Multilevel neural-network model for supercapacitor module in automotive applications

Due to their complex physicochemical characteristics, it is difficult to build a comprehensive model for supercapacitor taking into account dependency on device current rate, voltage, temperature and chemistry. In this study, we use a one-layer feed-forward artificial neural network, trained using the back-propagation algorithm, to model the behavior of supercapacitors module used in automotive applications. Possible improvements of the neural network model using a multilevel approach are discussed. Experimental and simulation results confirmed the accuracy of the model.