Hybrid Fault Diagnosis for High Speed Wire Rod Finishing Mill

High speed wire rod finishing mill is a significant equipment in steel production enterprise. However, due to the complex nature of this system-level equipment, fault location and real-time monitoring can be challenging. Aiming at this issue, this paper proposes a hybrid fault diagnosis method based on expert experience and data-driven for high speed wire rod finishing mill. Firstly, fault feature extraction is improved by analyzing mechanical structure of finishing mill in both time and frequency domains. In time domain, a combination of virtual value, peak value, and kurtosis index is utilized, while in frequency domain, speed adjustment and side frequency analysis are proposed to accurately identify component characteristic frequency and corresponding sidebands. Based on these time and frequency domain characteristics, an expert experience-based fault location approach is proposed to accurately position fault. Finally, the proposed method is illustrated using an on-site equipment fault as an example to demonstrate its effectiveness.