Data-driven robust coordination of generation and demand-side in photovoltaic integrated all-electric ship microgrids

Abstract

Fully electrified ships, which is known as the 'all-electric ships (AESs)', have the potentials to bring great economic /environmental benefits. To further improve the energy efficiency of AESs, PV generations are gradually integrated, which introduces uncertainties to the AES operation. However, current researches mostly focus on sizing problem whereas rarely concern the operation. In this perspective, a data-driven robust coordination of generation and demand-side is proposed to properly address the onboard PV generation uncertainties as well as reducing the fuel cost of AESs, which consists of an extreme learning machine (ELM) based PV uncertainty forecasting method and a two-stage operating framework, where the first stage for the worst PV generation case and the second stage targets at the uncertainty realization. A 4-DG AES is implemented into the case study and the simulation results show that the ELM-based method can well characterize the PV uncertainties, and the two-stage operating framework can well accommodate the onboard PV uncertainties. Further analysis also demonstrates the proposed method has enough flexibility when facing working condition variations.