Elaboration of forecasting methods of analysis of metallurgical facilities electric drives technical state

Abstract

To prevent losses of equipment downtime, analytical algorithms and data of automated control systems sensors are used to estimate time to the equipment failure approach. It was shown that modern technical systems, such as automated electric drives are equipped by measuring devices which allow to accomplish self-diagnostic of the system in real time mode at existing methods of data processing and analysis. This approach is more effective than traditional methods of diagnostic and does not include additional capital expenses for specialized diagnostic equipment and personal qualification. Hypothesis of forecasting of electromechanical driving system state was elaborated and checked for continuous rolling mill runaway roller, based on character of loading modes by using only one parameter – the electric motor current. Based on the data analysis of wide strip rolling mill 2000 runaway rollers operation, normalized curves of distribution of average values of the runaway roller drive current at its normal operation and origination of a malfunction were built. It was shown that a technical system state change is fixed at appearance of a deviation of load current distribution comparing with a standard deviation. Analysis of dynamics of a statistical parameter, the standard error (the difference between actual and standard distribution) at a transition process allowed to make a forecast of roller jamming several days before the malfunction took place. The proposed approach of equipment state estimation can become a base for elaboration of a principally new methods of diagnostic of metallurgical rotor equipment.