Multi-time Scaling Optimization for Electric Station Considering Uncertainties of Renewable Energy and EVs

The development of new energy vehicles, particularly electric vehicles (EVs) and hydrogen fuel cell vehicles (HFCVs), represents a strategic initiative to address climate change and foster sustainable development. Integrating PV with hydrogen production into hybrid electricity-hydrogen energy stations enhances land and energy efficiency but introduces scheduling challenges due to uncertainties. A multi-time scale scheduling framework, which includes day-ahead and intraday optimization, is established using fuzzy chance-constrained programming to minimize costs while considering the uncertainties of PV generation and charging/refueling demand. Correspondingly, trapezoidal membership function and triangular membership function are used for the fuzzy quantification of day-ahead and intraday predictions of photovoltaic power generation and load demands. The system achieves 29.37% lower carbon emissions and 17.73% reduced annualized costs compared to day-ahead-only scheduling. This is enabled by real-time tracking of PV/load fluctuations and optimized electrolyzer/fuel cell operations, maximizing renewable energy utilization. The proposed multi-time scale framework dynamically addresses short-term fluctuations in PV generation and load demand induced by weather variability and temporal dynamics. By characterizing PV/load uncertainties through fuzzy methods, it enables formulation of chance-constrained programming models for operational risk quantification. The confidence level – reflecting decision-makers’ reliability expectations – progressively increases with refined temporal resolution, balancing economic efficiency and operational reliability.