Hybrid Fault Identification Analysis for an Industrial Control Process

Fault detection and diagnosis play a key role in dealing with the malfunctioning of industrial systems. Different data-driven techniques have been explored for successful fault detection and diagnosis. One technique known as Principal Component Analysis is one of the prominent techniques in detecting anomalies. In this paper, an advanced novel technique known as Adaptive Principal Component with wavelet denoising is used to reduce anomalies. Results have shown that it aids in reducing false alarm rate, missed detection rate, and time delay while increasing the fault detection rate for the Penicillin Fermentation Process. The findings haveshown that the method used in this paper is more effective and successful for more robust, reliable, and sensitive fault detection.