Colloquium "Stochastic Model Predictive Control in Building HVAC systems"

Abstract:

We consider the problem of predictive control for linear systems subject to stochastic disturbances and its application to building heating, ventilation, and air conditioning (HVAC) systems. We repeatedly solve a stochastic finite-time constrained optimal control problem by using the scenario-based approach. We address the conservatism of the approach by presenting a new technique for fast scenario removal based on mixed-integer quadratic optimization. Probabilistic bounds are derived which quantify the benefits of the proposed technique.  In order to apply the proposed methodology the building HVAC system is modeled as a network of thermal zones controlled by a central air handling unit and local variable air volume boxes, with each zone modeled with simplified nonlinear models. The thermal load in each thermal zone is a function of occupancy and weather conditions. We present SMPC design that minimizes expected energy consumption and bounds the probability of thermal comfort violations. The SMPC uses predictive knowledge of the stochastic load in each zone during the design stage. A special attention is put on a trade-off between computational tractability and conservatism of the resulting SMPC scheme. The proposed SMPC scheme is compared with alternative SMPC designs, and the effectiveness of the proposed approach is demonstrated by simulation tests.

Slides: across_coll_Matusko13

Example

CV:

Jadranko Matuško received his BSc, M.Sc and PhD degree, from Faculty of Electrical Engineering and Computing, University of Zagreb, in 1999, 2003 and 2008, respectively. From October 2008 he has been working as an assistant professor at the Department of Electrical Machines, Drives and Automation, Faculty of Electrical Engineering and Computing, Zagreb. Period from July 2011 to September 2012 he spent at the Department of Mechanical Engineering, UC Berkeley, USA, as a visiting resercher, working on stochastic model predictive control with the application to building HVAC systems. His scientific interests include system identification, intelligent control, model predictive control and mechatronics systems.