Methodology for assessing the resource of power equipment of metallurgical workshops

Abstract

Prognostic is central to manufacturing and infrastructure management. The development of new methods for determining the residual resource of equipment is an important task aimed at increasing the efficiency of using industrial electrical complexes. Solving the problem of reliably determining the state of power equipment in the mining and metallurgical complex makes it possible to move from an outdated system of scheduled preventive maintenance to maintenance based on the current state of the unit. Complex technical systems are characterized by complex nonlinear interactions between their constituent elements, complex scenarios of cause-and-effect relationships between hazardous, probabilistic events and processes occurring during their operation. Therefore, methods and tools are being developed to assess and manage wear mechanisms in high-risk industries. The article presents the results related to the development of a methodology for assessing the residual life of industrial equipment. An algorithm for assessing the residual life of power equipment is proposed, based on the comparison of the results of calculation of predicted values of the criteria of the residual life of industrial equipment with the normative values. The methodology uses probabilistic mathematical methods to predict the remaining service life and information collected during audits and equipment monitoring. Along with the classical methods, methods are presented based on the use of the entire potential of the modern element base of microprocessor technology and technologies for the use of artificial neural networks, machine learning, and «big data». Based on the application of probabilistic analysis, a methodology for monitoring the predicted residual life of power equipment has been proposed, for which an algorithm for solving diagnostic problems using neural networks has been developed.