Charging lane deployment method and system based on dual-objective optimization

Abstract

The invention relates to the technical field of lane deployment, and provides a charging lane deployment method and system based on dual-objective optimization, and the method comprises the steps: initializing a plurality of particles, and coding whether each candidate road segment is deployed with a wireless charging lane or not and the deployment length through each particle; for each particle, executing a deployment-traffic distribution iteration process, and calculating an operation income and a user comprehensive cost; iterative selection and updating of particles are carried out by taking maximization of operation income and minimization of user comprehensive cost as targets to finally obtain an optimal solution, and the optimal solution is decoded into a wireless charging lane deployment scheme. The vehicle charging can be more effectively induced without excessively increasing congestion or detouring distance, and the practical feasibility and the overall social benefit of the scheme are improved.

Description

Technical Field

[0001] This invention belongs to the field of lane deployment technology, and in particular relates to a charging lane deployment method and system based on dual-objective optimization. Background Technology

[0002] The statements in this section are merely background information related to the present invention and do not necessarily constitute prior art.

[0003] In recent years, the market size of electric vehicles has expanded rapidly. With the increase in the number of EVs (pure electric vehicles), the demand for charging has surged. However, the construction of charging infrastructure has lagged behind, and the problem of charging difficulties has become increasingly prominent. Traditional plug-in charging mode is limited by physical interfaces and charging time, which restricts the promotion of electric vehicles. Battery replacement and other solutions also face challenges such as inconsistent standards and high costs.

[0004] Against this backdrop, dynamic wireless charging technology has emerged: by embedding inductive charging facilities into the road, it replenishes the vehicle's power while it is in motion, thereby effectively alleviating the "range anxiety" of EVs and reducing the reliance on large-capacity batteries.

[0005] The core feature of Wireless Charging Lanes (WCLs) is that they charge simultaneously with vehicle movement, eliminating the need for additional vehicle stops. Unlike fixed charging stations, WCLs extend charging functionality to road segments, integrating driving and charging. This unique model couples vehicle route selection with charging behavior, rendering traditional fixed-station-centric planning methods unsuitable. New layout planning research tailored to the characteristics of WCLs is required. Currently, the demand for dynamic wireless charging in cities is increasing daily. A key issue is how to scientifically plan the deployment locations and power supply capacity of WCLs across the entire road network. Inadequate WCL configuration may lead to low facility utilization or excessive detours for users to charge, harming user experience and hindering the maximization of operational revenue.

[0006] After the charging infrastructure changed from discrete sites to continuous road segments, new technical problems emerged in the planning process, such as conflicting objectives, coupled decision-making, a large decision space, and increased computational complexity. Summary of the Invention

[0007] To address the technical problems mentioned above, this invention provides a charging lane deployment method and system based on dual-objective optimization. It considers maximizing revenue and minimizing user costs as equally important objectives, balancing the demands of infrastructure providers and users. Furthermore, it incorporates vehicle route selection behavior into the deployment scheme evaluation, simulating vehicle responses through iterative traffic allocation. This allows the optimization process to predict drivers' choices when encountering different wireless charging lane layouts. In practical implementation, this method can more effectively guide vehicles to charge without excessively increasing congestion or detour distances, thus improving the practical feasibility and overall social benefits of the solution.

[0008] To achieve the above objectives, the present invention adopts the following technical solution: The first aspect of the present invention provides a charging lane deployment method based on dual-objective optimization, comprising: Initialize several particles, each particle encoding whether a wireless charging lane is deployed on each candidate road segment and the deployment length; For each particle, an iterative deployment-traffic assignment process is executed to calculate operational revenue and overall user cost. This process includes: calculating the unit segment travel cost for each candidate road segment based on whether a wireless charging lane is deployed and its deployment length; for any candidate road segment, calculating the overall travel cost for each vehicle choosing that candidate road segment based on the unit segment travel cost, determining the probability of each vehicle choosing that candidate road segment, and calculating the total traffic flow for that candidate road segment based on the probability of all vehicles choosing that candidate road segment; and for each candidate road segment, calculating operational revenue and overall user cost based on the total traffic flow, whether a wireless charging lane is deployed, and its deployment length. With the goal of maximizing operational revenue and minimizing overall user costs, the particles are iteratively selected and updated to obtain the optimal solution, which is then decoded into a wireless charging lane deployment scheme.

[0009] Furthermore, the operating revenue is the sum of the electrical energy obtained by all vehicles on each deployment segment.

