K-nearest Neighbors Based Nonlinear Process Monitoring Using Kernel EPCA

摘要

In this paper, we introduce a new technique for nonlinear monitoring process relying on kernel entropy principalcomponent analysis (KEPCA). KEPCA can transform input data into high-dimensional feature space using the nonlinearkernel function and determine the number of principal components (PCs) based on the computation of the entropy. Theretained PCs are the ones that explain the maximum entropy of data in the feature space. Then, we introduce a newapproach to calculate the upper control limits (UCLs) for the squared prediction error (SPE) and the T² Hotelling in thefeature space based on the density estimation via the k-nearest neighbors (kNN) estimator. The abovementionedapproaches were applied to fault detection for the benchmark Tennessee Eastman process (TE). Results were robust andsupply better performance than KPCA.