摘要
稀土生产设备状态的有效检测,是保证稀土生产过程可靠、连续运行以及产品质量的关键。通过对萃取槽上的皮带和电动机故障类型进行分析,提出一种基于改进贝叶斯算法的故障检测方法。以稀土萃取生产过程故障检测中的几个主要参量为研究对象,对比分析贝叶斯估计与正则化二次判别方法,引入级数逼近将二者结合起来,通过散度函数最小化贝叶斯分类算法的期望误差,对皮带和电动机的故障进行二分类,形成改进贝叶斯算法的故障检测方法。实验结果表明,此方法能够准确地检测出生产过程中的故障类型为皮带或者电动机的故障。
Effective detection of the status of rare earth production equipment is the key to ensuring reliable, continuous operation and product quality of the rare earth production process. A fault detection method based on improved Bayes algorithm is proposed by analyzing the fault types of the belt and motor on the extraction tank. Taking several main parameters in the fault detection of rare earth extraction production process as the research object, comparative analysis of Bayesian estimation and regularized quadratic discrimination method, the introduction of series approximation is used to combine the two, and the Bayesian is minimized by the divergence function, the expected error of the classification algorithm effectively classifies the faults of the belt and the motor to form a fault detection method of the improved Bayesian algorithm. The results show that this method can accurately detect that the fault type in the production process belongs to the belt or the motor.
作者
王占涛
罗奕
卢新佳
张筵凯
WANG Zhantao;LUO Yi;LU Xinjia;ZHANG Yankai(School of Mechatronic Engineering,Guilin University of Electronic Technology,Guilin 541004,China)
出处
《桂林电子科技大学学报》
2021年第5期420-425,共6页
Journal of Guilin University of Electronic Technology
基金
桂林市科技攻关项目(20190211-14)
桂林电子科技大学研究生教育创新计划(2019YCXS010)。
关键词
贝叶斯
正则化
生产线
故障检测
智能技术
皮带
电动机
Bayes
secondary discrimination
production line
fault detection
intelligent technology
belt
motor
