Fatigue Life Analysis of Cranes Based on Load Spectrum Prediction and Fracture Mechanics

ABSTRACT A residual fatigue life analysis method based on load spectrum prediction and fracture mechanics has been proposed. In this method, a crane load sequence prediction strategy based on convolutional neural network (CNN) and bidirectional long short‐term memory (BiLSTM) network is proposed to solve the problem of difficulty in obtaining accurate load spectra due to uncertain loads during the operation process in engineering practice. Moreover, a method for calculating the remaining fatigue life of structures based on a combination of fracture mechanics and sequential methods has been proposed. Compared with the linear cumulative damage theory, the influence of structural crack size and external load sequence effects on structural fatigue life is considered. In addition, the remaining fatigue life of the crane girder structure was analyzed using the above method, and the accuracy of the proposed method was verified through comparative analysis.