EV Charging Rate Prediction via Historical Pattern Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for predicting the charging rate of plug-in electric vehicles at charging stations are not accurate due to variations in actual charging rates from advertised rates, influenced by factors like time, day, ambient temperature, and vehicle conditions, leading to uncertainties in charging time estimates.
Innovation Solution
The method learns from historical charging patterns at the station, considering conditions such as time, day, and conditions under which they were observed, using numerical analysis, statistical approximation, or machine learning to predict the charging rate based on the vehicle's current state and station conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If historical charging patterns are analyzed using machine learning to predict charging rates, then prediction accuracy is improved, but computational complexity and data processing requirements increase
Solution Approach 1:
The system performs preliminary actions by collecting and storing historical charging pattern data in advance, tagging each pattern with time, day, and environmental conditions. This pre-processing of data enables the machine learning model to make accurate predictions without requiring complex real-time computations during actual charging events.
Solution Approach 2:
The patent introduces a machine learning model as an intermediary between raw historical charging data and prediction outputs. This intermediary processes the complex relationships between multiple variables (time, temperature, charge level, battery temperature) and transforms them into accurate charging rate predictions, managing the complexity through specialized computational processing.
2Reliability
If multiple vehicle conditions and environmental factors are considered in prediction, then prediction reliability is improved, but the number of parameters to be monitored and processed increases
Solution Approach 1:
The system segments the prediction process by determining vehicle conditions (maximum charging rate, current charge level, battery temperature) and environmental conditions (ambient temperature, time, day) separately. Each condition is independently monitored and then integrated into the overall prediction model, managing complexity through structured segmentation of multiple parameters.
Solution Approach 2:
The machine learning model utilizes feedback from historical charging patterns tagged with specific conditions to continuously improve prediction reliability. By comparing predicted versus actual charging rates from past events under similar conditions, the system refines its predictions for multiple parameters, enhancing reliability through iterative learning.
Data Source
Figure 1
Figure 2~3
AI summary
The invention provides a method of predicting a charging rate at a charging station for a plug-in electric vehicle characterized in capturing historical charging patterns (21, 22, 23) observed at the charging station, determining conditions of the vehicle at time of charge, and estimating the probable charging rate at the time of charge based on the charging rate advertised by the charging station and the historical charging patterns (21, 22, 23) relating to the determined conditions. The invention further provides a corresponding data processing apparatus, a corresponding computer program, and a corresponding medium.