting in a solution to be processed. The solution to be processed is the initial path planning scheme for each vehicle.

[0076] In this embodiment, the five neighborhood search operations are: node swapping, node movement, A-Opt transformation, node reversal, and multi-path merging. A schematic diagram of node swapping is shown below. Figure 4 As shown: Among them, Figure 4 (a) is a schematic diagram of node swapping operations on a single path, that is, the two nodes in a single path swap positions (such as swapping node i and node j in path A). Figure 4 (a) is a schematic diagram of node swapping operations between multiple paths (e.g., swapping node i in path A with node j in path B); a schematic diagram of node movement is shown below. Figure 5 As shown: Among them, Figure 5 (a) is a schematic diagram of a single node moving on a single path, such as node i moving between node j and node k within a single path; Figure 5 (b) is a schematic diagram of multi-node movement on a single path; Figure 5 (c) is a schematic diagram of a single node movement on multiple paths, such as node i moving from path A to a specified position on path B between multiple paths; Figure 5 (d) is a schematic diagram of multi-node movement on a multipath. A schematic diagram of the A-Opt transform is shown below. Figure 6 As shown: A schematic diagram of the A-Opt transform corresponding to A=2 is shown below. Figure 6 (a) and Figure 6 As shown in (b), Figure 6 (a) is a schematic diagram of the generation of intersecting edges. Figure 6(b) is a schematic diagram of independent edge generation, where black dashed arrows represent edges to be deleted, red solid arrows represent newly added edges, and black solid arrows represent existing edges. Two edges (ij, kl) are deleted and two edges (ik, jl) are added. The schematic diagram of the A-Opt transformation when A=3 is shown below. Figure 6 As shown in (c): Delete three edges and add three edges to reconstruct the path; a schematic diagram of node reversal is shown below. Figure 7 As shown: the reverse order of consecutive child node sequences (such as ijkl) in a path is lkji; a schematic diagram of multi-path merging is shown below. Figure 8 As shown: Two or more paths that satisfy capacity constraints (such as path A: 0-ij-0 and path B: 0-kl-0) are merged into one path (such as 0-ijkl-0).

[0077] Subsequently, the embodiments of this application generate candidate solutions through the above five operations. For example, performing node swapping (swapping 1 and 3) on the path 0-1-2-3-0 yields 0-3-2-1-0, or performing 2-Opt transformation (deleting 1-2 and 3-0, and adding 1-3 and 2-0) yields 0-1-3-2-0.

[0078] Specifically, the tabu search algorithm and simulated annealing algorithm are used to perform a neighborhood search operation on the solution to be processed according to a preset neighborhood search rule to obtain the corresponding candidate solution set. This can include: using the tabu search algorithm and simulated annealing algorithm to determine the preset neighborhood search rule corresponding to the solution to be processed; if the preset neighborhood search rule is a node swapping rule, then the positions of two nodes in the single path of the solution to be processed are swapped, and the node swapping operation is performed on the multi-path of the solution to obtain the corresponding candidate solution; if the preset neighborhood search rule is a node movement rule, then the node movement operation is performed on the single path of the solution to be processed, and the candidate solution is moved to the next node. The nodes in the first path of the solution are moved to the specified positions in the second path to obtain the corresponding candidate solutions. If the preset neighborhood search rule is the A-Opt transformation rule, the edges in the path of the solution to be processed are deleted and added based on the value of A to obtain the corresponding candidate solutions. If the preset neighborhood search rule is the node reversal rule, the sequence of child nodes in the path of the solution to be processed is reversed. If the preset neighborhood search rule is the multi-path merging rule, several paths in the solution to be processed that satisfy the preset vehicle capacity constraint are merged to obtain the corresponding candidate solutions. A corresponding candidate solution set is constructed based on each candidate solution.

[0079] It is worth mentioning that, in this embodiment of the application, completed neighborhood operations (such as swapping nodes i and j) can be stored in a tabu list, and a tabu duration (TABU_RANGE) can be set to avoid repeated searches in a short period of time. That is, if the operation of swapping nodes 1 and 3 is already in the tabu list, and the profit of the new solution does not exceed the current optimal solution, then the solution is skipped; if the profit of the new solution is higher, then the tabu is lifted and the optimal solution is updated.

