Electrothermal Continuous Crystallization Machine Multivariable Predictive Control Algorithm Research

Multivariable model predictive control 70 production in the 20th century, the field of industrial process control, after 30 years of development, has become the most popular advanced control algorithms and in practical industrial applications and achieved fruitful results.Since the continuous crystallization of electric machines still rely on manual has been put into operation and control of the semi-mechanized state, resulting in product yield and quality of instability. Continuous crystalline electric machine of the paragraphs tank water volume is an important factor in the temperature gradient, the temperature gradient maintained in a certain range is to ensure the quality of products and key. If you simply rely on the experience of workers on-site to adjust the water spray will inevitably lead to the low level of automation, production efficiency is not high, but will also cause the increase in energy consumption. Therefore, this paper carried out based on the dynamic matrix control (DMC) algorithm for continuous crystallization of the electric machine temperature gradient of multivariable model predictive control.Paper first outlines the development of multivariable predictive control and application, and analysis of electric machines continuous crystallization mechanism and the dynamic characteristics of the work; in a row has received the go electric machine dynamic mechanism of crystallization model based on the soaring curves were obtained Test, followed by crystal machine to obtain the transfer function of the temperature of each segment, and further to the multivariable predictive control required in the prediction model (dynamic matrix control the transfer function matrix); the last key to the continuous crystallization of electric control of the multi-variable model Algorithm, and using MATLAB7.8 as the development environment, the study of the multi-variable model predictive control algorithm was simulated, the results show that the continuous crystallization of the electric machine to implement predictive control effect is satisfactory.This study has some practical, can be realized in the future automation of electric continuous crystallization machine lay a foundation.