Optimized Adaptive Iterative Sparse Principal Component Analysis Methodology for Fault Detection and Identification in Control Valves

With the advancement of artificial intelligence (AI), modern industrial fault diagnosis has become a prominent area of research. Among them, PCA-based fault diagnosis has been widely applied across various industrial domains. However, the rapid progress in the industrial sector has revealed limitations of PCA in handling dynamic and time-varying data, rendering it unsuitable for deployment in industrial process control applications. The progress in machine learning and artificial intelligence has paved the way for the appropriate extension of traditional methods, expanding their practical applications in the real world. This paper presents an optimized adaptive iterative sparse PCA (OAISPCA) approach. By iteratively updating the sparse penalty term through an interior point-based method, adaptive sparsity is introduced into the original PCA method, thereby enhancing model interpretability and fault diagnosis accuracy. The effectiveness and advantages of this method are validated through its application to real-world industrial control valve data. The results demonstrate that OAISPCA significantly improves performance metrics such as fault detection rate (FDR) and false alarm rate (FAR).