[0080] Specifically, after obtaining the corresponding candidate solution set by performing a neighborhood search operation on the solution to be processed using the tabu search algorithm and simulated annealing algorithm according to the preset neighborhood search rules, the process may further include: determining the tabu list length based on the number of candidate solutions in the candidate solution set, constructing the tabu list based on the tabu list length, monitoring the neighborhood search operations performed on the solution to be processed, and recording the neighborhood search operations in the tabu list; the tabu list is used to record the neighborhood search operations performed within a preset time interval and the corresponding related attribute information; when a neighborhood search operation is performed, it is checked whether the neighborhood search operation is recorded in the tabu list. If the neighborhood search operation is not recorded in the tabu list, it is recorded in the tabu list; the simulated annealing algorithm is used to determine the current temperature parameter and the current temperature decrease rate, and a neighborhood search iterative operation is performed on the solution to be processed according to the preset neighborhood search rules; wherein, the value of the current temperature parameter corresponding to each iteration decreases sequentially according to the current temperature decrease rate; the value of the current temperature parameter is negatively correlated with the fineness of the search.

[0081] Step S13: Set the first candidate solution in the candidate solution set as the current candidate solution, and then determine whether the unprocessed profit of the current candidate solution is greater than the current maximum profit. If it is greater, set the unprocessed profit as the current maximum profit. If it is not greater, set the next candidate solution in the candidate solution set as the new current candidate solution, and jump back to the step of determining whether the unprocessed profit of the current candidate solution is greater than the current maximum profit, until all candidate solutions in the candidate solution set have been determined. Then set the current candidate solution as the target solution, and plan the path of the electronic waste recycling vehicle based on the target solution.

[0082] In this embodiment, if the profit of a candidate solution is higher than that of the current optimal solution, the desire criterion is triggered to lift the restrictions and update the optimal solution; otherwise, the acceptance probability is calculated according to the Metropolis criterion. :

[0083] ;

[0084] in, For profit difference, Given the current temperature, when The solution is accepted if the random number is greater than or equal to (0,1).

[0085] It is worth mentioning that after each iteration, according to ( (To reduce the cooling rate) the temperature until the final temperature is reached. Output the optimal solution, and the test dataset is shown in Table 1:

[0086] Table 1 shows the dataset for this test.

[0087]

[0088] In this embodiment, the comparison information of the results of constructing the initial solution using different algorithms is shown in Table 2:

[0089] Table 2. Comparison of Initial Solution Construction Results for Different Algorithms

[0090]

[0091] Furthermore, the results of using different neighborhood selection strategies are compared in Table 3:

[0092] Table 3. Comparison of results from different neighborhood selection strategies

[0093]

[0094] It is worth mentioning that when the temperature is high (e.g., T=100), the probability of accepting a poor solution in this application embodiment is high, which is conducive to escaping local optima; when the temperature decreases (e.g., T=10), the embodiment of this application only accepts slightly poor solutions and focuses on local optimization.

[0095] Specifically, the first candidate solution in the candidate solution set is set as the current candidate solution. Then, it is determined whether the unprocessed profit of the current candidate solution is greater than the current maximum profit. If it is, the unprocessed profit is set as the current maximum profit. If it is not, the next candidate solution in the candidate solution set is set as the new current candidate solution, and the process jumps back to the step of determining whether the unprocessed profit of the current candidate solution is greater than the current maximum profit. This process continues until all candidate solutions in the candidate solution set have been determined. This can include: setting the first candidate solution in the candidate solution set as the current candidate solution, then determining whether the unprocessed profit of the current candidate solution is greater than the current maximum profit. If the unprocessed profit of the current candidate solution is greater than the current maximum profit, the desire criterion is triggered to unbind the tabu list, and then the unprocessed profit is set as the current maximum profit. If the unprocessed profit of the current candidate solution is not greater than the current maximum profit, the current temperature and the profit difference between the unprocessed profit and the current maximum profit are determined, and then Metropolis is used to determine the profit difference. The criteria determine the acceptance probability based on the difference between the current temperature and the profit. Based on the acceptance probability and the preset acceptance threshold, it determines whether to accept the current candidate solution and then jumps back to the step of judging whether the unprocessed profit of the current candidate solution is greater than the current maximum profit, until all candidate solutions in the candidate solution set have been judged.

