Coordinating temporal parameters to optimize MPC deployment: A bi-level multi-criteria evaluation of integrated building energy systems under weather uncertainty

Integrated building energy systems (IBES) feature complex coupling among multiple energy sources, storage units, and dynamic building processes, making their operational control challenging. Model predictive control (MPC) has emerged as an effective approach to achieve flexible and economical operation of such systems, yet its performance is highly sensitive to the coordination of temporal parameters, i.e., model time step, MPC step, and control horizon. However, these parameters are often selected empirically without systematic guidelines, leading to inconsistent and sub-optimal implementations. To bridge this gap, this study proposes a bi-level multi-criteria evaluation framework that jointly assesses the coordinated impacts of temporal parameters on both prediction accuracy (model level) and operational performance (MPC level), while explicitly accounting for weather uncertainty. A representative 5550 m2 office building integrating photovoltaics, heat pumps, a thermal tank, and electric vehicles serves as the case study. Through comprehensive testing of 40 temporal parameter configurations, results demonstrate that system performance depends more on the synergistic configuration of these parameters than on any individual one. The optimal configuration, i.e., both model and MPC steps set to 15 min and a control horizon of 1–4 steps, reduces the average computation time per control action from 243 s in 5 min case to 10 s, while lowering operating costs by approximately 3% and thermal comfort deviations by 74–90% compared to cases with longer time steps. These findings remain consistent across different weather uncertainty scenarios, providing theoretical and practical guidance for optimizing temporal parameter determination to support effective MPC deployment in IBES.