Early warnings using no-code AI for predictive maintenance

Challenge

Vibration monitoring on large rotating equipment is an everyday use of predictive maintenance. Alarms are frequently employed but are often problematic due to the variation in vibrational sensors leading to many unwanted alarms.

Solution

With Intelecy anomaly detection, you can learn the exact vibration pattern on your equipment. In addition, it can recognize different modes of operations or even different "products" going through the mill, dryer, crusher, separator, or other equipment.

For example, when the engineers on site noticed that the average vibrational measurement increased over time before a breakdown, recognizing that they couldn't use a standard alarm system, they created machine learning models on various vibration sensors in their plant. Within days of activating the machine learning models, the company identified anomalies on two motors allowing them to perform a controlled shutdown.

In both cases, only minor damage was inflicted, and repairs were made during scheduled maintenance. In addition, by using Intelecy early warning, the client could avoid downtime of 3 days on two occasions within a month.

Result

• Predictive maintenance implemented
• Unplanned downtime reduced 
• Efficiency increased