EV Charging Demand Prediction Using Reachable Range Modeling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies fail to predict demand for charging spots at newly installed locations, as they do not account for the number of electric vehicles and their reachable ranges.
Innovation Solution
An information processing apparatus that acquires information about arbitrary points or areas, generates a reachable range for electric vehicles based on power, and predicts charging spot demand using the number of electric vehicles and generated ranges.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If existing technology is used to predict charging timing and position, then charging state information can be predicted, but demand for charging spots at arbitrary locations cannot be predicted
Solution Approach 1:
The system segments the prediction function into two independent components: (1) prediction of charging demand at arbitrary locations using reachable range and vehicle number data, and (2) prediction of charging state information using existing position data. This segmentation allows the new location-based prediction module to be added without disrupting the existing charging state prediction system, thereby improving both prediction accuracy and adaptability to new locations simultaneously.
2Loss of information
If charging spot demand is predicted only based on existing position information, then charging timing can be predicted, but information for newly installed charging spots is unavailable
Solution Approach 1:
The system performs preliminary action by pre-calculating and storing reachable range data for multiple arbitrary locations before actual charging demand occurs. By pre-establishing the relationship between vehicle numbers, power consumption, and reachable ranges at various locations, the system ensures that when new charging spots are installed, the prediction model can immediately provide accurate demand forecasts without requiring additional data collection or model retraining, thus eliminating information loss and expanding prediction coverage.
3Device complexity
If reachable range is not considered in prediction, then calculation is simpler, but power consumption influence on demand is ignored
Solution Approach 1:
The system applies local quality by making the reachable range parameter location-specific and power-specific. Instead of using a uniform prediction model, the system calculates reachable ranges individually for each arbitrary location based on local power consumption characteristics and vehicle numbers. This localized approach ensures that the prediction accuracy reflects actual local conditions while keeping the overall system structure modular and manageable, balancing calculation complexity with prediction accuracy.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
An information processing apparatus (100) includes an acquisition unit (121) that acquires information regarding an arbitrary point or an arbitrary area and information regarding the number of electric vehicles, a generation unit (122) that generates a reachable range of the electric vehicle with predetermined power based on the information regarding the arbitrary point or the arbitrary area acquired by the acquisition unit (121); and a prediction unit (123) that predicts a demand for a charging spot at the arbitrary point or in the arbitrary area based on the information regarding the number of electric vehicles acquired by the acquisition unit (121) and the reachable range generated by the generation unit (122).