Answers are grounded in this paper and cite their sources.
Fault diagnosis is particularly important in mechanical engineering systems, where early detection of problems can save costs and time.Traditional fault diagnosis methods rely heavily on features extracted by expert experience, but with the development of intelligent manufacturing, deep learning provides an effective idea for automatically extracting data features.Convolutional neural networks are very effective deep learning methods.In this study, an improvement scheme of convolutional neural network Alexnet algorithm is presented.First, the rolling bearing vibration signal is converted to a symmetric snowflake image in polar coordinates according to the symmetrized dot pattern (SDP) principle.Then, it is fed into the CNN, and the fault feature is extracted.The impact of manual feature extraction is eliminated.According to the ratio of accuracy and time proposed in this paper, the size of each layer of convolution kernel is adjusted accordingly to determine the optimal model, and test experiments are carried out to verify the robustness of the method under different bearing conditions, and the accuracy rate reaches 98.5%.Comparison with other methods shows that the proposed method has achieved significant improvement.
No takes yet. Share an insight, caveat, or question.
Wang et al. (2023) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: