EV Battery Health Prediction Using Full-Charge Driving Range
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing battery health monitoring systems for electric vehicles rely on complex metrics like battery capacity in ampere hour (Ah), which are difficult for average customers to understand and are influenced by various environmental, mechanical, and driver behavioral factors, making it challenging to accurately predict battery health.
Innovation Solution
A method and system using a machine learning model that factors out these influences by creating a distance-driven model based on historical telematics data, including odometer readings, battery state-of-charge, vehicle speed, and battery-module temperatures, to predict battery health, allowing for real-time monitoring and alerting when health thresholds are met.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If battery capacity in ampere hour (Ah) is used as the monitoring metric, then the measurement precision of battery health is improved, but the ease of operation for average customers deteriorates
Solution Approach 1:
The patent creates a virtual copy of the battery's operational capacity by modeling the distance driven on a full battery load based on historical telematics data. This modeled distance serves as an interpretable proxy that maintains the precision of technical measurement while being easily understood by customers, effectively copying the essence of battery health in human-friendly terms.
2Ease of operation
If distance driven on a full battery load is used as the monitoring metric, then the ease of operation for customers is improved, but the reliability of battery health prediction deteriorates due to multiple influencing factors
Solution Approach 1:
The patent extracts and isolates the battery capacity component from the complex distance driven metric by using machine learning to factor out environmental, mechanical, and driver behavioral influences. This extraction process separates the reliable battery health signal from the noisy operational variables, maintaining prediction reliability while preserving the interpretability of distance-driven metrics.
Solution Approach 2:
The patent introduces a machine learning model as an intermediary between the raw telematics data and the battery health assessment. This intermediary processes the complex relationships between multiple variables and produces a refined distance metric that reliably reflects battery health while remaining easy for customers to understand and interpret.
3Reliability
If a machine learning model is introduced to factor out influencing factors, then the reliability of battery health prediction is improved, but the device complexity increases
Solution Approach 1:
The patent implements a universal machine learning model that serves multiple functions: it processes historical telematics data, identifies trips on battery, extracts relevant features, models the distance driven relationship, and predicts battery health. This multi-functional approach consolidates complexity into a single versatile system rather than requiring separate specialized components for each function.
Data Source
AI summary
A method and a system for predicting battery health based on distance driven on a full battery load with machine learning model are provided. The method includes: obtaining historical vehicle telematics of vehicles, wherein the historical vehicle telematics comprise at least one of the following: odometer readings, battery SOC, vehicle speed, battery-module temperatures, and battery-cell voltages; creating a distance driven model according to a relationship between a distance driven on a full battery load of vehicles and the historical vehicle telematics; obtaining a distance driven on a full battery load of a vehicle based on the distance driven model by using real-time vehicle telematics of the vehicle as model input; predicting battery health of the vehicle by comparing the obtained distance with a reference distance value.