Fault Diagnosis Model of Silk Dryer Based on KPCA-POA-KELM

Identifying and detecting faults in the silk dryer is of great importance, as it serves as crucial equipment for tobacco drying. This article presents a novel fault diagnosis model for silk drying machine based on KPCA-POA-KELM. Firstly, the optimal feature set is obtained by utilizing Kernel Principal Component Analysis (KPCA) for feature optimization. Secondly, the regularization coefficient C and kernel function parameter σ of the Kernel-based Extreme Learning Machine (KELM) are optimized using Pelican Optimization Algorithm (POA). Finally, the proposed model is compared with other models, and the results show that the proposed model achieves a ten-fold cross-validation accuracy of 96.81 % in detecting complex faults. In comparison, the ten-fold cross-validation accuracy of the KPCA-KELM model is found to be the lowest at 68.88%.