Colloquium "MPC via Parametric Optimization and its Application for Wind Farm Optimal Control"

Abstract:

The colloquium began by reviewing the standard Model Predictive Control (MPC) concept. It is now well known that, in the case of linear system dynamics, linear constraints and linear/quadratic objective, the underlying optimization problem is a linear/quadratic program. Furthermore, such a program can be solved explicitly, via so-called parametric optimization, whereas the initial systems state is treated as a parameter. After a brief survey of useful results from parametric optimization, the talk focused on a particular application - optimization-based coordinated control for wind farms. Wind farm should produce demanded power while minimizing the loads on mechanical parts of wind turbines that occur due to turbulent winds. The load reduction is achieved by dynamic adaptation of individual wind turbine power references to the current (turbulence induced) disturbances. The changes in operation of all wind turbines are coordinated to ensure that the wind farm produces power according to the grid operator demands. The pursued approach uses parametric programming to find the optimal control actions for the wind farm with respect to the constraints inherent to the system. The considered control design is also motivated by the need for efficient controller computation that will allow implementations on large wind farms.

CV:
Mato Baotić received the B.Sc. and M.Sc. degrees, both in Electrical Engineering, from the Faculty of Electrical Engineering and Computing (FER Zagreb), University of Zagreb, Croatia, in 1997 and 2000, respectively. In 2005 he received the PhD from the Automatic Control Laboratory at ETH Zurich, Switzerland. He is currently an Associate Professor at the Department of Control and Computer Engineering, FER Zagreb, Croatia. His research interests include mathematical programming, hybrid systems, model predictive control and application of optimal control in renewable energy systems.