Fault Diagnosis Analysis of Angle Grinder Based on ACD-DE and SVM Hybrid Algorithm

Due to the complex structure of angle grinder and existence multiple rotating parts, coupling phenomenon data results in complexity chaos data. The market scale is huge. Manual diagnosis traditional are difficult meet requirements, so a fault method that based on adaptive parameters theory dual-strategy differential evolution algorithm (ACD-DE) SVM model hybrid proposed by combining chaos-mapping algorithm, dynamic factor, crossover factor. effectiveness robustness proven solving eight test functions. acceleration signal decomposed wavelet packet decomposition reconstruction, variety sensor signals processed constructed as feature vectors. training set divided. used optimized ACD-DE. Based grinder, compared with other optimization algorithms machine learning models; comparison show performance improved improved, which precision rate 98.81%, recall 98.74%, F1 score 0.9877. Experiments has strong accuracy robustness.