EV Charging Schedule Selection Using Multi-Objective Evolutionary Algorithm
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The integration of electric vehicles into the power grid poses challenges due to uncoordinated charging, which can lead to significant stress on distribution grids in terms of power losses and power quality, as existing technologies fail to effectively manage multiple conflicting objectives such as maximizing revenue, minimizing customer costs, and optimizing grid usage.
Innovation Solution
A computer-implemented method and apparatus that determine multiple electric charging schedules based on weighted charging objectives, using a multi-objective evolutionary algorithm to select an optimized schedule that balances various objectives like revenue, customer cost, battery degradation, and grid constraints, allowing for dynamic adjustment of weights to prioritize specific objectives.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If uncoordinated charging of electric vehicles is implemented, then charging simplicity and user convenience are improved, but power losses and power quality on the distribution grid deteriorate
Solution Approach 1:
The system implements feedback mechanisms by continuously monitoring grid conditions (power losses, power quality) and adjusting charging schedules accordingly. The multi-objective evolutionary algorithm uses real-time grid state information to optimize charging timing and power levels, creating a closed-loop control system that balances user convenience with grid efficiency.
Solution Approach 2:
The charging schedule is made dynamic rather than static. The system adapts charging parameters (timing, power level) based on real-time grid conditions and evolving objectives. The evolutionary algorithm continuously refines charging schedules as grid conditions change, allowing the system to respond dynamically to varying power loss conditions and power quality requirements.
2Ease of operation
If uncoordinated charging of electric vehicles is implemented, then charging simplicity is improved, but power quality on the distribution grid deteriorates
Solution Approach 1:
The system monitors power quality parameters (voltage stability, harmonic distortion) in real-time and uses this feedback to adjust charging schedules. The multi-objective evolutionary algorithm incorporates power quality constraints and objectives, automatically modifying charging patterns to maintain acceptable power quality levels while still providing convenient charging services.
Solution Approach 2:
The charging management system acts as an intermediary between electric vehicles and the distribution grid. It mediates the charging process by coordinating charging timing and power levels, preventing direct uncoordinated connections that would harm power quality. The system translates user charging requests into grid-friendly schedules that maintain power quality standards.
3Productivity
If multiple charging objectives with different weights are optimized, then charging service provider revenue and grid efficiency are improved, but system complexity increases
Solution Approach 1:
The system uses parameter changes in the form of weighted objectives within the evolutionary algorithm. Different charging objectives (revenue maximization, power loss reduction, power quality improvement) are represented as weighted parameters that can be adjusted to reflect changing priorities. This allows the system to handle multiple objectives through parameter optimization rather than complex structural changes.
Solution Approach 2:
The patent replaces complex mechanical or manual charging coordination systems with a computational intelligence approach. The multi-objective evolutionary algorithm uses software-based optimization instead of hardware-based control mechanisms, reducing physical system complexity while achieving sophisticated multi-objective optimization. The algorithm substitutes complex control logic with evolutionary computation that automatically finds optimal charging schedules.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The present description refers to a computer implemented method, computer program product, and computer system to determine a plurality of electric charging schedules for one or more electric vehicles, determine, for each of the electric charging schedules, a plurality of charging objective values, assign a weight to each of the charging objective values, wherein one set of weights is used for the plurality of electric charging schedules, calculate a fitness value for each of the electric charging schedules based on the plurality of charging objective values for each respective electric charging schedule and the set of weights, identify the electric charging schedule having a highest fitness value, and select the electric charging schedule having the highest fitness value for charging one or more electric vehicles.