EV Charging Recommendations Using Traffic Prediction and Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Electric vehicles face challenges such as limited accessibility of charging points, variability in driving ranges due to real-time factors like weather and driving behavior, and range anxiety caused by continuous battery drain and uncertainty in actual range, leading to inefficient charging strategies.
Innovation Solution
An apparatus and method using machine learning models to predict traffic congestion and optimize charging strategies by generating recommendations based on integer or linear programming, considering constraints like travel time, charging point availability, and user inputs, to determine the most efficient time and location for charging.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If EVs are charged periodically to maintain driving range, then driving performance is improved, but accessibility of charging points is limited due to high demand during peak hours
Solution Approach 1:
The system performs preliminary actions by predicting traffic congestion status before the user needs to charge, and recommends charging during off-peak hours when charging points are more accessible. This advance planning allows users to charge when charging points are available, avoiding the high-demand peak hours problem.
Solution Approach 2:
The system incorporates feedback loops by continuously monitoring traffic congestion information and charging point availability, then adjusting charging recommendations accordingly. This feedback mechanism enables the system to adapt to changing conditions and guide users to optimal charging times when accessibility is improved.
2Productivity
If charging strategies are optimized using machine learning and integer programming, then charging efficiency is improved, but device complexity increases
Solution Approach 1:
The patent employs an intermediary approach by using machine learning models and optimization algorithms as mediators between raw traffic data and charging recommendations. These intermediaries process complex computations internally while presenting simplified, actionable recommendations to users, thus achieving high charging efficiency without exposing the user to system complexity.
Solution Approach 2:
The system performs self-service by automatically executing complex integer programming and linear programming calculations to generate optimal charging strategies. The computational complexity is handled autonomously by the system's algorithms, eliminating the need for users to understand or manage the underlying complexity while still benefiting from optimized charging efficiency.
3Measurement precision
If real-time factors like weather and driving behavior are considered, then accuracy of driving range prediction is improved, but difficulty of detecting and measuring increases
Solution Approach 1:
The system applies a universal approach by using a single machine learning model that simultaneously processes multiple real-time factors including weather conditions, driving behavior patterns, traffic congestion, and battery characteristics. This multi-functional model achieves accurate driving range predictions while consolidating the complexity of monitoring multiple parameters into a unified system.
Data Source
AI summary
An apparatus, a method, and a non-transitory computer-readable storage medium for providing strategies for charging vehicles is provided. For example, the apparatus obtains, using a map database, traffic congestion information on a road segment, predicts a traffic congestion status on the road segment based on the traffic congestion information, generates an objective function based on the traffic congestion status, computes a solution of the objective function using an integer programming or a linear programming, generates a recommendation based on the solution, and outputs the recommendation.