Multi-grade processes have played an important role in the fine chemical and polymer industries. Anintegrated nonlinear soft sensor modeling method is proposed for online quality prediction of multi-gradeprocesses. Several single least squares support vector regression (LSSVR) models are first built for eachproduct grade. For online prediction of a new sample, a probabilistic analysis approach using the statisticalproperty of steady-state grades is presented. The prediction can then be obtained using the correspondingLSSVR model if its probability of the special steady-state grade is large enough. Otherwise, the querysample is considered located in the transitional mode because it is not similar to any steady-state grade.In this situation, a just-in-time LSSVR (JLSSVR) model is constructed using the most similar samplesaround it. To improve the efficiency of searching for similar samples of JLSSVR, a strategy combinedwith the characteristics of multi-grade processes is proposed. Additionally, the similarity factor andsimilar samples of JLSSVR can be determined adaptively using a fast cross-validation strategy with lowcomputational load. The superiority of the proposed soft sensor is first demonstrated through a simulationexample. It is also compared with other soft sensors in terms of online prediction of melt index in anindustrial plant in Taiwan.
Available at: http://works.bepress.com/inter_liu/10/