Battery Lifetime Estimation Using SoC and Temperature Aging Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for determining the remaining lifetime of energy storage units, such as batteries and capacitors, lack accuracy due to variability in operating conditions and assumptions about constant conditions, leading to unreliable predictions.
Innovation Solution
A method that estimates the state of charge (SoC) and present energy storage parameter values, combined with temperature, to derive an aging model that predicts the time evolution of these parameters, allowing for precise determination of the remaining useful life by extrapolating when a predetermined threshold is reached.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If the Miner rule is used to estimate SoH based on linear accumulation of stress and fatigue damages, then the estimation process is simple, but the remaining lifetime prediction accuracy is insufficient because it does not account for varying operating conditions
Solution Approach 1:
The patent transforms the estimation approach from using only cycle-based damage accumulation (Miner rule) to incorporating multiple dynamic parameters including temperature, state of charge, and calendar time. This multi-parameter model captures the actual aging mechanisms more accurately, resolving the contradiction between simplicity and accuracy by adding necessary complexity only where it impacts prediction precision.
Solution Approach 2:
The patent transitions from a static SoH estimation method to a dynamic model that continuously updates remaining lifetime predictions based on real-time operating conditions. The model adapts to varying temperature, charge states, and usage patterns, making the estimation process dynamic rather than static, thereby improving accuracy without requiring overly complex infrastructure.
2Loss of time
If a reference lifetime value is used for prediction, then the initial estimation is quick, but the accuracy deteriorates due to significant deviations caused by fluctuating operating conditions
Solution Approach 1:
The patent performs preliminary calibration by determining actual initial capacity and aging rates during early operation phases or factory testing. This preliminary data collection enables the model to establish baseline parameters that are specific to each energy storage unit, allowing for quick yet accurate predictions even under varying operating conditions, thus resolving the time-accuracy tradeoff.
Solution Approach 2:
The patent implements a feedback mechanism where actual performance data from operation is continuously fed back into the aging model to refine predictions. The system monitors real-time temperature, charge-discharge cycles, and capacity degradation, adjusting the remaining lifetime estimation dynamically. This feedback loop maintains high accuracy without requiring excessive computation time by using incremental updates rather than complete recalculations.
Data Source
Figure 1
Figure 2
Figure 3A~3B
AI summary
The present disclosure relates to a computer-implemented method for determining a remaining lifetime of an energy storage unit, the method comprising: acquiring a temperature of the energy storage unit, an electrical current supplied from or to the energy storage unit and a terminal voltage of the energy storage unit; estimating a state of charge, SoC, and a present value of an energy storage parameter of the energy storage unit based on the temperature, the electrical current and the terminal voltage; deriving an aging model of the energy storage unit based on the SoC and the temperature, wherein the aging model is indicative of a predicted time evolution of the energy storage parameter of the energy storage unit; and determining the remaining lifetime based on the derived aging model, the present value of the energy storage parameter and an initial value of the energy storage parameter.