SVM predictive control for calcination zone temperature in lime rotary kiln with improved PSO algorithm

To improve the control performance of calcination zone temperature in a lime rotary kiln, a predictive control method based on a support vector machine (SVM) and improved particle swarm optimization (PSO) algorithm is proposed. As high-temperature thermal equipment, the lime rotary kiln requires accurate modelling because of its complex non-linearity and long delay characteristics. SVM has strong normalization and good learning ability compared with other modelling models such as neural network, partial least squares model and other non-linear regression models, which can avoid overfitting and local minimization problems. At the same time, it is sometimes difficult to obtain a large number of production sample data of lime rotary kiln. The modelling process based on SVM requires only a small amount of sample data. SVM is appropriate for the modelling of calcination zone temperature of the lime rotary kiln. The predictive control method in this paper utilizes SVM to establish a non-linear prediction model ...