Load identification of the gearbox using artificial neural networks

This paper proposes an approach to identify loads acting on the gearbox, which uses an artificial neural network to model the load-strain relationship in structural analysis. The identified results meet accuracy demands by comparing these outputs with experiment values The paper demonstrates that artificial neural network can be used very effectively in load identification. The well-trained neural network reveals an extremely fast convergence and a high degree of accuracy in the gearbox load identification. Due to BP algorithm cannot work perfectly n the actual application, so many improved algorithm come up. In this study, an Improved Error Back Propagation algorithm is adopted to train artificial neural network.