[0010] Furthermore, the user's comprehensive cost is the total cost incurred in all actual trips, including basic travel time cost, energy consumption cost, and detour cost.

[0011] Furthermore, each particle needs to satisfy the following constraints: endurance constraints and path selection behavior constraints.

[0012] Furthermore, the unit road segment travel cost is: ;in, The vehicle speed changes dynamically with the deployment ratio. , As the baseline vehicle speed, The velocity attenuation coefficient, Indicate whether it is on the road section Deploy wireless charging lanes, Indicates the section of road The length of the wireless charging lane to be deployed, Indicates road segment Total length, The vehicle is on the road section Energy consumption per unit of driving, For vehicles on road sections where wireless charging lanes have been deployed The equivalent charging amount obtained above, and This is the conversion factor.

[0013] Furthermore, the probability is ;in, For each vehicle Select road segment The overall travel cost Refers to the path selection sensitivity coefficient. For vehicles The set of feasible paths; For vehicles Select road segment The probability of.

[0014] Furthermore, the total traffic flow of a candidate road segment is the weighted sum of the probabilities of all vehicles choosing that candidate road segment.

[0015] A second aspect of the present invention provides a charging lane deployment system based on dual-objective optimization, comprising: The initialization module is configured to initialize several particles, each particle encoding whether a wireless charging lane is deployed on each candidate road segment and the deployment length; The target calculation module is configured to: for each particle, execute a deployment-traffic allocation iterative process to calculate operational revenue and overall user cost; wherein, the deployment-traffic allocation iterative process includes: calculating the unit segment travel cost of each candidate road segment based on whether a wireless charging lane is deployed and the deployment length of each candidate road segment; for any candidate road segment, calculating the overall travel cost of each vehicle choosing that candidate road segment based on the unit segment travel cost, determining the probability of each vehicle choosing that candidate road segment, and calculating the total traffic flow of that candidate road segment based on the probability of all vehicles choosing that candidate road segment; for each candidate road segment, calculating operational revenue and overall user cost based on the total traffic flow, whether a wireless charging lane is deployed, and the deployment length. The iterative optimization module is configured to iteratively select and update particles with the goal of maximizing operational revenue and minimizing overall user costs, and finally obtain the optimal solution, which is then decoded into a wireless charging lane deployment scheme.

[0016] A third aspect of the invention provides a computer-readable storage medium having a computer program stored thereon that, when executed by a processor, implements the steps of a charging lane deployment method based on dual-objective optimization as described above.

[0017] A fourth aspect of the present invention provides a computer device including a computer-readable storage medium, a processor, and a computer program stored on the computer-readable storage medium and executable on the processor, wherein the processor executes the program to implement the steps of the charging lane deployment method based on dual-objective optimization as described above.

[0018] Compared with the prior art, the beneficial effects of the present invention are: This invention considers maximizing revenue and minimizing user costs as equally important objectives, balancing the demands of infrastructure providers and users. Furthermore, it incorporates vehicle routing behavior into the deployment scheme evaluation and simulates vehicle responses through iterative traffic allocation. This allows the optimization process to predict drivers' choices when encountering different wireless charging lane layouts. In practical implementation, this can more effectively guide vehicles to charge without excessively increasing congestion or detour distances, thus improving the practical feasibility and overall social benefits of the solution. Attached Figure Description

[0019] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0020] Figure 1 This is a flowchart of a charging lane deployment method based on dual-objective optimization according to Embodiment 1 of the present invention; Figure 2 This is a flowchart of the multi-objective particle swarm optimization solution in Embodiment 1 of the present invention; Figure 3 This is a schematic diagram of the structure of a computer device according to Embodiment 4 of the present invention. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings.

[0022] It should be noted that the following detailed description is illustrative and intended to provide further explanation of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0023] Example 1 This embodiment provides a charging lane deployment method based on dual-objective optimization.

[0024] This embodiment provides a charging lane deployment method based on dual-objective optimization, which realizes hierarchical dynamic routing with real-time path planning, computing task scheduling and energy replenishment collaborative optimization in a large-scale vehicle-computing-energy integrated network.

[0025] This embodiment provides a charging lane deployment method based on dual-objective optimization, including mathematical modeling, deployment and traffic assignment feedback coupling, solution algorithm design, Pareto solution selection and decision-making, such as... Figure 1 As shown, the specific steps include the following: Step 1: Establish a mathematical model by modeling urban road network data, electric vehicle parameters, and wireless charging lanes (WCL), and pass them as input to a multi-objective optimization solver based on the two objective functions of maximizing operational revenue and minimizing overall user costs.

