Adjustable Robust Optimization for Hybrid Hydrogen-Battery Microgrids with LSTM-QR Forecaster

To address the challenges posed by renewable energy variability and carbon emissions in traditional fossil fuel-powered microgrids, this paper proposes an emission-free microgrid system integrated with hybrid hydrogen-battery energy storage. First, a Long Short-Term Memory with Quantile Regression (LSTM-QR) forecaster is employed to estimate the upper and lower bounds of renewable energy generation, which are then used to construct an uncertainty set. Next, an adjustable robust optimization model is developed to minimize the total operational cost while accounting for the uncertainty in renewable energy generation. The model’s robustness can be flexibly controlled by adjusting the time budget. Using linear decision rules, the model is reformulated as a mixed-integer linear programming problem, which can be efficiently solved by standard optimization solvers. Numerical simulations demonstrate the effectiveness of the model and solution approach.