Improved Process Monitoring Based on Sparse Principal Component Analysis (SPCA)

The principal components analysis was (PCA) intensively developed and widely applied in industrial processes for monitoring. The purpose of using the PCA is to reduce the extractable dimension of the still valid feature space to the most information in the primary dataset. Sparse Principal Component Analysis (SPCA) is a relatively new mechanism proposed for the creation of Principal Components (PCs) with sparse loads via a variance tradeoff. Using the SPCA, some of the loads on the PCs can be limited to zero. In this article, we applied this method for the diagnosis and monitoring of a biological process. The results obtained are discussed.

• Process Monitoring
• Principal Component Analysis (PCA)
• Sparse Principal Component Analysis (SPCA)
• fault Detection and Isolation.

• Process Monitoring
• Principal Component Analysis (PCA)
• Sparse Principal Component Analysis (SPCA)
• fault Detection and Isolation.