Abstract: For the assessment of remaining fatigue life, prevention of catastrophic accidents and safety of the cranes in service, in view of the highly random and uncertain working condition of cranes, a large number of data investigations are conducted for bridge cranes for general purpose. In a certain period, the numbers of work cycles corresponding to different lifting loads for different rated lifting capacity are collected. Firstly based on artificial neural network (ANN), the equivalent load spectrum, which is equivalent to the actual load spectrum of the estimated crane, is acquired. Meanwhile, according to Paris-Erdogan equation along with Miner’s fatigue damage accumulation theory, the linear elastic fracture mechanics theory and rainflow algorithm, the remaining fatigue life formula could be deduced and the estimation of the remaining fatigue life for the crane could be completed. The example demonstrates: it could be quick to acquire the equivalent load spectrum of the estimated crane and to estimate its remaining fatigue life by this approach and the time-consuming, tedious process and the massive investment for the cranes field testing could be avoided. Compared with the remaining fatigue life predicted through the field testing of stress spectrum, it has been proven that this method is feasible and effective with better consistency and application.

Key words: Artificial neural network (ANN), Bridge cranes, Equivalent load spectrum, Remaining fatigue life