[0026] (1) Modeling decision variables.

[0027] This embodiment considers an urban road network consisting of a set of candidate road segments. Composition, each road segment Dynamic wireless charging lanes (WCLs) can be deployed in all of these scenarios. Therefore, the following decision variables are introduced: Indicate whether it is on the road section Deploy WCL; Indicates the section of road The length (meters) of the deployed charging section; the constraint is: if no charging section is deployed ( ), then set If deployed ( If so, the deployment length must meet the following requirements. (It shall not exceed the total length of the road section) These variables provide a complete description of a given WCL deployment scheme. .

[0028] (2) Setting the objective function.

[0029] Based on the above decision variables, this embodiment constructs a "dual-objective" optimization model, which includes two objectives: operational revenue and overall user cost.

[0030] (A) Maximizing operating revenue ( The price per unit of electricity is set as follows: The total electrical energy collected by all vehicles traveling through the deployed WCL sections is Then operating revenue The first objective of this embodiment is to maximize this benefit, namely ,in, This can be further represented as the sum of electrical energy obtained by all vehicles on each deployment segment: ,(1); Among them, the function Indicates the section of road The deployment length is After a charging section is established, the electrical energy obtained by all vehicles passing through that section; for any given section... If it is deployed with WCL (i.e. ), and the deployment length is Then, the power received by this section of road within a certain period of time is: (2); in, For time Passing sections during certain time periods Electric vehicle density (vehicles / h). WCL is the power density per unit length (kW / m). For road section In time The average traffic speed (m / s). To indicate that the vehicle is in the WCL section Travel time (in seconds), integration period The above formula outputs kWh by multiplying the average value by the time weight. This function expresses that the longer the deployment, the denser the traffic flow, and the slower the driving, the more can be charged per unit time, and the higher the operating revenue.

[0031] (B) Minimize overall user cost This embodiment sets the user cost during the electric vehicle travel process. If the total user cost for all vehicle trips is C, then the second objective is to minimize it. ;in, This represents the total cost incurred by each vehicle k during its actual journey, comprising the following three components: Basic travel time cost: i.e., vehicle Along the path Total time cost of normal driving; Energy consumption cost: the cost of a vehicle's journey The equivalent cost of the total electrical energy consumed during driving; Detour costs: Due to the influence of charging layout on route selection, vehicles may have to detour or reduce speed, resulting in additional delays and energy consumption.

[0032] To facilitate standardized measurement, this embodiment converts these costs into a common currency unit (e.g., yuan) for weighted summation. Therefore, the vehicle... In its selected path The total travel cost is recorded as follows: (3); in, For vehicles A set of feasible paths The vehicle is on the road section The unit travel time (in minutes) can be expressed as ,in It is length. It is the average traffic speed after deployment; It is the energy consumption per unit of driving on a road segment (kilowatt-hours). Vehicles on sections of road where WCL has been deployed The equivalent charging amount obtained is used to offset the cost of future additional energy replenishment, and thus it is reflected as a benefit (i.e. a cost negative item) in the total cost. These are the conversion value coefficients for unit time, unit energy consumption, and unit electricity revenue (units: yuan / minute, yuan / kWh, yuan / kWh).

[0033] (3) Constraints.

[0034] While optimizing the above objectives, the model needs to meet some key constraints to ensure the feasibility and rationality of the solution.

[0035] (A) Range Constraint: The amount of electricity that all vehicles obtain through the deployed WCL on their respective chosen travel routes should not be less than the amount of electricity required for their journey. For any vehicle k and its chosen route... , (4); in, Let be the function of the amount of charge that the vehicle can obtain on the WCL segment of road segment i; The minimum amount of electricity required for vehicle k to complete this trip; ensuring that each vehicle can complete the journey without having to stop and charge, is equivalent to requiring the deployment plan to provide sufficient charging along the way to guarantee the vehicle's range.

[0036] (B) Route Selection Behavior Constraints: The vehicle's route selection follows the principle of user rationality, that is, it is assumed that each user will choose the route with the lowest overall cost to themselves. To characterize this behavior, this embodiment considers the charging factor along the way as part of the route cost. Let the vehicle... The set of feasible paths is Record the path The overall cost is Under the ideal condition of deterministic user equilibrium (UE), the vehicle The optimal path actually chosen is denoted as It should satisfy the condition for any feasible alternative path. have: (5); The above inequality ensures that the overall cost of each user's path is no higher than the cost of any other feasible path, reflecting the requirement of deterministic user equilibrium: all actually selected paths are the shortest (lowest cost) paths for the corresponding trip.

