EV Route Planning With On-Route Charging and Battery Swapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing electric vehicle routing systems face challenges in optimizing multi-service deliveries and on-route energy replenishment, particularly in large-scale scenarios with hundreds of users, due to unpredictable charging patterns, non-availability of captive chargers, and high capital costs, which impact operational efficiency and user experience.
Innovation Solution
A system utilizing a trained learning agent and neural network to generate optimized route maps for electric vehicles, considering factors like battery swapping and discharging stations, to minimize trip cost and time duration, while incorporating energy replenishment strategies such as vehicle-to-grid and battery swapping.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If electric vehicles are used for last mile delivery, then operational cost and emission are reduced, but charging pattern unpredictability and capital cost increase
Solution Approach 1:
The system performs preliminary routing planning that anticipates future charging needs by integrating charging station locations and availability into the route optimization. The planner pre-identifies optimal charging stops and timing before the vehicle departs, transforming unpredictable charging into a planned operational element.
Solution Approach 2:
The system introduces a centralized routing planner as an intermediary between the vehicle operator and charging infrastructure. This planner accesses real-time charging station data, battery status, and delivery constraints to mediate the charging process, converting unpredictable external charging availability into controlled, optimized charging decisions.
2Adaptability or versatility
If multi-service deliveries are implemented, then service capability increases, but route complexity and computational requirements increase
Solution Approach 1:
The system segments the complex multi-service routing problem into distinct components: delivery locations, charging stations, battery swap stations, and service time windows. Each node type is handled with specific constraints and optimization criteria, making the overall complex problem manageable through modular processing.
Solution Approach 2:
The routing system dynamically adapts to different service requirements by allowing flexible node types and constraints. The planner can adjust route parameters in real-time based on changing conditions such as battery status, delivery priorities, and charging availability, transforming a static complex problem into a dynamic adaptable system.
3Loss of time
If battery swap stations are added to the route, then energy replenishment time is reduced, but infrastructure requirements and system complexity increase
Solution Approach 1:
The system introduces a routing planner as an intermediary that manages the complexity of coordinating with battery swap infrastructure. The planner translates vehicle energy needs into optimized swap station visits, handling the complexity of infrastructure coordination while presenting a simplified route to the driver.
Solution Approach 2:
The system performs preliminary routing that pre-identifies optimal battery swap locations and timing based on predicted energy consumption and station availability. This preliminary planning reduces actual swap time by eliminating decision delays during operation.
Data Source
AI summary
Method and system of electric vehicle route planning for multi-service delivery and on-route energy replenishment is disclosed. Last mile delivery is a critical component of supply chains that impacts both customer experience and delivery cost. User request is received by a depot of a delivery environment. The user request comprising a current location of the user, one or more required services, and a time window rendered between each of the required services is processed and a graph is generated for the user request. Further, trained learning agent generates a route map indicating a plurality of waypoint locations for the electric vehicle to visit each node in accordance with minimized trip cost of fleet and time duration where each node has a state action pair for the electric vehicle. The learning agent learns continuously during the interaction with delivery environment and obtains feedback for every associated action.