The soft-sensing and optimal control of microbial fermentation process has become an important research content in the biotechnology field. It is of great significance to improve the level of equipment automation, promote the innovation of biotechnology technology, improve production efficiency and economic efficiency.In this book, the inverse soft sensing model of fermentation process is constructed by combining inverse system method and least square support vector machine theory(LS-SVM), LS-SVM is used to approximate the nonlinear function which is difficult to express explicitly in the inverse soft-sensing model, and the support is established. Secondly, in order to realize decoupling and optimal control of multivariable nonlinear systems, the inverse control method of LS-SVM is proposed, and the inverse system method is combined with IMC. A new method of internal model control for inverse system based on LS-SVM is proposed, which improves the robustness and anti-jamming ability of the inverse control method. Furthermore, the theoretical results are applied to the development of engineering software, and good results are obtained.