A system for automatic monitoring of conveyor belt integrity using neural networks has appeared in Russia – the “ML Sense Conveyor Belts” module from Nord Clan. The product is included in the register of domestic software and is already being implemented at enterprises in Russia, a company representative told CNews.

Cameras and sensors are installed on the conveyor line, recording belt movement in real time. The neural network algorithm analyzes the video stream and detects tears, ruptures, cuts, nicks, and punctures from 10 mm with an accuracy of up to 98%. The system records defects with coordinates and immediately notifies the operator; data can automatically be sent to the MES system and “1C”. Monitoring is continuous, and implementation occurs without stopping production.

Scheduled operator rounds do not allow for the detection of damage at the moment it occurs – the system fills this gap. For businesses, this means reducing emergency downtime, lowering costs for belt replacement due to early defect detection and extended service life, and freeing personnel from routine inspections.

The solution works with rubber-fabric, rubber-cord, food-grade, and solid-rolled metal belts in the mining, metallurgical, chemical, food, and machine-building industries. It integrates with PLC, SCADA, MES, “1C”, and industrial safety systems, and operates stably in poor lighting, high humidity, and temperature fluctuations.