Fault diagnosis of reel motor in hot continuous rolling based on Bayesian decision
XIE Feng1, FU Wen-peng2, LI Yang1, XIE Xiang-qun2, LI Wei-gang1
1. School of Information Science and Engineering, Wuhan University of Science and Technology, Wuhan 430081, Hubei, China; 2. Hot Rolling Plant, Meishan Iron and Steel Co., Nanjing 210039, Jiangsu, China
Abstract：The reel motor is the power source of the reel, which has an important influence on the coil shape quality and coiling stability. Aiming at the squirrel cage broken problem of the reel motor in the coiling process of the hot strip mill, the current signal of the reel motor is collected from the steel mill, and the signal characteristics are extracted through signal interception, filtering, fitting and other operations. The multivariate Gaussian model is used for modeling, and the fault diagnosis is performed based on the Bayesian classification algorithm with minimum risk. Experimental results prove that the Bayesian decision based on minimum risk can better diagnose the failure phenomenon of the squirrel cage strip of the reel motor. Compared with support vector machines and Fisher classifiers, the proposed method has a good classification effect and the ability to capture abnormal samples can be enhanced by adjusting the risk coefficient.
谢丰, 付文鹏, 李阳, 谢向群, 李维刚. 基于贝叶斯决策模型的热轧卷筒电机故障诊断[J]. 中国冶金, 2021, 31(4): 68-73.
XIE Feng, FU Wen-peng, LI Yang, XIE Xiang-qun, LI Wei-gang. Fault diagnosis of reel motor in hot continuous rolling based on Bayesian decision[J]. China Metallurgy, 2021, 31(4): 68-73.
2021, Vol. 31 