[0096] In this embodiment, after obtaining the target solution and planning the route of the electronic waste recycling vehicle using the target solution, the present application embodiment can use an optimization algorithm to adjust the adjusted route optimization model. Specifically, after setting the current candidate solution as the target solution and planning the route of the electronic waste recycling vehicle based on the target solution, it may further include: determining adaptive adjustment rules based on the actual application scenario, and using the adaptive adjustment rules to adaptively adjust the model parameters corresponding to the route optimization model to obtain the adjusted route optimization model; wherein, the adaptive adjustment rules include adjusting the model parameters corresponding to the route optimization model based on newly added or canceled scheduled recycling tasks, adjusting the model parameters corresponding to the route optimization model based on real-time identified street recycling demand opportunities, and adjusting the model parameters corresponding to the route optimization model based on changes in travel time caused by changes in road traffic conditions; using a preset hybrid optimization algorithm and based on the adjusted route optimization model, the vehicle route of the vehicle used for electronic waste recycling is re-optimized to obtain the target vehicle route, and the vehicles are dynamically and intelligently scheduled based on the target vehicle route.

[0097] In one specific implementation, the vehicle capacity Q = 15 (capable of recycling 15 small electronic devices), the driver's daily wage W = 500 yuan, overtime pay O = 1.5 yuan / minute, fuel cost C = 0.5 yuan / minute, and time window penalty V = 5 yuan / minute; the recycling unit price: official node recycling P node = 100 yuan / unit, unofficial edge recycling P edge = 400 yuan / group (continuous recycling along the street); time window example: node demand ei = 9:00, li = 10:30 (allowing waiting 30 minutes in advance, with a penalty for being late within 30 minutes). Furthermore, the initial temperature T0 = 100, the cooling rate a = 0.95, the maximum number of iterations L = 100, and the tabu list length TABU_RANGE = 20.

[0098] Furthermore, in this embodiment, 120 demand nodes (30 official reservations and 90 potential street-side demands) in a certain area of ​​Shanghai can be input. Then, after calculating the node polar angles, nodes with angles between 0° and 60° are assigned to path 1, nodes with angles between 60° and 120° are assigned to path 2, and the remaining nodes are assigned to path 3. Subsequently, node swapping is performed on path 1 (swapping nodes 15 and 28), increasing profit by 200 yuan. A 2-Opt transformation is performed on path 2, shortening the driving distance by 1.2 kilometers and reducing fuel costs by 6 yuan. The final output is: a total profit of 17954.6 yuan for the three optimized paths, a 12.4% improvement over the traditional algorithm, with a time window violation of less than 5 minutes.

[0099] Furthermore, this application embodiment uses 24 standard instances (including four sizes from Very-small to Large) generated based on preset regional road network data, covering low-density (few demand edges) and high-density (many demand edges) scenarios. Table 4 shows the comparative information on profit determination based on different instances using the TSSA-TS algorithm.

[0100] Table 4. Comparison of Profit Determination Using the TSSA-TS Algorithm Based on Different Instances

[0101]

[0102] Understandably, TSSA-TS achieved a profit of 4484.36 yuan in the V15High instance, a 0.19% increase compared to TS (4476.05 yuan); and a profit of 17954.6 yuan in the L120High instance, a 12.4% increase compared to TS (16694.8 yuan). Furthermore, the average runtime per instance was 1.2 seconds (compared to 1.0 seconds for TS), a 20% increase, but the profit improvement was significant. Moreover, in 100 repeated experiments, the standard deviation of the solution was < 5%, demonstrating better stability than the SA algorithm.

[0103] As can be seen from the above, the embodiments of this application first need to define the official reservation recycling demand as node demand and the unofficial street recycling demand as edge demand. Then, the node demand and edge demand are integrated to obtain the target recycling demand. A path optimization model is constructed based on preset soft time window constraints, preset vehicle capacity constraints, preset path closure conditions, preset profit maximization function, and target recycling demand. Second, the polar coordinate angle between each node demand node and the target warehouse is determined using the AD algorithm and angle partitioning algorithm. Then, each node is sorted based on each polar coordinate angle to obtain the sorting result. The path optimization model is used and the path is allocated to each node based on the sorting result to obtain the solution to be processed. Then, tabu is used. The search algorithm and simulated annealing algorithm perform a neighborhood search operation on the solution to be processed according to a preset neighborhood search rule to obtain the corresponding candidate solution set. Then, the first candidate solution in the candidate solution set is set as the current candidate solution. Next, it is determined whether the unprocessed profit of the current candidate solution is greater than the current maximum profit. If it is, the unprocessed profit is set as the current maximum profit; otherwise, the next candidate solution in the candidate solution set is set as the new current candidate solution, and the process returns to the step of determining whether the unprocessed profit of the current candidate solution is greater than the current maximum profit. This process continues until all candidate solutions in the candidate solution set have been determined. Finally, the current candidate solution is set as the target solution, and the path of the e-waste recycling vehicle is planned based on the target solution. This improves the efficiency of path planning for intelligent vehicles used for e-waste recycling, thereby enhancing the user experience.

