Battery Temperature Estimation Using Extended Kalman Filter
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for estimating the temperature of a battery system in vehicles are not accurate, leading to poorer vehicle drivability, increased energy usage, and rapid battery capacity degradation, and are costly due to the use of multiple temperature sensors.
Innovation Solution
The use of an extended Kalman filter (EKF) to estimate battery system temperature, which reduces the need for multiple sensors by utilizing a series of temperature measurements over time, operating recursively with new input measurements and a process model to provide accurate, real-time temperature estimates.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple temperature sensors are used to estimate battery system temperature, then measurement coverage is improved, but manufacturing cost and device complexity increase
Solution Approach 1:
The patent introduces an intermediary computational model (thermal model with state variables) that mediates between limited sensor measurements and the desired comprehensive temperature estimation. The model acts as a bridge, using a small number of actual sensor readings combined with thermal physics principles to infer temperatures throughout the battery system without requiring dense sensor placement.
Solution Approach 2:
The patent replaces the mechanical/physical system of multiple temperature sensors with a computational system (algorithmic temperature estimation). Instead of physically measuring temperature at multiple points with hardware sensors, the system uses computational models and algorithms to estimate temperatures based on limited measurements and thermal physics principles.
2Measurement precision
If multiple temperature sensors are deployed throughout the battery system, then temperature monitoring accuracy improves, but initial production costs and ongoing repair costs increase
Solution Approach 1:
The patent employs inexpensive temperature sensors (thermistors) rather than expensive alternative sensors, accepting that these simple sensors may have limited longevity or precision but compensating through the computational model that fuses multiple data sources and uses thermal physics to maintain accurate temperature estimation over time.
3Device complexity
If conventional temperature estimation methods are used, then device simplicity is maintained, but battery performance and drivability deteriorate due to inaccurate temperature data
Solution Approach 1:
The patent implements feedback mechanisms where the computational model continuously refines temperature estimates by comparing model predictions with actual sensor measurements and adjusting state variables accordingly. This feedback loop ensures that temperature estimates remain accurate and reliable for battery control decisions.
Solution Approach 2:
The computational temperature estimation system serves multiple functions simultaneously: it provides temperature monitoring, predicts thermal behavior, supports battery management decisions, and enables diagnostic capabilities. This multi-functionality achieves high reliability without proportionally increasing device complexity.
Data Source
AI summary
System and methods for estimating a temperature of a battery are presented. In some embodiments, a method of estimating a temperature of a battery system may utilize measured battery system temperature data and measured ambient temperature data. Based on the measured temperature data, an average estimated temperature of the battery system may be determined using, at least in part, an extended Kalman filter and an energy balance process model associated with the battery system.