[0037] The aforementioned bi-objective optimization model aims to maximize operational revenue. At the same time, minimize the overall cost to users. Find a set of non-dominated solutions (Pareto solutions) that satisfy the constraints, where To deploy location decision variables, indicating whether it is on the road segment Wireless charging lanes are deployed on top. The deployment length decision variable represents the specific deployment length on the selected road segment. Since the two objectives are often conflicting (increasing revenue usually requires more vehicles to detour for charging, thus increasing user costs, while reducing user costs may decrease revenue), the model does not have a single global solution that simultaneously optimizes both objectives. Instead, it generates a series of Pareto optimal solutions, each with its own advantages and disadvantages: any solution that attempts to improve one objective will deteriorate in another. The next step is to find this series of compromise solutions using a solution algorithm.

[0038] Step 2: Using the introduced parameters, the multi-objective optimization solver generates an abstract deployment scheme. Specifically, based on the iterative feedback mechanism of scheme-traffic assignment, it simulates the impact of WCL deployment on vehicle route selection, thereby realistically evaluating the benefits and costs of each scheme.

[0039] In the process of solving the model, this embodiment considers the impact of WCL deployment on vehicle route selection, that is, the two-way feedback coupling relationship between deployment scheme and traffic travel. As mentioned above, vehicles will choose the optimal route based on the comprehensive travel cost on the road network, and the deployment of wireless charging lanes will directly change the travel cost of the road.

[0040] On the one hand, road sections with wireless charging (WCL) allow vehicles to charge while driving, effectively replenishing their battery power. From the driver's perspective, this reduces future static charging time or avoids additional charging stops, thus lowering overall travel costs. On the other hand, if the wireless charging section is too long or a vehicle attempts to acquire more power, it may need to reduce its speed to extend charging time, increasing travel time costs for that section. This process forms a closed loop between deployment decisions (upper layer) and traffic allocation (lower layer). Upward feedback influences path costs, while downward feedback influences traffic allocation and revenue.

[0041] To assess the interaction between deployment plans and vehicle travel, this embodiment introduces a feedback coupling mechanism of deployment-route selection-benefit evaluation during the optimization process: the deployment plan affects route costs and vehicle selection behavior, while route allocation determines the utilization and actual benefits of different road segments, thus in turn affecting the fitness evaluation of the deployment plan. This mechanism has been embedded in the solution algorithm and implemented through an iterative traffic assignment process.

[0042] To accurately assess the aforementioned interactions during the solution process, this embodiment introduces an iterative traffic allocation process when the algorithm evaluates each candidate deployment scheme. Specifically, for any given deployment scheme generated by the algorithm... This embodiment performs the following steps to evaluate the corresponding traffic operation state and objective function value: Mapping changes in travel costs: In each iteration, the algorithm first reads the WCL settings (i.e., deployment location and deployment length) of each road segment in the current deployment plan, and dynamically updates the comprehensive travel cost function of each road segment in the road network accordingly. For any road segment... The unit travel cost per road segment is defined as: (6); in, The vehicle speed, which changes dynamically with the deployment ratio, can be represented by an empirical function as follows: (7); in, The baseline vehicle speed in the undeployed state. The formula represents the speed attenuation coefficient. This formula illustrates that when the deployment length of a road segment is long, vehicles may moderately reduce their speed due to the extended charging time, thereby increasing time costs. Conversely, shorter deployments only bring charging benefits, with minimal impact on time costs. Therefore, each vehicle Select path Expected overall travel cost Updatable is represented as: (8); After updating the costs of each road segment, a vehicle route selection model based on the User Experience (UE) principle is constructed. It is assumed that each driver possesses complete road network cost information and tends to choose the route that minimizes their overall travel cost. Once all vehicles have completed their route adjustments, the system reaches a new equilibrium state, where the costs of all selected routes are equal and no higher than the costs of other feasible routes.

[0043] Vehicle route reallocation: After updating the travel costs of each road segment, the algorithm reallocates vehicle traffic using a route selection model to reflect drivers' travel response behavior under the new cost conditions. Vehicle route selection follows a Stochastic User Equilibrium (SUE) model, which assumes that each driver chooses a route with a certain probability given differences in perceived costs. While the model analysis assumes deterministic equilibrium, the SUE model is used to approximate user route selection behavior in the solution to improve realism. For vehicles... Its set of feasible paths is Each path The selection probability is defined as: (9); in, For path The expected overall travel cost; This refers to the route selection sensitivity coefficient; the larger the value, the more sensitive the driver is to cost differences. For vehicles Select path The probability of , the probability of all paths sum to 1, that is .

