Multi-Stage Robust Scheduling for Microgrids with Energy Storage and Renewable Energy

The increasing integration of renewable energy sources (RES) in microgrids introduces significant uncertainty, posing challenges to operational stability and economic efficiency. Traditional two-stage robust optimization (TSRO) methods typically assume that uncertainties can be simultaneously observed and utilized, which may render scheduling solutions infeasible in real-time operation. To address this issue, we propose a multi-stage robust optimization (MSRO) model that explicitly enforces the non-anticipativity requirement, ensuring that scheduling decisions at each stage rely only on available uncertainty observations. The proposed model incorporates boundary variables to regulate energy storage system (ESS) operations, enhancing flexibility while maintaining computational tractability. Numerical experiments on a microgrid system demonstrate that MSRO achieves comparable day-ahead scheduling costs to TSRO while reducing intra-day operational costs by 3.66%. Furthermore, MSRO ensures more stable ESS operation and power exchange strategies, mitigating short-sighted scheduling decisions and improving system robustness. These findings highlight the potential of MSRO for practical microgrid scheduling applications.