Beer Boiling Temperature Application Of Predictive Functional Control Based On Support Vector Machine

Brewery is a traditional industry which not only its control process is very complex but also contains many operation steps, and there still is a long way to go to catch up with the advanced level in the world for our country. So, it is an urgent issue to improve our technical level in brewing industry, so that we can actively participate in international competition with stronger comprehensive strength. Although there are many reasons that result in the gap between our level and advanced level, such as management level, raw material quality and so on, but the automatic level in this area is one of the most important factors.The main content of this paper is as follows:(1) The boiling process of Beer Saccharification is generally presented; the temperature control and equipments in this process are highlighted. And then least squares support vector machines (LS-SVM) modeling method is introduced, and regressive model of boiling temperature process with LS-SVM through modeling of mechanism and experiments is designed. The result of simulation shows that the method can achieve better result. It not only gets over the difficulties of mechanism modeling but also shortens the period of experiments modeling.(2) Based on the LS-SVM model of boiling process in Beer Saccharification, through the study of single model predictive functional control, a new multi-model predictive functional control algorithm is designed for the temperature control of the boiling process, and the simulated result shows that it meets the desired control request well.