Unlike the machines that operate at higher speed, condition monitoring of low-speed machinery presents a challenge as traditional acceleration sensors make it difficult to differentiate the machine and fault induced vibration from the ‘normal’ noise from the process, as typical amplitude levels of interest are low. Furthermore, it is difficult to monitor low-speed machines reliably using conventional monitoring techniques.

This case study details how the AI-based predictive maintenance solution from Nanoprecise Sci Corp detected a stage 4 fault, with the help of wireless sensors and machine learning algorithms. The solution notified the maintenance and reliability professionals, who took appropriate action and prevented unplanned downtime, resulting in 20x ROI for the mining operation.