Soft sensor study and application based on WSVM

JU Xuan, LIU Chunbo, PAN Feng
 （Key Laboratory of Advanced Process Control for Light Industry, Ministry of Education,
Jiangnan University, Wuxi 214122, China ）

Abstract: In order to solve the problem associated with random disturbance in the process of soft sensor on wastewater disposal process, a new method called support vector machine(SVM) with multiresolution wavelet kernel function was presented. A soft measure of BOD & COD on soft sensors model of wastewater treatment process control parameters was achieved by using a rare way of building a mixed model based on SVM and parameters traits. The experimental results indicate that the soft measure can overcome a series of problems through the process of sewage disposal, such as random interfere, strong nonlinearity, big time change, serious lag, and so on. Therefore, this method will create conditions for getting parameters online and make wastewater disposal automatically.
Key words: soft measure; wavelet kernel function; support vector machine(SVM); sewage disposal