Online Scheduling Strategy for Microgrids: An Approximate Dynamic Programming Approach

With the increasing integration of renewable energy, microgrids face significant scheduling challenges due to uncertainties in generation, demand, and electricity prices. Traditional optimization methods are inadequate for real-time decision-making, while dynamic programming suffers from the curse of dimensionality. This paper proposes an online scheduling strategy based on approximate dynamic programming (ADP), where the microgrid is modeled as a Markov decision process and the value function is approximated using a piecewise linear structure (PLF-VFA) to reduce computational complexity. The strategy combines offline training with online execution to achieve efficient and adaptive decision-making under uncertainty. Case studies verify that the proposed method balances accuracy and speed, making it suitable for real-time microgrid optimization.