[0104] Accordingly, see Figure 9 As shown, this application also provides an intelligent vehicle route planning device for electronic waste recycling, comprising:

[0105] The path optimization model construction module 11 is used to define the official reservation recycling demand as node demand and the unofficial street recycling demand as edge demand. Then, the node demand and the edge demand are integrated to obtain the target recycling demand. The path optimization model is constructed based on the preset soft time window constraints, preset vehicle capacity constraints, preset path closed loop conditions, preset profit maximization function and the target recycling demand.

[0106] The candidate solution set determination module 12 is used to determine the polar coordinate angle between the node required by each node and the target warehouse using the AD algorithm and the angle partitioning algorithm. Then, it sorts the nodes based on the polar coordinate angle to obtain the sorting result. The path optimization model is used to allocate paths to each node based on the sorting result to obtain the solution to be processed. Then, the tabu search algorithm and the simulated annealing algorithm are used to perform a neighborhood search operation on the solution to be processed according to the preset neighborhood search rules to obtain the corresponding candidate solution set.

[0107] The path planning module 13 is used to set the first candidate solution in the candidate solution set as the current candidate solution, and then determine whether the unprocessed profit of the current candidate solution is greater than the current maximum profit. If it is greater, the unprocessed profit is set as the current maximum profit. If it is not greater, the next candidate solution in the candidate solution set is set as the new current candidate solution, and the process jumps back to the step of determining whether the unprocessed profit of the current candidate solution is greater than the current maximum profit, until all candidate solutions in the candidate solution set have been determined. Then, the current candidate solution is set as the target solution, so as to plan the path of the electronic waste recycling vehicle based on the target solution.

[0108] In some specific embodiments, the path optimization model construction module 11 may specifically include:

[0109] The target recycling demand determination unit is used to define official reservation recycling demand as node demand and unofficial street recycling demand as edge demand, and then integrate the node demand and the edge demand to obtain the target recycling demand with a graph structure; the target recycling demand includes nodes and edges; the nodes correspond one-to-one with the node demand; the edges correspond one-to-one with the edge demand;

[0110] A path optimization model construction subunit is used to train and construct a path optimization model based on preset soft time window constraints, preset vehicle capacity constraints, preset path closure conditions, a preset profit maximization function, and the target recycling demand. The preset soft time window constraint allows vehicles used for electronic waste recycling to arrive and wait before the specified time window corresponding to the node, and incurs a penalty if the vehicle arrives after the specified time window. The preset vehicle capacity constraint ensures that the loading capacity of each vehicle does not exceed its corresponding rated load. The preset path closure condition ensures that the starting and ending points of the path for each vehicle are both the target warehouse. The preset profit maximization function is a total profit maximization function determined based on total recycling revenue, fuel costs, labor costs, and penalties. The total recycling revenue includes the total revenue obtained by each vehicle in fulfilling node and edge demands. The labor costs include driver wages and overtime pay, and the penalty is the penalty imposed when each vehicle fails to meet the preset soft time window constraints.

[0111] In some specific embodiments, the candidate solution set determination module 12 may specifically include:

[0112] The polar coordinate angle determination unit is used to set the target warehouse as the coordinate origin, determine the node coordinates corresponding to each node based on the coordinate origin, and then use the angle division algorithm and the AD algorithm to determine the corresponding polar coordinate angle based on the coordinate origin and the coordinates of each node.

[0113] The sorting result determination unit is used to sort each node according to a preset direction and based on each polar coordinate angle to obtain the corresponding sorting result, and to divide the target service area into several sector partitions using the path optimization model, and then to assign each node located in the same sector partition to the electronic waste recycling service path corresponding to the same vehicle to obtain the corresponding solution to be judged.

[0114] The first solution determination unit is used to determine whether the solution to be determined satisfies the preset vehicle capacity constraint after being assigned to the service path. If the solution to be determined satisfies the preset vehicle capacity constraint after being assigned to the service path, the path load corresponding to the service path is updated to obtain the solution to be determined.

[0115] The second solution determination unit is used to close the service path and create a new service path if the solution to be judged does not meet the preset vehicle capacity constraint after being assigned to the service path, so as to assign the solution to be judged to the new service path and obtain the solution to be processed; the solution to be processed is the initial path planning scheme corresponding to each vehicle.

