Control and Automated Monitoring of Electromechanical Belt Conveyor Systems

Current implementation of Predictive Maintenance systems forindustrial equipment, particularly belt con-veyor systems, is hindered by technical complexity and the necessity of adapting solutions to specific technological processes. Existing methods often require profound knowledge in vibration diagnostics or substantial computational resources. The relevance of this researchis justified by the urgent need to de-velop accessible and efficient monitoring tools capable of detecting faults (loop breakage, chain sagging) at early stages without production shutdown. The aim of this work is to develop and validate a predictive maintenance system for curved belt conveyors based on the concept of a digital twin and machine learning methods. Materials and methods. Experimental investigations were conducted on a laboratory test rig equipped with a comprehensive sensor array (accelerometers, microphones, current sensors). Data acquisition was performed at high sampling frequency (51.2 kHz), followed by downsampling to 10 kHz using linearinter-polation to optimize computational costs. For data analysis and condition classification («normal» vs. «fault»), four machine learning algorithms were implemented and compared: Random Forest, logistic regression, support vector machines (SVM), and decision trees. Model training and testing were performed on a combined dataset with 80/20 train-test split ratio.