Nearly optimal demand side management for energy, thermal, EV and storage loads: An Approximate Dynamic Programming approach for smarter buildings

This paper proposes a distributed feedback-based optimization method, based on the principles of approximate dynamic programming, aiming for the optimal management and energy efficient operation of grid connected buildings. Modern building management faces multiple challenges, aiming for, energy efficiency, RES integration, EV operation, cost reduction and of course user comfort. Therefore, this paper adopts a multi-criterion cost objective, where minimum energy costs are required without sacrificing user preferences and satisfaction. Different type of loads are presented, including both thermostatically controllable loads (simulated and modelled in Energy Plus), and controllable electric vehicles (EV) and energy storage systems (ESS). Extensive simulations showcase the effectiveness of the proposed method introducing comparisons with different strategies by exploiting renewable energy integration, varying pricing tariffs and the stored energy of EVs and ESS units. A robustness evaluation of the proposed method is presented, validating the performance under different conditions. Finally, the paper demonstrates that the proposed strategy can be utilized for the management of real-world problems and buildings. (c) 2021 Elsevier B.V. All rights reserved.

Automated summary

This paper proposes a distributed feedback-based optimization method for the optimal management and energy efficient operation of grid connected buildings. The method considers a multi-criterion cost objective to minimize energy costs while satisfying user preferences and satisfaction. Extensive simulations demonstrate the effectiveness of the proposed method and its robustness is evaluated.