[0116] In some specific embodiments, the candidate solution set determination module 12 may specifically include:

[0117] The first candidate solution determination unit is used to determine the preset neighborhood search rule corresponding to the solution to be processed using the tabu search algorithm and the simulated annealing algorithm. If the preset neighborhood search rule is a node swapping rule, then the two nodes in the single path of the solution to be processed are swapped, and the nodes in the multi-path of the solution to be processed are swapped to obtain the corresponding candidate solution.

[0118] If the preset neighborhood search rule is a node movement rule, then the node movement operation is performed on the single path in the solution to be processed, and the node in the first path in the solution to be processed is moved to the specified position of the second path to obtain the corresponding candidate solution.

[0119] The third candidate solution determination unit is used to delete and add edges in the path of the solution to be processed based on the value of A if the preset neighborhood search rule is the A-Opt transformation rule, so as to obtain the corresponding candidate solution.

[0120] The fourth candidate solution determination unit is used to reverse the sequence of each child node in the path of the solution to be processed if the preset neighborhood search rule is a node reversal rule; and to merge several paths in the solution to be processed that satisfy the preset vehicle capacity constraint condition to obtain the corresponding candidate solution if the preset neighborhood search rule is a multi-path merging rule.

[0121] The candidate solution set determination subunit is used to construct a corresponding candidate solution set based on each of the candidate solutions.

[0122] In some specific embodiments, the intelligent vehicle route planning device for electronic waste recycling may further include:

[0123] The tabu list construction unit is used to determine the tabu list length based on the number of candidate solutions in the candidate solution set, construct the tabu list based on the tabu list length, monitor the neighborhood search operation performed on the solution to be processed, and record the neighborhood search operation in the tabu list; the tabu list is used to record the neighborhood search operation performed within a preset time interval and the corresponding related attribute information;

[0124] The taboo list checking unit is used to check whether the neighborhood search operation is recorded in the taboo list when the neighborhood search operation is performed. If the neighborhood search operation is not recorded in the taboo list, the neighborhood search operation is recorded in the taboo list.

[0125] The temperature parameter determination unit is used to determine the current temperature parameter and the current temperature decrease rate using a simulated annealing algorithm, and to perform a neighborhood search iterative operation on the solution to be processed according to a preset neighborhood search rule; wherein, the value of the current temperature parameter corresponding to each iteration decreases sequentially according to the current temperature decrease rate; the value of the current temperature parameter is negatively correlated with the fineness of the search.

[0126] In some specific embodiments, the path planning module 13 may specifically include:

[0127] The pending profit judgment unit is used to set the first candidate solution in the candidate solution set as the current candidate solution, and then determine whether the pending profit of the current candidate solution is greater than the current maximum profit. If the pending profit of the current candidate solution is greater than the current maximum profit, the desire criterion is triggered to unban the tabu list using the desire criterion, and then the pending profit is set as the current maximum profit.

[0128] The step jump unit is used to determine the current temperature and the profit difference between the unprocessed profit and the current maximum profit if the unprocessed profit of the current candidate solution is not greater than the current maximum profit. Then, it uses the Metropolis criterion and the profit difference to determine the acceptance probability. Based on the acceptance probability and a preset acceptance threshold, it determines whether to accept the current candidate solution and jumps back to the step of determining whether the unprocessed profit of the current candidate solution is greater than the current maximum profit until all candidate solutions in the candidate solution set have been judged.

[0129] In some specific embodiments, the intelligent vehicle route planning device for electronic waste recycling may further include:

[0130] The model parameter adjustment unit is used to determine adaptive adjustment rules based on actual application scenarios, and to adaptively adjust the model parameters corresponding to the path optimization model using the adaptive adjustment rules to obtain the adjusted path optimization model; wherein, the adaptive adjustment rules include adjusting the model parameters corresponding to the path optimization model based on newly added or canceled scheduled recycling tasks, adjusting the model parameters corresponding to the path optimization model based on real-time identified opportunities for recycling demand along the street, and adjusting the model parameters corresponding to the path optimization model based on changes in travel time caused by changes in road traffic conditions;

[0131] The target vehicle path determination unit is used to re-optimize the vehicle path of the vehicle used for electronic waste recycling by using a preset hybrid optimization algorithm and based on the adjusted path optimization model to obtain the target vehicle path, so as to dynamically and intelligently schedule each vehicle based on the target vehicle path.

