Model predictive control based speed control in DC motors

The main objective of the research is to realize the speed control of a DC motor by making the closest prediction about the future behavior of the process with Model Predictive Control (MPC) and Kalman Filter, which is the most recent and advanced feedback control algorithm of the control strategies used to increase the efficiency in the control phase of the systems. Model Predictive Control is a model-based controller that predicts the future behavior of the system using a direct model of the system and calculates the future control signal of the system to optimize a given objective criterion. MPC, which is capable of generating high-performance control systems and handling constraints, and the computationally efficient Kalman Filter, which is used in the development of the prediction model, can work in harmony. The Kalman Filter is a powerful tool that can be used to improve the performance of MPC and make more accurate control decisions. Since it provides information about the current and future states of the system, the Kalman Filter and MPC are used together to improve the predictions of the MPC and to have more accurate information about the actual states of the system. High performance can be achieved by using advanced algorithms such as Finite Set Control Model Predictive Control, a specialized Model Predictive Control method, and the Kalman filter. These algorithms reduce the uncertainties and noise encountered in real-time systems and provide accurate results. In this thesis, the speed control simulation of a Permanent Magnet Direct Current Motor using MPC and Kalman Filter is realized in MATLAB Simulink environment. In this study, a DC-DC converter with H-bridge connection is used to drive the DC motor. FCS-MPC provides the control of the switching states of this converter. For this reason, a rate-based cost function in FCS-MPC is selected to select the switching state. To be used in this controller, the instantaneous armature current is estimated by Kalman Filter. In addition to the armature current, motor speed and load torque estimation are also provided to obtain a more sensitive estimation to the load variation. This study shows that advanced algorithms such as FCS-MPC and Kalman Filter can achieve high performance for DC motor control.