[0044] The traffic distribution of vehicles in the network is obtained by superimposing the probabilities of each path, denoted as... For road section The total traffic flow on the road is as follows: (10); in, For vehicle assembly, For vehicles The corresponding travel demand weights, and the second summation symbol indicates that only the route segments are counted. The formula represents the mapping relationship between path selection and traffic flow on road segments. Deployment scheme Changed expected path cost Therefore, the path selection probability is changed through equation (9). Ultimately, this affects the traffic flow distribution on each road segment. This process is executed in real time during algorithm iteration, enabling traffic allocation to dynamically respond to adjustments in the deployment plan, thereby providing updated network status for subsequent benefit and cost evaluations.

[0045] Evaluation metrics for the statistical scheme: Based on the new traffic allocation results, two target values ​​corresponding to this deployment scheme are calculated: operational revenue. Total User Cost In specific calculations, the number of charging vehicles and the amount of electricity received on each deployment segment can be summarized: for each deployment segment... It calculates how many vehicles pass by and charge at this location, and how much energy they accumulate. Then sum them up to get In turn, one can gain benefits. At the same time, the total travel costs after all vehicle upgrades are calculated. It needs to be emphasized that in calculation The changes in vehicle behavior due to charging (such as detour distance, increased time due to reduced speed, etc.) have already been taken into account. ,therefore The value may increase or decrease compared to when it is not deployed, depending on the overall effect of the deployment plan.

[0046] The deployment scheme alters the costs of various paths on the road network, causing vehicles to reselect routes and recharge. The resulting revenue and costs are then fed back to evaluate the effectiveness of the deployment scheme, guiding the upper-level optimization algorithm to adjust the scheme. This step embeds a complex problem, originally involving upper-level network design and lower-level traffic assignment, into a single-objective evaluation through simulation and iteration, providing the optimization algorithm with an accurate fitness function.

[0047] Step 3: Design an improved multi-objective particle swarm optimization algorithm to solve the model and obtain a set of Pareto optimal solutions with different trade-offs between benefits and costs.

[0048] After establishing the model and determining the evaluation mechanism, this embodiment needs to select an appropriate optimization algorithm to solve the bi-objective problem. This embodiment uses the Multi-Objective Particle Swarm Optimization (MOPSO) algorithm to solve the above model. Particle Swarm Optimization (PSO) is an evolutionary algorithm that simulates swarm intelligence behavior. It gradually approaches the optimal solution through information sharing among particles; its multi-objective extension is called MOPSO, which can obtain a set of solutions approximating the Pareto front in a single run. MOPSO is chosen because PSO has advantages such as fast convergence speed, simple implementation, and easy parameter adjustment. Furthermore, a single run can obtain multiple non-dominated solutions and approximate non-convex or discontinuous Pareto fronts. These characteristics make it a powerful tool for solving this problem (with conflicting objectives and complex constraints), such as... Figure 2 As shown, the solution process for MOPSO is as follows: (1) Particle Encoding and Initial Population: In MOPSO, this embodiment represents each particle encoding as a candidate WCL deployment scheme. A length of [missing information] can be used. The vector represents a particle, containing genetic information such as whether each candidate road segment is deployed and its deployment length. For each road segment... The genes of particles can correspond to and The process consists of two parts. During initialization, a random population of particles is generated, each representing a feasible deployment scheme (which must satisfy basic constraints, such as ensuring a vehicle is not rendered inaccessible). Then, using the aforementioned traffic assignment evaluation process, two objective values ​​are calculated for each particle. This serves as its fitness level. At this point, each particle corresponds to a specific benefit and user cost value.

[0049] (2) Particle Swarm Iterative Update: The MOPSO algorithm enters the main iterative loop. In each iteration, the particles generate new schemes according to the velocity-position update rule of PSO. Specifically, the particles... The velocity and position update formulas are:

[0050] ; in, It is inertial weight. It is the acceleration coefficient. It is a random number between 0 and 1; It is a particle Its best position in its own history (explanation). This represents a kind of global optimal solution. In the multi-objective scenario, "optimal" is no longer a single solution, but a set of Pareto optimal solutions. Therefore, MOPSO improves upon PSO by introducing an external archive to maintain all currently found non-dominated solutions and employing mechanisms such as crowding to ensure the diversity of the solution set. Specifically, each iteration includes the following operations: (A) Fitness Evaluation: For each particle's new position (i.e., new deployment scheme), the deployment-traffic assignment iterative process is executed again to calculate its new fitness. This yields the target values ​​for all candidate solutions generated in this iteration.