[0132] Furthermore, embodiments of this application also disclose an electronic device, Figure 10 This is a structural diagram of an electronic device 20 according to an exemplary embodiment. The content of the diagram should not be construed as limiting the scope of this application. The electronic device 20 may specifically include: at least one processor 21, at least one memory 22, a power supply 23, a communication interface 24, an input / output interface 25, and a communication bus 26. The memory 22 stores a computer program, which is loaded and executed by the processor 21 to implement the relevant steps in the intelligent vehicle route planning method for electronic waste recycling disclosed in any of the foregoing embodiments. Furthermore, the electronic device 20 in this embodiment may specifically be an electronic computer.

[0133] In this embodiment, the power supply 23 is used to provide operating voltage for each hardware device on the electronic device 20; the communication interface 24 can create a data transmission channel between the electronic device 20 and external devices, and the communication protocol it follows can be any communication protocol applicable to the technical solution of this application, and is not specifically limited here; the input / output interface 25 is used to acquire external input data or output data to the outside world, and its specific interface type can be selected according to specific application needs, and is not specifically limited here.

[0134] In addition, the memory 22, as a carrier for resource storage, can be a read-only memory, random access memory, disk or optical disk, etc. The resources stored thereon can include operating system 221, computer program 222, etc., and the storage method can be temporary storage or permanent storage.

[0135] The operating system 221 is used to manage and control the various hardware devices on the electronic device 20 and the computer program 222, which may be Windows Server, Netware, Unix, Linux, etc. In addition to including a computer program capable of performing the intelligent vehicle route planning method for electronic waste recycling executed by the electronic device 20 as disclosed in any of the foregoing embodiments, the computer program 222 may further include computer programs capable of performing other specific tasks.

[0136] Furthermore, this application also discloses a computer-readable storage medium for storing a computer program; wherein, when the computer program is executed by a processor, it implements the aforementioned intelligent vehicle path planning method for electronic waste recycling. Specific steps of this method can be found in the corresponding content disclosed in the foregoing embodiments, and will not be repeated here.

[0137] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to in the method section.

[0138] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0139] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented directly by hardware, a software module executed by a processor, or a combination of both. The software module can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.

[0140] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0141] The technical solutions provided in this application have been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for intelligent vehicle route planning in electronic waste recycling, characterized in that, include: Official reservation recycling demand is defined as node demand, and unofficial street recycling demand is defined as edge demand. Then, the node demand and the edge demand are integrated to obtain the target recycling demand. A path optimization model is constructed based on preset soft time window constraints, preset vehicle capacity constraints, preset path closed loop conditions, preset profit maximization function and the target recycling demand. The polar coordinate angle between the required nodes and the target warehouse is determined using the AD algorithm and the angle partitioning algorithm. Then, the nodes are sorted based on the polar coordinate angles to obtain the sorting results. The path optimization model is then used to allocate paths to the nodes based on the sorting results to obtain the solution to be processed. Then, the tabu search algorithm and the simulated annealing algorithm are used to perform a neighborhood search operation on the solution to be processed according to the preset neighborhood search rules to obtain the corresponding candidate solution set. The first candidate solution in the candidate solution set is set as the current candidate solution. Then, it is determined whether the unprocessed profit of the current candidate solution is greater than the current maximum profit. If it is greater, the unprocessed profit is set as the current maximum profit. If it is not greater, the next candidate solution in the candidate solution set is set as the new current candidate solution, and the process jumps back to the step of determining whether the unprocessed profit of the current candidate solution is greater than the current maximum profit. This process continues until all candidate solutions in the candidate solution set have been determined. Then, the current candidate solution is set as the target solution, and the path of the electronic waste recycling vehicle is planned based on the target solution.

2. The intelligent vehicle route planning method for electronic waste recycling according to claim 1, characterized in that, The process involves defining official scheduled recycling demand as node demand and unofficial street-side recycling demand as edge demand. Then, by integrating the node and edge demands, a target recycling demand is obtained. A path optimization model is constructed based on preset soft time window constraints, preset vehicle capacity constraints, preset path closure conditions, a preset profit maximization function, and the target recycling demand. This includes: Official scheduled recycling needs are defined as node needs, and unofficial street-side recycling needs are defined as edge needs. Then, the node needs and edge needs are integrated to obtain the target recycling needs, which are graph-structured. The target recycling needs include nodes and edges. Each node corresponds to a node need, and each edge corresponds to an edge need. A path optimization model is trained and constructed based on preset soft time window constraints, preset vehicle capacity constraints, preset path closure conditions, a preset profit maximization function, and the target recycling demand. The preset soft time window constraint allows vehicles used for electronic waste recycling to arrive and wait before the specified time window corresponding to the node, and incurs a penalty if the vehicle arrives after the specified time window. The preset vehicle capacity constraint ensures that the loading capacity of each vehicle does not exceed its corresponding rated load. The preset path closure condition ensures that the starting and ending points of the path for each vehicle are both the target warehouse. The preset profit maximization function is a total profit maximization function determined based on total recycling revenue, fuel costs, labor costs, and penalties. The total recycling revenue includes the total revenue obtained by each vehicle in fulfilling node and edge demands. The labor costs include driver wages and overtime pay, and the penalty is the penalty imposed when each vehicle fails to meet the preset soft time window constraints.

