稀土萃取过程组分含量神经网络软测量方法.pdfVIP

Vol. 32, No. 4 ACTA AUTOMATICA SINICA July, 2006 Component Content Soft-sensor Based on Neural Networks in Rare-earth Countercurrent Extraction Process1) YANG Hui1 CHAI Tian-You2 1(School of Electronics and Electrical Engineering, East China Jiaotong University, Nanchang 330013) 2(Research Center of Automation, Northeastern University, Shenyang 110004) (E-mail: yhshuo@263.net) Abstract Throught fusion of the mechanism modeling and the neural networks modeling, a compo- nent content soft-sensor, which is composed of the equilibrium calculation model for multi-component rare earth extraction and the error compensation model of fuzzy system, is proposed to solve the prob- lem that the component content in countercurrent rare-earth extraction process is hardly measured on-line. An industry experiment in the extraction Y process by HAB using this hybrid soft-sensor proves its effectiveness. Key words Rare-earth, countercurrent extraction, soft-sensor, equilibrium calculation model, neural networks 1 Introduction China has the most abundant rare-earth resource in the world[1,2] . But the extraction process automation is still in the stage that component content is measured off-line, and the process is controlled by experience and parameters are regulated by hands. This situation leads to low efficient production rate, high resource consumption and unstable production quality, and limits the rare-earth industry development[2]. To implement automation in the rare-earth extraction process, the on-line component content measuring must be achieved at first. Main component content meas