A novel dense residual network based on Adam-S optimizer for fault diagnosis of bearings under different working conditions

Abstract

In recent years, the residual network has been widely used in the field of intelligent diagnosis because of its powerful functioning. This paper proposes a novel dense residual network (DRNet) for the efficient fault diagnosis of rolling bearings, which combines the advantages of dense connections and residual learning to prevent the gradient disappearance and network degradation caused by network deepening. First, each sub-block in the dense network (DesNet) is deeply processed so that it has better nonlinear expressive ability to extract deep fault features. Then, the residual learning is embedded into each sub-block of the DesNet, so that each sub-block processed by deepening will not show the phenomenon of network degradation. Finally, an Adam-subtracted momentum optimization algorithm is proposed, which adds the first-order momentum and the second-order momentum of the previous gradient into the expression of the second-order momentum of the current gradient, which enhances the connection between the parameters in the two adjacent gradients in the Adam algorithm. It makes the algorithm more reliable and the gradient prediction more accurate. Without adding additional parameters, the training stability of the algorithm in complex environments is further improved. Experiments on two kinds of data sets under different working conditions are carried out many times, and in comparison with the random forest, support vector machine, dense network, residual network, AlexNet and DRNet-Adam. This proves the effectiveness and feasibility of the proposed model and optimization algorithm.