Volume 13 | Issue 4
Volume 13 | Issue 4
Volume 13 | Issue 4
Volume 13 | Issue 4
Volume 13 | Issue 4
This study examines how deep learning improves mechanical fault diagnostics. As equipment becomes more complex, diagnostic methods must improve. Intelligent Industrial Fault Diagnosis utilizing Sailfish Improved Inception with Residual Network (IIFD-SOIR) Model is introduced in this paper. The model performs signal portrayal, highlight extraction, and arrangement. Continuous Wavelet Transform (CWT) pre-processes the vibration signal in the proposed model. High-level features are generated using Inception with ResNet v2 feature extraction. A sailfish optimizer tunes Inception's ResNet v2 model parameters. A multilayer perceptron (MLP) classification method is used to accurately diagnose problems. Extensive experimentation ensures the model's gearbox and motor bearing dataset results. On the gearbox and bearing datasets, the IIFD-SOIR model had a higher average accuracy of 99.8% and 99.68%. Compared to other methodologies, the simulation showed that the proposed model performed well. Advanced deep learning approaches can improve mechanical system failure diagnostics, improving dependability and maintenance efficiency in industrial applications.