[0051] (B) Archive Update: The new solutions generated in this iteration are combined with the previously stored solutions and the external archive is updated according to the Pareto dominance relationship. That is: first, archive solutions dominated by new solutions are discarded; then, new solutions not dominated by any solutions in the archive are added to the archive; if there are too many solutions in the archive, some overly dense solutions can be deleted according to indicators such as crowding distance to maintain archive capacity and uniform distribution. The Archive always stores the set of currently discovered Pareto front approximate solutions.

[0052] (C) Individual extreme value update: Update the historical best position for each particle. In multi-objective problems, "historical optimality" refers to the solution that the particle has experienced during its evolution, which is currently not dominated and is related to the current solution. A comparable solution. If the new position is similar to the original... If they do not control each other, they can choose to update randomly with a 50% probability according to the preset strategy.

[0053] (D) Global Guided Selection: Unlike single-target PSO, which directly selects the best particle in the current population, MOPSO requires selecting one from the many Pareto solutions in the archive as the global guide. This method evaluates the sparsity of solutions in an archive based on crowding, randomly selecting solutions that are less crowded (i.e., have fewer nearby solutions and are relatively sparse regions) as the basis for evaluation. This encourages particles to search for weak regions at the Pareto front, maintaining the diversity of the solution set.

[0054] (E) Particle velocity and position update: using the selected and Substituting into the velocity update formula above, calculate the new velocity of each particle and update its position (deployment plan). At this point, it's important to adjust the particle solutions to meet constraints (e.g., ensure...). No more than (etc.), if the particle solution appears after the update or but If any unreasonable situation occurs, the solution is truncated or reset to ensure that the constraints are satisfied again. After the update is complete, the next generation loop begins.

[0055] The above iterations continue until a preset algebraic or convergence criterion is reached. Ultimately, the external archive stores a set of Pareto-optimal deployment schemes with different trade-offs between benefits and costs. Because no weights are pre-assigned to the two objectives during the optimization process, the algorithm's search process maintains a diverse exploration of various schemes, resulting in a solution set that simultaneously includes schemes prioritizing benefit maximization, cost minimization, and compromise solutions in between. This provides decision-makers with a rich set of candidate schemes from which they can select the deployment scheme that best meets their needs based on preferences.

[0056] Step 4: Using the entropy weight multi-criteria decision-making method, select a satisfactory deployment scheme that takes into account the interests of both parties from the Pareto solution set.

[0057] While all Pareto solutions generated by multi-objective optimization are valid solutions, they may be numerous and each has its own advantages and disadvantages. Therefore, this embodiment introduces a multi-criteria decision analysis (MCDA) step after optimization, using objective weighting methods such as entropy weighting to score and rank the Pareto solutions, and select the final wireless charging lane deployment scheme.

[0058] Standardization of evaluation metrics: First, the two objective values ​​of each scheme in the Pareto solution set are made dimensionless. Because... It's about the returns (the higher the better). Costs (the lower the better) can be converted into "benefit-oriented indicators" (the higher the better) for all options. , Normalize them separately to ensure that different indicators are comparable.

[0059] Entropy weighting is an objective weighting method that determines weights based entirely on the inherent differences in the data. Specifically, it calculates the dispersion of each objective's value across all options: the greater the numerical difference of an objective among candidate options, the more information it provides to distinguish different options, resulting in lower information entropy and a higher weight. Conversely, if an objective's value changes very little across all options, it means different options perform almost identically on that objective, leading to higher information entropy, lower importance, and a lower weight. In short, higher information entropy means a more uniform distribution of indicators, providing less information and thus a lower weight; lower information entropy means greater differences in indicators, providing more information and thus a higher weight. Using this objective weighting method, this approach defines the weight coefficients for two objectives. and (and This reflects their relative importance in distinguishing the merits of different schemes.

