A new crankshaft bending fatigue test method: both residual life prediction and statistical analysis

In modern engineering application, for the metal engine parts such as the crankshafts and some other related objects, sufficient high-cycle fatigue strength is necessary to guarantee the system reliability during the working period. The traditional resonant crankshaft fatigue test bench can determine the load–life relationship of the given crankshaft directly, but the experiment process usually lasts tens of days, which may result in the large consumption of time. In this paper, the fatigue test speed was accelerated based on the prediction of the residual fatigue life during the experiment process. Then the system state-space equation was modified based on the theory of fracture mechanics to improve the accuracy of the predictions. Finally, the statistical analysis based on the predicted data was adopted to determine the fatigue limit load of the crankshaft. The main conclusion of this paper is that the combination of the particle filter algorithm and the dynamic response signal can predict the residual fatigue life of the crankshaft conveniently based on the modified sampling range, and thus is able to shorten the experiment process and has very wide popularization and application prospects in actual engineering.