基于JSOA-VMD分解的回转窑故障诊断研究  

Research on Rotary Kiln Fault Diagnosis Based on JSOA-VMD Decomposition

作　　者：王彦峰 张云[2] WANG Yanfeng;ZHANG Yun(School of Mechanical and Electronic Engineering,Wuhan University of Technology,Wuhan 430070,China;Rotary Kiln Testing Technology Center in Wuhan University of Technology,Wuhan 430070,China)

机构地区：[1]武汉理工大学机电工程学院,武汉430070 [2]武汉理工大学回转窑检测技术中心,武汉430070

出　　处：《数字制造科学》2024年第3期194-199,共6页

摘　　要：在对回转窑的运转状况进行诊断时,能否准确的提取窑故障特征信号显得非常重要。通过对托轮的受力分析,得出托轮振动信号中包含能够表征窑故障的信号分量。为解决采集的信号中故障信号难以准确提取问题,提出使用变分模态分解算法VMD进行分解和提取,为解决模态数和二次惩罚项两个参数难以准确设置问题,使用跳蛛优化算法JSOA对其进行优化。最后使用JSOA-VMD算法对托轮振动信号进行分解和故障诊断,对窑筒体截面和椭圆度进行检测,验证了该方法在窑故障诊断方面的准确性和可靠性,为窑故障诊断提供新的方法。Accurate extraction of kiln fault characteristic signals is critical for diagnosing rotary kiln operations.Through stress analysis of the supporting roller,it is concluded that the vibration signal of the roller contains components that can indicate kiln fault.To address the difficulty in extracting fault signals from collected data,the variational mode decomposition(VMD)algorithm is proposed.The jumping spider optimization algorithm(JSOA)is used to optimize the mode number and the quadratic penalty term parameters.The JSOA-VMD algorithm was then applied to decompose the supporting roller’s vibration signal for fault diagnose.Using our lab’s new patent for kiln shell section and ovality detection,the method’s accuracy and reliability in kiln fault diagnosis were verified,providing a novel approach for fault detection in rotary kilns.

关 键 词：回转窑 变分模态分解 跳蛛优化算法 托轮 故障诊断 信号分析 

分 类 号：TQ172.622[化学工程—水泥工业]