文章基本信息

标题：Deep Learning Explicit Differentiable Predictive Control Laws for Buildings
本地全文：下载
作者：Ján Drgoňa ; Aaron Tuor ; Elliott Skomski 等
期刊名称：IFAC PapersOnLine
印刷版ISSN：2405-8963
出版年度：2021
卷号：54
期号：6
页码：14-19
DOI：10.1016/j.ifacol.2021.08.518
语种：English
出版社：Elsevier
摘要：AbstractWe present a differentiable predictive control (DPC) methodology for learning constrained control laws for unknown nonlinear systems. DPC poses an approximate solution to multiparametric programming problems emerging from explicit nonlinear model predictive control (MPC). Contrary to approximate MPC, DPC does not require supervision by an expert controller. Instead, a system dynamics model is learned from the observed system’s dynamics, and the neural control law is optimized offline by leveraging the differentiable closed-loop system model. The combination of a differentiable closed-loop system and penalty methods for constraint handling of system outputs and inputs allows us to optimize the control law’s parameters directly by backpropagating economic MPC loss through the learned system model. The control performance of the proposed DPC method is demonstrated in simulation using learned model of multi-zone building thermal dynamics.
关键词：Keywordsconstrained deep learningneural state space modelssystem identificationdifferentiable predictive controlexplicit nonlinear MPC