Comparison of Wireless Temperature Profiles With Generalized Predictive Control

Abstract

In this study, Generalized Predictive Control (GPC) algorithm is applied to a process simulator which wireless temperature experiments were achieved and the results of the experiments were compared under the same conditions obtained. To achieve the data transfer between computer in Process Control Laboratory and the process simulator in Unit Operations Laboratory, wireless communication system was established and wireless experiments were performed on-line by means of MATLAB/Simulink program. Wireless data transfers during the experiments were carried out by using radio waves at a frequency of 2.4 GHz. In wireless temperature control experiments which are conducted using with algorithm of GPC, all L (control weighting) values which are bigger than 1.0 are not suitable for temperature control because the heater made very big oscillatory behaviour and consequently temperatures also made very big oscillatories at any Nu (control costing horizon), N1 (minimum costing horizon) and N, (maximum costing horizon) values. Therefore GPC experiments were carried out in L values smaller than 1.0. Changes of temperature profiles were observed with time as a result of experiments carried under the same conditions, using the Nu, N-1 ve L parameters in the GPC algoritm to the N-u=1.0; N-1=1.0; N-2=2.0; N-U =1.0; N-1=1.0; N-2=4.0; N-U=2.0; N-1=1.0; N-2=2.0 ve N-U=2.0; N-1 1.0; N-2 =4.0 values for the L=0.001; 0.005; 0.01; 0.05; 0.1 and 0.5 values. According to the experimental results analyzing the temperature profiles, the temperatures the N-1=1.0 and N-2=2.0 values of control was closer to the set point and oscillations were found to be less than according to the values of N-U=1.0; N-1=1.0; N-2=4.0; N-U=2.0; N-1=1.0; N-2=2.0 ve N-U=2.0; N-U=1.0; N-2=4.0. Also by increasing the value of Nu and N, temperatures are begin to move away from the set point and the oscillations are increased. Wireless experiments conducted under the same conditions for Tset=50 degrees C and Tset 60 degrees C which using the best temperature control parameters with the L=0.001; 0.005; 0.01; 0.05; 0.1 and 0.5 values were determined with the integral of the square of the error (ISE) and the integral of the absoluteof the error (IAE) values. According to the experiments carried out at the lowest values of ISE and IAE were determined that the value L 0.01. In addition, lower and higher values of L= 0.01 calculated ISE and IAE values were higher and that can be seen the temperature away from set point. In conclusion, the best temperature control is obtained N-U=1.0; N-1=1.0; N-2=2.0 and L=0.01 values.