Demand-Aware EV Fleet Routing and Power Procurement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for optimizing power procurement for enterprises with electric vehicle fleets fail to consider the mutual dependency between source selection, fleet routing, and demand profile shaping, particularly due to the influence of external charging points and battery storage, leading to suboptimal electricity cost management.
Innovation Solution
A method and system that iteratively solve routing and procurement sub-problems using interior point techniques to optimize EV routes and electricity procurement, considering constraints and market price risks, reducing the problem complexity from MINLP to MIQP.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If existing methods for optimizing EV routing and charging are used, then the cost of charging and travel distance can be minimized, but the mutual dependency between source selection, fleet routing, and demand profile shaping is not considered, leading to suboptimal power procurement
Solution Approach 1:
The patent divides the complex power procurement optimization problem into two separate sub-problems: (1) EV routing and charging optimization, and (2) power source selection and procurement optimization. This segmentation allows each sub-problem to be solved independently using appropriate methods, reducing overall complexity while still achieving near-optimal results by iteratively refining solutions between the two sub-problems.
Solution Approach 2:
The patent introduces an intermediary iterative process that connects the EV routing sub-problem and the power procurement sub-problem. The demand profile generated by EV routing serves as input to power procurement optimization, and the resulting procurement costs feed back to refine routing decisions. This intermediary mechanism handles the mutual dependency between the two sub-problems without requiring a single complex optimization model.
2Device complexity
If the cost of electricity is assumed to be independent of EV charging demand, then the optimization problem becomes simpler, but this assumption does not hold for enterprises with EV fleet charging load where demand influences pricing
Solution Approach 1:
The patent makes the electricity cost dynamic by incorporating demand-dependent pricing into the power procurement optimization. Instead of assuming fixed electricity rates, the model considers how EV charging demand affects procurement costs from different sources. This dynamic approach allows the system to adapt routing and charging decisions to varying cost conditions, achieving better cost optimization despite increased model complexity.
3Adaptability or versatility
If external charging points are utilized to offload EV charging demand, then enterprise power procurement complexity increases, but this offloading fraction is determined by multiple factors including availability, constraints, and cost relationships
Solution Approach 1:
The patent segments the charging infrastructure into enterprise-controlled charging points and external third-party charging points. This segmentation allows the system to independently optimize the fraction of charging demand allocated to each type based on their respective characteristics, constraints, and cost structures. The EV routing sub-problem handles both charging sources simultaneously, while the power procurement sub-problem focuses on optimizing enterprise power purchases based on the resulting demand profile.
Data Source
Figure 1~2
Figure 3A~3B
Figure 4A
AI summary
Accurate estimation of average procurement cost becomes increasingly pivotal to ensure optimal procurement of electricity for an enterprise owning EV fleets. Existing techniques available for joint optimization of EV routing and charging are not usable for enterprise with EV fleet charging loads as cost of electricity is assumed to be independent of the EV charging demand. Further, they require total demand to be known a priori. Present disclosure provides a method and a system for optimizing power procurement for enterprises with electric vehicle fleet charging load. The system first performs a route optimization by modelling set of routing constraints which are then used to obtain optimal EV routes. Thereafter, system performs a procurement cost optimization by modelling set of procurement constraints using the plurality of inputs and the optimal EV routes to obtain an allocation information of each external energy source which is then utilized to obtain final procurement cost.