Fuzzy Based Nonlinear Principal Component Analysis for Process Monitoring

Nowadays, process monitoring to improve product quality and safety has become an important issue in control engineering. One of the most used methods in process monitoring is principal component analysis. This method uses process data to make process model with lower dimension than original dimension. In this paper, a new PCA based monitoring is proposed that uses fuzzy logic capability. The reason to use fuzzy logic is its good ability to approximate nonlinear function with arbitrary accuracy. The new method is tested on Tennessee Eastman Process.