3. The intelligent vehicle route planning method for electronic waste recycling according to claim 1, characterized in that, The process involves using the AD algorithm and angle partitioning algorithm to determine the polar coordinate angles between the nodes requiring each node and the target warehouse. Then, based on these polar coordinate angles, the nodes are sorted to obtain a sorting result. Finally, the path optimization model is used, and based on the sorting result, paths are allocated to each node to obtain the solution to be processed. This includes: The target warehouse is set as the origin of the coordinate system, and the node coordinates corresponding to each node are determined based on the origin of the coordinate system. Then, the corresponding polar coordinate angle is determined by using the angle division algorithm and the AD algorithm based on the origin of the coordinate system and the coordinates of each node. The nodes are sorted according to a preset direction and based on the polar coordinate angles to obtain the corresponding sorting results. The target service area is divided into several sector partitions using the path optimization model. Then, the nodes located in the same sector partition are assigned to the electronic waste recycling service path corresponding to the same vehicle to obtain the corresponding solution to be judged. Determine whether the solution to be determined satisfies the preset vehicle capacity constraint after being assigned to the service path. If the solution to be determined satisfies the preset vehicle capacity constraint after being assigned to the service path, update the path load corresponding to the service path to obtain the solution to be processed. If the solution to be judged, after being assigned to the service path, does not meet the preset vehicle capacity constraint, the service path is closed and a new service path is created to assign the solution to be judged to the new service path, thus obtaining the solution to be processed; the solution to be processed is the initial path planning scheme corresponding to each vehicle.

4. The intelligent vehicle route planning method for electronic waste recycling according to claim 1, characterized in that, The process involves using the tabu search algorithm and simulated annealing algorithm, and performing a neighborhood search operation on the solution to be processed according to a preset neighborhood search rule, to obtain a corresponding candidate solution set, including: The tabu search algorithm and simulated annealing algorithm are used to determine the preset neighborhood search rule corresponding to the solution to be processed. If the preset neighborhood search rule is a node swapping rule, then the two nodes in the single path of the solution to be processed are swapped, and the nodes in the multi-path of the solution to be processed are swapped to obtain the corresponding candidate solutions. If the preset neighborhood search rule is a node movement rule, then a node movement operation is performed on the single path in the solution to be processed, and the node in the first path in the solution to be processed is moved to the specified position of the second path to obtain the corresponding candidate solution. If the preset neighborhood search rule is the A-Opt transformation rule, then based on the value of A, the edges in the path of the solution to be processed are deleted and added to obtain the corresponding candidate solution; If the preset neighborhood search rule is a node reversal rule, then the sequence of each child node in the path of the solution to be processed is reversed; if the preset neighborhood search rule is a multi-path merging rule, then several paths in the solution to be processed that satisfy the preset vehicle capacity constraint are merged to obtain the corresponding candidate solutions. Construct a corresponding set of candidate solutions based on each of the aforementioned candidate solutions.

5. The intelligent vehicle route planning method for electronic waste recycling according to claim 1, characterized in that, After obtaining the corresponding candidate solution set by performing a neighborhood search operation on the solution to be processed using the tabu search algorithm and the simulated annealing algorithm according to a preset neighborhood search rule, the method further includes: The tabu list length is determined based on the number of candidate solutions in the candidate solution set. A tabu list is constructed based on the tabu list length. Neighborhood search operations performed on the solution to be processed are monitored, and the neighborhood search operations are recorded in the tabu list. The tabu list is used to record the neighborhood search operations performed within a preset time interval and the corresponding related attribute information. When performing a neighborhood search operation, check whether the neighborhood search operation is recorded in the taboo table. If the neighborhood search operation is not recorded in the taboo table, then record the neighborhood search operation in the taboo table. The simulated annealing algorithm is used to determine the current temperature parameter and the current temperature decrease rate, and the solution to be processed is subjected to a neighborhood search iterative operation according to a preset neighborhood search rule; wherein, the value of the current temperature parameter corresponding to each iteration decreases sequentially according to the current temperature decrease rate; the value of the current temperature parameter is negatively correlated with the fineness of the search.

