An artificial neural network module for simulation of the energy storage system of a hybrid electric vehicle

A hybrid electric vehicle (HEV) is a complex system integrating interactive subsystems of disparate degrees of complexity. The simulation of an HEV th us poses a challenge. An accurate simulation requires highly accurate model s of each subsystem. Without these, the system has a poor overall performan ce. Typically, modelling problems are not amenable to physical solutions wi thout simplifying assumptions that impair their accuracy. Conventional empi rical models, on the other hand, are time consuming and data intensive and falter where extensive non-linearity is encountered. An artificial neural n etwork (ANN) approach to simulation of an HEV is presented in this paper. A n ANN model of the energy storage system (ESS) of an HEV was deployed in th e ADVISOR simulation software developed by the National Renewable Energy La boratories (NREL) of the US Department of Energy. The ANN model mapped the state of charge (SOC) and the power requirement of the vehicle to the volta ge and current at the ESS output. An ANN model was able accurately to captu re the complex, non-linear phenomena underlying the ESS. A novel performance-enhancing technique for design of ANN training data, Sm art Select, is described here. It resulted in a model of 0.9978 correlation (R-2 error) with data. ANNs can be data hungry. The issue of knowledge sha ring between ANN models to save development time and effort is also address ed in this paper. The model transfer technique presents a way of levering t he expertise of one ANN into the development of another for a similar model ling task. Lastly, integration of the ANN model of the ESS into the ADVISOR software, on the MATLAB software platform, is described.