[0060] Comprehensive Scoring and Solution Selection: After obtaining the weights, this embodiment uses a weighted normalized distance method to score each Pareto solution. Each solution is evaluated based on its distance from the two ideal points: the closer the solution is to the optimal point and the farther it is from the worst point, the higher its comprehensive score. Combining the weights determined by entropy weights, this method calculates the weighted performance value of each candidate solution and ranks them accordingly. Finally, the solution with the highest score is considered the optimal compromise solution and recommended.

[0061] Through the aforementioned entropy-weighted multi-criteria decision-making steps, this embodiment objectively selects an optimal solution from the Pareto solution set that balances the interests of both operators and users. The final deployment scheme will achieve a balance between revenue and user costs, maximizing infrastructure operational efficiency while meeting vehicle charging needs, thus providing strong support for the construction of urban dynamic wireless charging networks.

[0062] Compared with existing infrastructure planning methods that only perform single-objective optimization or do not consider the impact of driving and charging, the WCL deployment method based on dual-objective optimization proposed in this embodiment has the following advantages: (1) Simultaneously optimize operational revenue and user experience: Traditional planning often focuses on either the interests of the operator or the convenience of the user. However, this embodiment considers maximizing revenue and minimizing user costs as equally important goals, balancing the demands of the infrastructure provider and the user. The optimization results enable the deployment scheme to increase charging revenue while minimizing the additional driving delay and energy consumption of vehicles for charging, achieving a "win-win" effect.

[0063] (2) Fully tapping the charging potential of the road network: Since it allows the selection of deployment sections across the entire network without a preset budget limit, this embodiment can automatically select key road sections that contribute the most to revenue and cost through optimization. For example, the result often prioritizes the deployment of WCLs on trunk roads with high electric vehicle traffic and long travel distances, and reasonably determines the laying length of each road section, so that vehicles can obtain sufficient power replenishment without significant detours. This not only improves the utilization rate of facilities and the return on unit investment, but also avoids the waste caused by blindly laying on a large scale.

[0064] (3) Considering vehicle behavior feedback is more realistic: This embodiment incorporates vehicle route selection behavior into the deployment scheme evaluation. By simulating vehicle response through iterative traffic allocation, the optimization process "predicts" the driver's choice response when encountering different WCL layouts. Compared with static planning that does not consider traffic flow adjustment, the scheme obtained in this embodiment can more effectively induce vehicles to charge without excessively increasing congestion or detour distance in actual implementation, thus improving the practical feasibility and overall social benefits of the scheme.

[0065] (4) The algorithm is efficient and provides decision support: The MOPSO algorithm combined with Pareto solution archiving is adopted, which enables efficient optimization to obtain an approximate Pareto front solution set even on complex large-scale road networks. Then, the MCDM method with objective entropy weighting selects the optimal compromise solution from many preferred solutions, avoiding the subjectivity of human experience in selection. The whole method has a clear process and high computational efficiency, and can be used as a decision support tool to help urban planners quickly formulate scientific wireless charging lane layout schemes.

[0066] Example 2 This embodiment provides a charging lane deployment system based on dual-objective optimization, comprising: The initialization module is configured to initialize several particles, each particle encoding whether a wireless charging lane is deployed on each candidate road segment and the deployment length; The target calculation module is configured to: for each particle, execute a deployment-traffic allocation iterative process to calculate operational revenue and overall user cost; wherein, the deployment-traffic allocation iterative process includes: calculating the unit segment travel cost of each candidate road segment based on whether a wireless charging lane is deployed and the deployment length of each candidate road segment; for any candidate road segment, calculating the overall travel cost of each vehicle choosing that candidate road segment based on the unit segment travel cost, determining the probability of each vehicle choosing that candidate road segment, and calculating the total traffic flow of that candidate road segment based on the probability of all vehicles choosing that candidate road segment; for each candidate road segment, calculating operational revenue and overall user cost based on the total traffic flow, whether a wireless charging lane is deployed, and the deployment length. The iterative optimization module is configured to iteratively select and update particles with the goal of maximizing operational revenue and minimizing overall user costs, and finally obtain the optimal solution, which is then decoded into a wireless charging lane deployment scheme.

[0067] It should be noted that each module in this embodiment corresponds one-to-one with each step in Embodiment 1, and their specific implementation processes are the same, so they will not be repeated here.

[0068] Example 3 This embodiment provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps in the dual-objective optimization-based charging lane deployment method described in Embodiment 1 above.

