A GRNN informed ECMS-ALPF operational management strategy for reducing fuel consumption of diesel/battery/supercapacitor hybrid vehicles

Abstract

The fuel consumption and pollution emission of ships have become the main focuses in ocean environment management. This paper proposes a new marine vehicular hybrid propulsion system (VHPS), which is composed of a diesel generator and a hybrid energy storage system consisted of batteries and supercapacitors. However, the distinct power and energy characteristics of the three power sources result in complex operational management and a negative impact on energy conservation and emission reduction. Therefore, a general regression neural network informed equivalent consumption minimization strategy and adaptive low pass filter (GRNN informed ECMS-ALPF) operational management strategy (OMS) is proposed to optimize the fuel consumption and energy loss. Finally, a hardware in the loop experimental platform is utilized to verify the performance of proposed OMS. From experimental results, the GRNN informed ECMS-ALPF strategy has accurate load predictive capability, the predictive average root mean square error value is only 2.1633 kW, whose fuel consumption is 14.9% lower compared with a conventional rule-based strategy. And the energy loss of the given hybrid energy storage system can reduce 0.0178 kWh (about 14.55%) compared with a conventional low pass filter strategy. The proposed VHPS and OMS offer insights for the study of ocean and coastal management.