Comparative analysis of fuzzy logic and machine learning algorithm for predictive analysis of control valve

A control valve is a device used to control the flow of the fluid in the process industry. Detecting the occurrence of the failure is very important for the condition monitoring of the control valve. Vibration data collected from the control valve is used for determining the condition of the valve. For determining the failures in the control valve two techniques are used viz. Fuzzy logic control and Predictive analysis. Fuzzy logic control gives the type of the failure and percentage of the occurrence of the failure, which requires to define the membership function for each failure event. The predictive analysis gives the score according to the condition of the valve. If vibration in the control valve is more than the score will be less and vice versa. The predictive analysis does not need particular failure data to identify failures in the valve. The experimental result shows that the algorithm implemented for the identification of various failures in the control valve and can say that the predictive analysis is more efficient than fuzzy logic as the fuzzy logic requires data to be collected for all the failure events and define membership functions for each type of failure.