Optimal Control and Real-Time Simulation of Hybrid Marine Power Plants

Abstract

With significantly increasing concerns about greenhouse effects and sustainable economy, the marine industry presents great potential for reducing its environmental impact. Recent developments in power electronics and hybridisation technologies create new opportunities for innovative marine power plants which utilize both traditional diesel generators and energy storage like batteries and/or supercapacitors as the power sources. However, power management of such complex systems in order to achieve the best efficiency becomes one of the major challenges. Acknowledging this importance, this research aims to develop an optimal control strategy (OCS) for hybrid marine power plants. First, architecture of the researched marine power plant is briefly discussed and a simple plant model is presented. The generator can be used to charge the batteries when the ship works with low power demands. Conversely, this battery energy can be used as an additional power source to drive the propulsion or assist the generators when necessary. In addition, energy losses through braking can be recuperated and stored in the battery for later use. Second, the OCS is developed based on equivalent fuel consumption minimisation (EFCM) approach to manage efficiently the power flow between the power sources. This helps the generators to work at the optimal operating conditions, conserving fuel and lowering emissions. In principle, the EFCM is based on the simple concept that discharging the battery at present is equivalent to a fuel burn in the future and vice-versa and, is suitable for real-time implementation. However, instantaneously regulating the power sources’ demands could affect the system stability as well as the lifetime of the components. To overcome this drawback and to achieve smooth energy management, the OCS is designed with a number of penalty factors by considering carefully the system states, such as generators’ fuel consumption and dynamics (stop/start and cranking behaviour), battery state of charge and power demands. Moreover, adaptive energy conversion factors are designed using artificial intelligence and integrated in the OCS design to improve the management performance. The system therefore is capable of operating in the highest fuel economy zone and without sacrificing the overall performance. Furthermore, a real-time simulation platform has been developed for the future investigation of the control logic. The effectiveness of the proposed OCS is then verified through numerical simulations with a number of test cases.