Neural networks based model and voltage control for lithium polymer batteries

The voltage of a lithium battery is a nonlinear function of its current rate, temperature, chemistry and history, and hence cannot easily be determined. In this study, we use a one-layer, feed-forward, artificial neural network (ANN), trained using the back-propagation algorithm, to model the behavior of a lithium-polymer (Li-Po) battery. Then, possible improvements to the neural network model using a multilevel approach are discussed. A neural controller was then developed on the basis of this model to control the battery voltage. Experimental and simulation results confirmed the accuracy of the model and the performance of the control technique.