EEMD-HMM在采煤机轴承故障诊断中的研究与应用

Research and application of EEMD-HMM in fault diagnosis of shearer bearings

采煤机是煤矿采煤过程中的重要设备,因其工作环境和组成结构复杂,关键部位发生损伤故障的情况时有发生。其轴承部位承压巨大,如果采煤机的轴承发生故障未及时发现,将对机器产生严重后果,进而影响工作面的推进进度。为提高开机率,对采煤机故障诊断与故障预测系统进行了研究,提炼出一种集合经验模态分解(EEMD)降噪与隐马尔科夫模型(HMM)的采煤机摇臂滚动轴承故障诊断方法。降噪预处理借鉴的是基于峭度准则的EEMD,先找出拥有大部分特征频率的函数,然后运用公式求解出信息熵然后选出它的敏感特征集,依靠HMM识别轴承的故障类型。

Shearer is an important equipment in the process of coal mining.Because of its complex working environment and composition structure,damage and failure of key parts often occur.The bearing parts bear huge pressure.If the failure of the shearer bearing is not detected in time,it will have serious consequences for the machine,which will affect the progress of the working face.In order to improve the start-up rate,the fault diagnosis and fault prediction system of shearer was studied,and an ensemble empirical mode decomposition(EEMD)noise reduction and hidden Markov model(HMM)for shearer rocker rolling bearing faults were extracted.The noise reduction preprocessing drew on the EEMD based on the kurtosis criterion,first found the function with most of the characteristic frequencies,then used the formula to solve the information entropy and then selected its sensitive feature set,and relied on the HMM to identify the bearing fault type.