AUTOMATION OF INDUSTRIAL EQUIPMENT DIAGNOSTICS USING GO AND ML

UDC 330.34
DOI: 10.36871/ek.up.p.r.2025.05.07.020

Authors

Abdul-Khalid A.-M. Aybuyev,
Hamid Sh. Nasurov,
Grozny State Petroleum Technical University named after Academician M. D. Millionshchikov, Grozny, Russia

Abstract

The article discusses an approach to the automation of industrial equipment diagnostics using the Go programming language and machine learning methods. The main focus is on the system architecture aimed at reducing the total cost of ownership (TCO) by optimizing capital (CAPEX) and operating expenses (OPEX). The use of Go is justified by its high performance, ease of scaling, and costeffectiveness in the development of microservices. Machine learning methods, including classification, regression, and anomaly detection algorithms, are used for predictive diagnostics, which reduces the number of unscheduled downtimes, increases equipment utilization, and improves key financial indicators — ROI, NPV, and IRR. An experimental evaluation based on SCADA telemetry data confirmed that the proposed system allows for a reduction in operating expenses of up to 30% and a return on investment in less than 12 months. The results demonstrate the economic feasibility of implementing automated diagnostic systems based on Go and ML in an industrial environment.

Keywords

machine learning, equipment diagnostics, predictive maintenance, CAPEX, OPEX, TCO, ROI, SCADA, automation