LS-SVM Generalized Predictive Control Based on PSO and Its Application of Fermentation Control

Fermentation process is a complex time-varying, nonlinear and multivariable biochemical process. The traditional fed-batch fermentation conditions are difficult to satisfy the control request. A control algorithm based on Generalized Predictive Control (GPC) is proposed. Firstly, the algorithm utilizes least square support vector machine (LS-SVM) and GPC to construct the prediction model and forecast the output value. And then, the particle swarm optimization (PSO) algorithm is applied to realize rolling optimization and obtain the control values. Finally, the control algorithm is applied to control the substrate concentration (S) of lysine fermentation. The simulation results show that the LS-SVM Generalized Predictive Control based on PSO has an excellent adaptive ability with rapid control response speed, high precision, and good performance.