A fault early warning and health status rating method for ensuring safe operation of rotating equipment

Rotating equipment safe operation requires to recognize the current health state and forecast their future health state. In the field of rotating equipment, research into the modeling of early failure warning and health status rating of equipment is still immature. Therefore, it is important to carry out research in this area. Based on learning information from training fault cases samples, and expert opinion from condition monitoring diagnosis analysis, and health degree statistical classification from engineering case study, the cross-correlation function, the spectral distance function, the coherence function health evaluation models and the equipment health status rating criterion have been constructed. Combined with the early warning threshold self-learning theory, health status rating and the ℓ1 trend filtering algorithm, a kind of health evaluation method has been proposed in this paper. The fault cases of the rotor imbalance of a centrifugal compressor and a laboratory bearing were used to verify the proposed method. The results showed that this approach can be used to evaluate the health status of rotating equipment or a part and can be used to guide the decision-making process for equipment operation and maintenance to ensure the safe operation of rotating equipment.