Dynamic Modeling of Biotechnical Process Based on Online Support Vector Machine

Abstract

Due to the complexity and high non-linearity of biotechnical process, most simple mathematical models cannot describe the behavior of biochemistry systems very well. Therefore, dynamic modeling of biotechnical process is indispensable. Support vector machine (SVM) is a novel machine learning method, which is powerful for the problem characterized by small sample, non-linearity, high dimension and local minima, and has high generalization. But currently most support vector machine regression (SVR) training algorithms are offline, which could not be suit for time-variant system. So an improved SVM called online support vector machine was presented to modeling for the dynamic feature of fermentation process. The model based on the modified SVM was developed and demonstrated using simulation experiments. Some models based on SVM were also presented. The result shows that the modeling based online SVM is superior to modeling based on SVW.