An industrial microcontroller design for servo control systems

In this thesis, various microcontrollers which are widely used in industry such as PIC16F877A, dsPIC and Arduino Due are analyzed and an industrial microcontroller is designed for some processes and servo systems. Sampling frequency and processing speed of microcontrollers are the most important factors affecting the performance of process control having high time constant and electrical drives having low time constant. On the other hand, computing load of the control algorithms which will be implemented in micro-controller, such as Proportional, Integral and Derivative (PID) or intelligent control algorithms such as artificial neural networks and fuzzy logic control are also imortant factors affecting performance of microcontroller and control system. Therefore, to investigate the feasibility of intelligent control method in the micro-controller based industrial control system is the main objective of this study. Firstly, the performances of PIC16F877A and dsPIC developed by Microchip and Arduino Due development platform designed by Atmel microprocessor with AT91SAM3X8E microcontroller are compared under same operating conditions. A sinusoidal signal with constant frequency at the analog input of the microcontroller are taken into microcontroller using analog-to-digital converter (ADC) and, the PID control algorithm and some matrix operations are implemented in microcontroller and then the sinusoidal signal are loaded to digital-to-analog converter (DAC). Thus, the processing speed of the microcontroller and the most suitable sampling frequency for control systems are determined by measuring the delay between the input and output signal. Simulation and experimental results showing the performance of microcontrollers are presented. Application software of the PIC16F877 microcontroller are programmed with mikro C PRO program and the simulation is implemented in ISIS subroutine of Proteus 7 Professional. Application results for Arduino Due's are obtained from the prepared program by using the Arduino Due's interface program. Results shows that the performance of Arduino Due is better than the others. In the second stage of the thesis, by using the Arduino Due platform, closed loop control of a temperature control system which requires low sampling frequency and the speed control of DC motor which requires higher sampling frequency are implemented. Speed information of the DC motor for feedback are obtained both from encoders and tacho generators. PID and fuzzy logic control methods are applied to temperature and DC motor control systems. Experimental results showing the performance of the control systems are evaluated in MATLAB program by getting from the serial port of the Arduino IDE program. Experimental results showing effectiveness and drawbacks of the control system under tests are evaluated.