[0069] Example 4 This embodiment provides a computer device, such as... Figure 3As shown, the system includes a computer-readable storage medium 1003, a processor 1001, a communication interface 1002, and a computer program stored on the computer-readable storage medium 1003 and executable on the processor 1001. The processor 1001, communication interface 1002, and computer-readable storage medium 1003 can be connected via a bus or other means. The communication interface 1002 is used to receive and transmit data. When the processor 1001 executes the program, it implements the steps in the dual-objective optimization-based charging lane deployment method described in Embodiment 1 above.

[0070] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A charging lane deployment method based on dual-objective optimization, characterized in that, include: Initialize several particles, each particle encoding whether a wireless charging lane is deployed on each candidate road segment and the deployment length; For each particle, an iterative deployment-traffic assignment process is executed to calculate operational revenue and overall user cost. This process includes: calculating the unit segment travel cost for each candidate road segment based on whether a wireless charging lane is deployed and its deployment length; for any candidate road segment, calculating the overall travel cost for each vehicle choosing that candidate road segment based on the unit segment travel cost, determining the probability of each vehicle choosing that candidate road segment, and calculating the total traffic flow for that candidate road segment based on the probability of all vehicles choosing that candidate road segment; and for each candidate road segment, calculating operational revenue and overall user cost based on the total traffic flow, whether a wireless charging lane is deployed, and its deployment length. With the goal of maximizing operational revenue and minimizing overall user costs, the particles are iteratively selected and updated to obtain the optimal solution, which is then decoded into a wireless charging lane deployment scheme.

2. The charging lane deployment method based on dual-objective optimization as described in claim 1, characterized in that, The operating revenue is the sum of the electrical energy obtained by all vehicles on each deployed road segment.

3. The charging lane deployment method based on dual-objective optimization as described in claim 1, characterized in that, The user's comprehensive cost is the total cost incurred in all actual trips, including basic travel time cost, energy consumption cost, and detour cost.

4. The charging lane deployment method based on dual-objective optimization as described in claim 1, characterized in that, Each particle needs to satisfy the following constraints: endurance constraints and path selection behavior constraints.

5. The charging lane deployment method based on dual-objective optimization as described in claim 1, characterized in that, The unit travel cost per road segment is: ;in, The vehicle speed changes dynamically with the deployment ratio. , As the baseline vehicle speed, The velocity attenuation coefficient, Indicate whether it is on the road section Deploy wireless charging lanes, Indicates the section of road The length of the wireless charging lane to be deployed, Indicates road segment Total length, It is a vehicle on the road section Energy consumption per unit of driving, For vehicles on road sections where wireless charging lanes have been deployed The equivalent charging amount obtained above, and This is the conversion factor.

6. The charging lane deployment method based on dual-objective optimization as described in claim 1, characterized in that, The probability is ;in, For each vehicle Select road segment The overall travel cost Refers to the path selection sensitivity coefficient. For vehicles The set of feasible paths; For vehicles Select road segment The probability of.

7. The charging lane deployment method based on dual-objective optimization as described in claim 1, characterized in that, The total traffic flow of a candidate road segment is the weighted sum of the probabilities of all vehicles choosing that candidate road segment.

8. A charging lane deployment system based on dual-objective optimization, characterized in that, include: The initialization module is configured to initialize several particles, each particle encoding whether a wireless charging lane is deployed on each candidate road segment and the deployment length; The target calculation module is configured to: for each particle, execute a deployment-traffic allocation iterative process to calculate operational revenue and overall user cost; wherein, the deployment-traffic allocation iterative process includes: calculating the unit segment travel cost of each candidate road segment based on whether a wireless charging lane is deployed and the deployment length of each candidate road segment; for any candidate road segment, calculating the overall travel cost of each vehicle choosing that candidate road segment based on the unit segment travel cost, determining the probability of each vehicle choosing that candidate road segment, and calculating the total traffic flow of that candidate road segment based on the probability of all vehicles choosing that candidate road segment; for each candidate road segment, calculating operational revenue and overall user cost based on the total traffic flow, whether a wireless charging lane is deployed, and the deployment length. The iterative optimization module is configured to iteratively select and update particles with the goal of maximizing operational revenue and minimizing overall user costs, and finally obtain the optimal solution, which is then decoded into a wireless charging lane deployment scheme.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the steps in the dual-objective optimization-based charging lane deployment method as described in any one of claims 1-7.

10. A computer device comprising a computer-readable storage medium, a processor, and a computer program stored on the computer-readable storage medium and executable on the processor, characterized in that, When the processor executes the program, it implements the steps in the charging lane deployment method based on dual-objective optimization as described in any one of claims 1-7.

Images