Intelligent Condition Monitoring System for Predictive Maintenance on Pump
Cavitation is the flow abnormality which has detrimental effect to pump impellers. This condition occurs over time and affect the efficiency of distribution. Currently, there are no direct monitoring measures on cavitation which relates cavitation to the pump variables such as flow rate and vibration. This project investigates cavitation as a variable, along with flow rate and vibration while applying elements of Machine Learning (ML) and Internet of Things (IoT) for an auto scheduling ecosystem to predict the Remaining Useful Life (RUL) of a pump. With this information, pump failure due to cavitation can be predicted and maintenance can be schedules. Results show that the system can achieve over 80% in predicting pump failure.