A Multiobjective Optimization Algorithm Based on Gaussian Quantum Behavior of DFAW Network for Fault Detection

In the field of fault diagnosis, intelligent fault diagnosis based on deep learning has attracted much attention in recent years. However, due to the low efficiency of traditional manually designed neural networks, the huge computational burden, and the large amount of processing time they require, their real-time applications have been greatly hindered. Therefore, this article proposes an improved method based on the DFAW network. This method improves the DFAW network structure by introducing the global average pooling (GAP) technique and the MO Gaussian quantum PSO (MOGQPSO) algorithm. The method includes two stages: first, the structure of the DFAW network is changed to reduce the computational burden of the network, and then Gaussian perturbations and quantum behavior are added to improve the multiobjective PSO (MOPSO) algorithm. The optimized MOGQPSO algorithm is used to optimize the key parameters of the model and improve the overall accuracy of the network. After applying the method to the XJTU-SY bearing dataset and the MFPT dataset, the comparison results show that the method has obvious advantages in accuracy and model efficiency.