Deep Learning Explicit Differentiable Predictive Control Laws for Buildings

About: We demonstrate the use of differentiable predictive control (DPC) methodology for learning constrained neural control laws for building thermal comfort control. Authors: Jan Drgona, Aaron Tuor, Soumya Vasisht, Elliott Skomski, Draguna Vrabie Venue: 7th IFAC Conference on Nonlinear Model Predictive Control 2021 paper: https://www.sciencedirect.com/science/article/pii/S2405896321012933 This research was partially supported by the Mathematics for Artificial Reasoning in Science (MARS) and Physics Informed Machine Learning (PIML) initiatives via the Laboratory Directed Research and Development (LDRD) investments at Pacific Northwest National Laboratory (PNNL). PNNL is a multi-program national laboratory operated for the U.S. Department of Energy (DOE) by Battelle Memorial Institute under Contract No. DE-AC05-76RL0-1830.