Adaptive Power Control Based on Neural Network Design and Implementation for the Hybrid Shipboard Microgrids

With the increase in carbon emissions, climate change, and reduction in fossil fuels, there is a need for energy storage systems and renewable energy sources for ships and ferries. Currently, seaports do not have shore-based cold ironing facilities in most ship ports. Therefore, electricity is provided by the main grid. Due to the lack of a cold ironing facility on the ship, expansive diesel generators must be equipped on the ship. Furthermore, the ship must be functional to provide supply to auxiliary loads on the ship. To provide a solution to these issues, we need to provide cold ironing facilities to every seaport. However, this solution will take time to implement, and it will be very costly. On the other hand, ship-based microgrid systems will enable clean and economical power generation facilities on the ship. The idea is to implement a low-cost hybrid microgrid system on the ship so that we receive alternate energy resources. We proposed a nonlinear droop control design for the improvement of voltage and frequency droops in the bidirectional power converter model. A comparison of conventional and proposed control designs has been presented in simulation results. Hard-ware implementation on the Jetson Nano system and practical results have been validated.