A three-layer energy management system for hydrogen-powered ships combined instantaneous load forecasting

Abstract

This paper proposes a three-layer energy management system (EMS) for hydrogen-powered ships (HPS) that combines instantaneous load forecasting. The EMS includes a one-layer instantaneous load forecasting subsystem (ILFS) and a two-layer optimization and scheduling subsystem (OSS). The ILFS is composed of variational mode decomposition (VMD) and an improved autoregressive integrated moving average model based on Itô process (IP-ARIMA). First, considering high volatility and randomness of ship load, VMD is used to decompose ship load into multiple relatively stable intrinsic mode functions (IMFs), and an IP-ARIMA forecasting model is established for each IMF. Then, the forecasted sequences of each IP-ARIMA are superimposed to obtain the final forecasted load sequence (FLS). The upper and lower layers of the OSS use model predictive control and receive the FLS from ILFS. After the upper layer optimizes the efficiency of the hydrogen fuel cell (HFC) and state of charge (SOC) of the storage battery (SB), the lower layer minimizes the operating and degradation costs of the HFC. Finally, the HFC and SB perform energy scheduling according to the instruction signals output by the EMS. Experimental results show that, compared to other methods, the proposed EMS significantly enhances both economy and durability of HPS operation.