How to Analyze a Lithium Ion Battery Voltage Curve
When a battery is charged or discharged the terminal voltage increases or decreases very slightly. This is primarily due to the internal impedance of the cell. In order to reduce this, it is often desirable to incorporate a voltage regulator in the system. However, this robs power from the cell. Most cell chemistries also exhibit a steep slope in the discharge curve which robs more power from the cell and can result in the loss of active chemicals within the cell.
Lithium ion batteries in particular can be very sensitive to detecting full charge. They have much tighter voltage tolerance when compared to lead acid batteries and do not like to be trickle or float charged. This can cause the cell to overheat which not only shortens their life but in extreme cases can lead to them catching fire or even exploding.
To address this issue a calibration charge indicator can be used which uses a constant current method to measure the battery's capacity and states of charge. This can be combined with a modelled load curve to estimate the state of charge very precisely. This is an important component of a battery management system which can be integrated into the vehicle control unit to provide a real-time view of the battery status.
Previously, the analysis of battery charging voltage curves and IC curves (capacity V/V) has been carried out using point counting methods or polynomial curve fitting methods. In this article a new method is proposed, based on a neural network model, that can be used to fit and analyze the charging voltage curve. In addition, the trained neural network model can be used to analyze the battery capacity relative to the phase change reactions that occur inside the battery.
