Intelligent indices for online monitoring of stress and condition

摘要

Automatic fault detection with stress and condition indices enables reliable condition monitoring to be combined with process control. Useful information on different faults can be obtained by selecting suitable features from generalised norms, which are defined by the order of derivation, the order of the norm and sample time. The methodology selects sensitive signals for different faults: the frequency ranges are selected with the order of derivation and the signal amplitude ranges by the order of norm. The nonlinear scaling based on generalised norms and skewness extends the idea of dimensionless indices to nonlinear systems and provides good results for the automatic generation of stress and condition indices. The stress index is -2 when the stress is negligible, and levels {-1, 0, 1} are analogue to the lower limits of the vibration severity ranges {usable, still acceptable, not acceptable}. The damage risk is predicted by the cumulative weighted sum of the stress contributions. Trends of the simulated sums could be used to predict the failure time if the operating conditions are not changed considerably. However, the approach allows varying operating conditions: the stress indices monitor the equipment in all operation conditions, whereas the condition indices need be compared in similar conditions throughout the operation. Tests in two applications: a Kaplan water turbine and an underground load haul dump (LHD) machine, demonstrate that the effects of the varying stress provide useful information for damage prediction. The approach generates from signals new informative indices to be used in the same way as normal process measurements.