Research on Diesel Engine Fault Status Identification Method Based on Synchro Squeezing S-Transform and Vision Transformer

The reliability and safety of diesel engines gradually decrease with the increase in running time, leading to frequent failures. To address problem that it is difficult for traditional fault status identification methods identify engine faults accurately, a method based on synchro squeezing S-transform (SSST) vision transformer (ViT) proposed. This can effectively combine advantages SSST processing non-linear non-smooth signals powerful image classification capability ViT. vibration reflecting are collected by sensors. solve problems low time-frequency resolution weak energy aggregation signal analysis methods, used convert into two-dimensional maps; ViT model extract features training achieve assessment. Pre-set experiments carried out using condition monitoring experimental bench, proposed compared three namely, ST-ViT, SSST-2DCNN FFT spectrum-1DCNN. results show overall accuracy public dataset actual laboratory data reaches 98.31% 95.67%, respectively, providing new idea identification.