Abstract
In this study, using data-driven methods, we develop a soft senor based on a multiple local model for a nonlinear industrial process. The soft sensor is based on a novel learning algorithm, which uses online subtractive clustering to recursively update the structure and parameters of a local model network. We also propose rules for updating the centers and local model coefficients of existing clusters, for generating new clusters and new models as well as for merging existing clusters and their corresponding models. As an industrial example, the proposed algorithm is applied to an o-xylene purification column, and it is shown that it is possible to track dynamic trends and compactly accumulate operating experiences. The performance of the proposed approach is compared with that of adaptive principal component regression, adaptive linear models based on key variables selection, fixed partial least-squares, and radial basic function neural network. The results demonstrate the effectiveness of the proposed modeling approach. © 2010 American Chemical Society.