Design and Simulation of a Model Predictive Controller (MPC) for a Seismic Uniaxial Shake Table

Abstract

Shake table is one of the apparatus that aids in researches to generate techniques, structural developments, and strategies to prevent, prepare, and minimize an earthquake’s devastating effects. One important factor that should be considered in a shake table is the system dynamics due to control-structural interactions, which could either be linear or non-linear. To accurately model both has always been the challenge but becomes more plausible with the availability of faster hardware and computers and the continuous decrease in latency. Model Predictive Controller (MPC) is a type of controller extensively used in the industry that can be used on linear and non-linear systems. This study presents the design and simulation of an MPC for a uniaxial shake table intending to analyze the system’s behavior and accuracy. MATLAB Simulink was utilized to handle the simulation analysis of the controller. Different MPC parameters such as sample time, prediction horizon, control horizon, and closed-loop performance were manipulated and adjusted to observe their effects on the output of the system. A signal that mimics the actual earthquake data was inputted into the controller, and the system's behavior and outputs were measured and presented through graphical representations. To determine the accuracy of the system’s output, its relationship with the reference signal was compared. From the simulation produced, the system demonstrated high accuracy levels and could be adjusted depending on the set performance aggressiveness of the system.