Development of an adaptive neuro-fuzzy inference system–based equivalent consumption minimisation strategy to improve fuel economy in hybrid electric vehicles

The most viable option to achieve the goals of saving energy and protecting the environment is to replace conventional vehicles with hybrid electric vehicles (HEVs). In HEVs, the operational characteristics of an internal combustion engine (ICE) and an electric motor (EM) are different from each other and thus require an adaptive control strategy to achieve higher fuel economy along with smooth operation and better performance of the vehicle. An energy management control strategy is proposed for an HEV based on an adaptive network-based fuzzy inference system (ANFIS). The proposed adaptive equivalent consumption minimisation strategy decides the power to be drawn from ICE and EM based on input parameters such as the speed of the vehicle, the state of charge of the battery, the EM torque and the ICE torque. The whole system is simulated in an advanced vehicle simulator tool. The proposed non-linear controller has also been tested for real-time behaviour using a field-programmable gate array–based MicroLabBox hardware controller to compare its performance against existing controllers. The authors compared the fuel economy obtained using the proposed method with several other methods available in the literature. The comparison clearly reveals that the proposed ANFIS-based method results in better optimization of energy and hence offers better fuel economy. The urban dynamometer driving schedule has been employed for this analysis.