Bioprocess soft sensing based on multiple kernel support vector machine

Abstract

Soft sensing technology is one of the topics of general interest in study on current process control, which has recently drawn considerable attention worldwide, and has stimulated researchers and engineers to make greater effort to reduce the cost/benefit-ratio for development and manufacture of bio-industrial processes both economically and environmentally. This paper introduced a kind of soft-sensor based on an improved support vector machine (SVM) for a polyacrylonitrile productive process. The improved SVM called the multiple kernel support vector machine was presented, and the mathematical formulation of multiple kernel learning is given. Through the implementation for average molecular weight in polyacrylonitrile productive process, it demonstrates the good performance of the proposed method compared to single kernel.