Application of graphic processing unit in model predictive control

Sadrieh, A. and Bahri, P.A. (2011) Application of graphic processing unit in model predictive control. Computer Aided Chemical Engineering, 29 . pp. 492-496.

Abstract

This study seeks to pave the way for the implementation of a model predictive controller method using a Graphic Processing Unit (GPU). The GPU has been adapted and used as a real time co processor for Nonlinear Model Predictive Control (NMPC) algorithms, providing a means to improve the computational performance of the MPC algorithm in an economic manner. In this approach, a parallel version of Nelder-Mead simplex algorithm was used to solve MPC optimization problem. In order to show the effectiveness of the proposed approach, the implementation was applied in model predictive control of a crystallizer unit operation. The results show a considerable improvement in computational performance compared to standard CPU based implementation.