| Early detection, diagnosis, and even prognosis of a spall in the bearing for the safe and economic operation of rotating machinery have been a field of challenging ongoing research. The objective of this study is to establish model based bearing diagnostic algorithms to enhance the sensitivity in detection bearing localized defect and to assess the damage severity in terms of impact magnitude and spall sizes. For detection, an embedded model was developed to identify the "gross stiffness" of a rotor-bearing system. For impact magnitude and spall size estimation, an inverse method was developed to estimate the impact impulse from which the spall size can be induced. In addition to the analytical work, experimental studies were conducted to evaluate the proposed algorithms in terms of strength and limitations. Future work is suggested based on the findings. |
