Model Predictive Control System Based on FPGA and a Case Study

Model predictive control (MPC) algorithm has been widely applied in industry process control systems since last century, like petro-refining and other chemical processes. However, it is rarely used in field controllers due to the complexity of online optimization. This paper introduces a new MPC system based on the field programmable gate array (FPGA) chip. In order to make MPC controller more efficient, a hybrid way is adopted to design the quadratic programming (QP) solver on the FPGA chip. Thus, the controller can solve the optimization problem of MPC in a pretty short time. The whole control system is implemented on a Virtex-4 FPGA chip and applied to the angle servo control of two motors with a satisfactory result. The experiment shows that it is possible to apply MPC algorithm in field controller by the proposed design.