Chapter 34 Model Based Monitoring of Dynamic Loads and Remaining Useful Life Prediction in Rolling Mills and Heavy Machinery

摘要

Operation of industrial metallurgical plants is associated with significant wear in spindles, gearboxes and bearings where difficult to implement digital diagnostic tools due to harsh operating conditions. Angular and radial gaps produce extremely high dynamic loads and abrupt failures in the multi-stand hot rolling mills. Reliable vibration monitoring is very complicated due to inherent changes of technological regimes, treated material and drive speed. It appears more beneficial to monitor dynamic torques in addition to vibration signals, but this is restricted to the installation of strain gauges. The more acceptable approach is to monitor static torques of electric motors and, having identified multi-body models, to calculate remaining useful life (RUL) of elements. Based on this approach, the new monitoring system is developed for the multi-stand mill with integration into plant automation infrastructure. Parameters adaptation of nonlinear dynamical models is provided and technological loads optimization by the criterion of RUL in rolling stands. System supports a database of maintenance actions and elements failures. Reports are generated on overloading and RUL.