EV Charging Scheduling Using Predicted Usage Patterns
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing charging infrastructure for electric vehicles lacks a smart scheduling method to efficiently distribute charging spots among multiple vehicles, leading to inefficiencies and long wait times.
Innovation Solution
A method and apparatus that utilize machine learning and data mapping to predict usage patterns based on historical data, adjusting schedules dynamically to optimize the distribution of charging spots and supply phases, incorporating various scheduling schemes like FCFS, round robin, and priority-based approaches.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If a simple first-come-first-serve scheduling method is used, then the system is easy to operate and implement, but the charging infrastructure efficiency and throughput are reduced due to long wait times and poor resource utilization
Solution Approach 1:
The patent implements dynamic scheduling by continuously monitoring real-time usage patterns and adjusting the distribution of charging spots accordingly. The system transitions from static first-come-first-serve to dynamic optimization based on predicted demand, vehicle types, and infrastructure capacity, thereby improving throughput while maintaining operational simplicity through automated decision-making
Solution Approach 2:
The system performs preliminary actions by predicting future usage patterns before actual charging demand occurs. By analyzing historical data and contextual factors (time, weather, events), the system pre-determines optimal scheduling strategies, allowing it to proactively allocate charging spots rather than reactively responding to queues, thus improving overall productivity
2Ease of operation
If charging spots are distributed equally among all electric vehicles, then fairness is improved, but the overall charging efficiency and utilization rate deteriorate due to ignoring varying vehicle needs and usage patterns
Solution Approach 1:
The patent applies local quality by customizing charging spot allocation based on specific vehicle characteristics and local conditions. Different vehicle types (passenger cars, buses, trucks), different charging needs, and different contextual factors receive differentiated treatment rather than uniform distribution, optimizing both fairness and efficiency through localized optimization strategies
Solution Approach 2:
The system dynamically changes allocation parameters based on multiple factors including vehicle type, predicted residence time, current queue length, and infrastructure capacity. By adjusting these parameters in real-time, the system achieves fair distribution that also maximizes utilization, moving from fixed equal allocation to flexible parameter-based allocation
3Quantity of substance
If the charging infrastructure serves a large number of electric vehicles simultaneously, then the coverage and service capability are improved, but the wait time and scheduling complexity increase significantly
Solution Approach 1:
The patent ensures continuous useful action by maintaining optimal charging spot utilization throughout operation. The predictive scheduling system continuously allocates spots to minimize idle time and reduce vehicle wait times, ensuring that the charging infrastructure operates at peak efficiency continuously rather than experiencing periodic bottlenecks or idle periods
Solution Approach 2:
The system implements feedback mechanisms by monitoring actual charging outcomes and comparing them with predicted usage patterns. This feedback loop allows the system to refine its predictions and adjust scheduling strategies in real-time, thereby managing large numbers of vehicles efficiently while minimizing wait times through data-driven optimization
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The present disclosure provides a new and improved method and apparatus of scheduling for a charging infrastructure serving a plurality of electric vehicles. In an exemplary embodiment, a computer-implemented method for scheduling a charging infrastructure serving a plurality of electric vehicles is provided, characterized in comprising: making a prediction for a usage pattern of the charging infrastructure with a context based on historical usage patterns of the charging infrastructure and the contexts of the historical usage patterns, and determining a schedule scheme for deciding a distribution of charging spots of the charging infrastructure among the electric vehicles based on the predicted usage pattern.