6. The intelligent vehicle route planning method for electronic waste recycling according to claim 5, characterized in that, The process of setting the first candidate solution in the candidate solution set as the current candidate solution, then determining whether the unprocessed profit of the current candidate solution is greater than the current maximum profit, if it is greater, setting the unprocessed profit as the current maximum profit; if it is not greater, setting the next candidate solution in the candidate solution set as the new current candidate solution, and then returning to the step of determining whether the unprocessed profit of the current candidate solution is greater than the current maximum profit, continues until all candidate solutions in the candidate solution set have been determined, including: The first candidate solution in the candidate solution set is set as the current candidate solution. Then, it is determined whether the unprocessed profit of the current candidate solution is greater than the current maximum profit. If the unprocessed profit of the current candidate solution is greater than the current maximum profit, the desire criterion is triggered to unban the tabu list using the desire criterion. Then, the unprocessed profit is set as the current maximum profit. If the unprocessed profit of the current candidate solution is not greater than the current maximum profit, then the current temperature and the profit difference between the unprocessed profit and the current maximum profit are determined. Then, the Metropolis criterion is used and the acceptance probability is determined based on the current temperature and the profit difference. Based on the acceptance probability and a preset acceptance threshold, it is determined whether to accept the current candidate solution. Then, the process jumps back to the step of determining whether the unprocessed profit of the current candidate solution is greater than the current maximum profit, until all candidate solutions in the candidate solution set have been judged.

7. The intelligent vehicle route planning method for electronic waste recycling according to claim 1, characterized in that, After setting the current candidate solution as the target solution and planning the path of the electronic waste recycling vehicle based on the target solution, the method further includes: Adaptive adjustment rules are determined based on actual application scenarios to adaptively adjust the model parameters corresponding to the path optimization model, thereby obtaining an adjusted path optimization model. These adaptive adjustment rules include adjusting the model parameters based on newly added or canceled scheduled recycling tasks, adjusting the model parameters based on real-time identified opportunities for recycling along the street, and adjusting the model parameters based on changes in travel time caused by changes in road traffic conditions. The vehicle routes of vehicles used for electronic waste recycling are recalculated using a preset hybrid optimization algorithm and based on the adjusted path optimization model to obtain target vehicle routes, so as to dynamically and intelligently schedule each vehicle based on the target vehicle routes.

8. An intelligent vehicle route planning device for electronic waste recycling, characterized in that, include: The path optimization model construction module is used to define the official reservation recycling demand as node demand and the unofficial street recycling demand as edge demand. Then, the node demand and the edge demand are integrated to obtain the target recycling demand. The path optimization model is constructed based on preset soft time window constraints, preset vehicle capacity constraints, preset path closed loop conditions, preset profit maximization function and the target recycling demand. The candidate solution set determination module is used to determine the polar coordinate angle between the node requiring each node and the target warehouse using the AD algorithm and the angle partitioning algorithm. Then, the nodes are sorted based on the polar coordinate angle to obtain the sorting result. The path optimization model is used to allocate paths to each node based on the sorting result to obtain the solution to be processed. Then, the tabu search algorithm and the simulated annealing algorithm are used to perform a neighborhood search operation on the solution to be processed according to the preset neighborhood search rules to obtain the corresponding candidate solution set. The path planning module is used to set the first candidate solution in the candidate solution set as the current candidate solution, and then determine whether the unprocessed profit of the current candidate solution is greater than the current maximum profit. If it is greater, the unprocessed profit is set as the current maximum profit. If it is not greater, the next candidate solution in the candidate solution set is set as the new current candidate solution, and the process jumps back to the step of determining whether the unprocessed profit of the current candidate solution is greater than the current maximum profit, until all candidate solutions in the candidate solution set have been determined. Then, the current candidate solution is set as the target solution, so as to plan the path of the electronic waste recycling vehicle based on the target solution.

9. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor for executing the computer program to implement the intelligent vehicle route planning method for electronic waste recycling as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, Used to store a computer program, wherein the computer program, when executed by a processor, implements the intelligent vehicle path planning method for electronic waste recycling as described in any one of claims 1 to 7.

Images