Dissolved oxygen generic model control of wastewater treatment process based on immune optimization least squares support vector machine

Biological wastewater treatment is a nonlinear, time-varying and time-delay process. It is difficult to achieve precise control. In order to solve the instability problem of dissolved oxygen (DO) control, a method of DO generic model control (GMC) based on immune optimization least squares support vector machine (LS-SVM) is proposed. The method doesn't require model simplification and linearization. A soft sensor method based on LS-SVM is proposed to solve the difficulty of oxygen uptake rate on-line measurement in DO process model. The immune algorithm is used to optimize the parameters of LS-SVM. In order to improve the distribution diversity of antibodies, an immune selection operator is proposed. Experimental results indicate that the proposed method is of high control accuracy.