A Hierarchical Multi-Timeframe Multi-Energy Sharing Framework for a Self-Sustained Energy-Transportation Nexus

This article proposes a hierarchical multi-timeframe multi-energy sharing framework for the peer-to-peer hydrogen (H
₂
)-electricity trading among the integrated hydrogen-electricity energy stations (IHESs) scattered at the freeway traffic system. Specifically, the upper layer aims to increase the IHESs’ day-ahead incomes while determining the amount of shared H
₂
among IHESs on long timeframes, where the H
₂
sharing is modeled as a distributionally robust optimization (DRO) problem. The obtained shared amount of H
₂
from the upper layer is served as the optimal reference to be followed in the lower level. The objective of the lower layer is to maximize the real-time profits of IHESs with rolling horizon-based DRO model, where the flexible electricity sharing is coordinated and optimized on small timeframes to overcome the uncertain sources and slow dynamic characteristics of shared H
₂
transportation. A model-data driven forecasting approach containing the partial differential fluid dynamic model and long short-term memory (LSTM) neural network is formulated to estimate the diversified time-varying load demands at IHESs. Moreover, an ADMM-based distributed pricing strategy is developed to obtain the incentive compatible prices for multi-energy sharing, and a closed-form optimal price solution is derived to accelerate the convergence speed. Comparative studies have demonstrated the superior performance of the proposed methodology on the improvement of the self-sustained efficiency and economic benefits for the energy-transportation nexus.