摘要
为了提高对直流有刷电机轴承寿命预测的精准度,设计提出了基于CHSMM预测的电机寿命预测方法。从时域和频域两方面提取轴承震动信号,利用PRA数据算法进行降维,在传统CHSMM预测算法的基础上,引入高斯分布密度函数,将预测算法与函数数据结合,生成预测概率密度函数,求取轴承运行状态驻留时间,基于时间参数推导轴承当前退化状态,进而计算预期寿命,实现轴承寿命预测。实验数据表明,利用新设计的寿命预测方法,轴承退化状态识别精准度可以提高15%左右,剩余价值系数求值精准度提高19.5%左右,可以有效提高轴承预测精准度。
In order to improve the brush in dc motor bearing life prediction accuracy,the design was proposed based on motor life prediction method for prediction of CHSMM is bearing vibration signal extracted from two aspects of time domain and frequency domain,and the PRA data algorithm is used to reduce the dimension.Based on the traditional CHSMM prediction algorithm,is calculated gaussian distribution density function is introduced,to generate the change of state density function,and calculate the bearing running state dwell time.Based on the current state of degradation time parameter derived bearing,calculate life expectancy to realize,bearing life prediction.Experimental data show that using the new design of life prediction method,bearing degradation state recognition accuracy can be improved by about 15%,coefficient of surplus value evaluation precision increased by 19.5% or so,and it can effective improve forecasting precision bearings.
作者
何颋
郑怀华
李灵兵
HE Ting;ZHENG Huaihua;LI Lingbing(State Grid Zhejiang Hangzhou Fuyang District Power Supply Co.,Ltd.Hangzhou 311400.China;Hangzhou ark Electric Power Technology Co.,Ltd.Hangzhou 311202,China)
出处
《自动化与仪器仪表》
2019年第11期72-75,共4页
Automation & Instrumentation
基金
国网浙江省电力公司集体企业科技项目:配电网柔性接地系统(No.JT2017-06)
关键词
轴承
有刷电机
寿命预测
退化状态
bearing
a brush motor
life prediction
degradation state
