Application of discrete kalman filter algorithm for speed regulation of permanent magnet stepper motor

Abstract

The permanent magnet stepper motor (PMSM) is a robust, low cost electromechanical device which output a mechanical step position for every electrical pulse. The current in the phase winding of the PMSM is responsible for the step angle movement. An abundant number of applications such as combustion engines, automotive industry, telecommunication industry and medical applications use the PMSM for position control. Hence existing literature review focus on step position control and accuracy of the stepper motor. Theoretically position and speed are closely related to each other. If the position of a moving object is maintained constant then, the rate of change in position i.e speed is zero. Similarly if the speed is maintained constant then, the rate of change in speed i.e acceleration is zero. Certain applications of the PMSM like printers, copiers, recorders, scanners require the speed to be regulated after accurate positioning so that an even print can be obtained on the paper feed. This thesis attempts to provide an insight on regulating the speed of the PMSM and thereby validating its performance. The driver plays a vital role in the operation of the PMSM. There are three types of driver used in the literature such as constant current drive, constant voltage drive and constant torque (or) microstep drive for controlling the operation of the PMSM. But if these are operated with the motor in open loop, then, for a fast moving input pulse train, the rotor of the motor may not respond to some of the input pulses, which will lead to a loss of synchronism. Consequently the open loop control is insufficient to improve the performance of the PMSM at high speed. A first solution to improve the performance of the PMSM is closed loop control by providing a feedback path via mechanical sensors. However presence of sensors increases the size and cost of the system. Due to this, sensorless control has emerged, and today is in existence Despite model improvements and presence of diverse control techniques, much work remains to be done to attain maximum performance of the PMSM. In this work, both steadystate and dynamic analysis of the PMSM is presented. The steady-state analysis of the PMSM is derived from the Kuo model. The consolidated results for position and speed obtained from the steady state equations when the motor is operated in single step and microstep mode is presented. The dynamic analysis is carried out in MATLABSIMULINK for free running operation of the motor and also under constant load. The PMSM is modeled in closed loop adopting the field oriented control strategy. Initially conventional Proportional and Integral (PI) based speed controller is designed and the performance of the PMSM with the above stated controller is observed. The simulated results in terms of speed overshoot, quick response and torque ripples are compared with those obtained when the PMSM is simulated with Proportional, Integral and Derivative (PID) based speed controller. Consecutive simulations are performed by replacing the conventional controller with the Fuzzy controller. The advantage of fuzzy theory is that knowledge of the mathematical equations for representing the system is not obligatory. The performance of the PMSM is further improved by adapting a combination of traditional and modern control theory by means of Fuzzy-PID controller in the feedback. newline