Real-time schedule of integrated heat and power system: A multi-dimensional stochastic approximate dynamic programming approach

Abstract This paper proposes a multi-dimensional approximate dynamic programming (ADP) algorithm for the real-time schedule of integrated heat and power system (IHPS) with battery and heat storage tank (HST). The multi-time period optimization problem is reformulated under the Markov Decision Process. The high dimensional state variables are aggregated into the state of charge (SOC) of battery and overall available heat (OAH) of HST to reduce the computation of value function approximation (VFA) while ensuring approximate accuracy. Under sufficient training by uncertainty scenarios of wind power, electricity price, electrical and heat load, the approximate value function (AVF) can derive empirical knowledge and help IHPS make decisions to cope with uncertainties. The proposed ADP algorithm can efficiently take advantage of multi-energy integration and provide a near-optimal operation strategy to ensure the economy of IHPS by recursively solving the Bellman’s equation. Simulation results compared with existing methods validate the superiority of the proposed algorithm.