基于数据驱动的故障诊断方法已被广泛应用于旋转机械零部件故障诊断领域。目前,大多数诊断方法主要依赖于定长数据分割产生的大量数据,但分割的数据通常为短周期的小片段信号,而实际长周期冗余信号由于数据尺度不匹配,无法直接作为测试...基于数据驱动的故障诊断方法已被广泛应用于旋转机械零部件故障诊断领域。目前,大多数诊断方法主要依赖于定长数据分割产生的大量数据,但分割的数据通常为短周期的小片段信号,而实际长周期冗余信号由于数据尺度不匹配,无法直接作为测试样本进行故障识别。针对以上不足,提出了一种新的基于数据概率密度与一维卷积神经网络(Data Probability Density and One-Dimensional Convolutional Neural Network,DPD-1DCNN)的故障诊断方法,其具有两个特点:①提取信号的密度特征可抵抗数据的冗余;②适应不同长度的冗余信号可作为诊断模型的输入。该方法采用DDS试验台产生的行星齿轮箱故障数据进行了验证;其在保证高诊断精度的同时,又增强了诊断模型的适应性。展开更多
A new distribution for the fluctuation of materials＇ lifetime cumulative hazard rate is firstly proposed. The new distribution is extended from the Weibull distribution by adding a sine function. After that, the prop...A new distribution for the fluctuation of materials＇ lifetime cumulative hazard rate is firstly proposed. The new distribution is extended from the Weibull distribution by adding a sine function. After that, the properties of its hazard rate function, cumulative hazard rate function, probability density function and cumulative distribution function are studied. The analysis result shows this distribution can well model the lifetime with variable and periodic hazard rate. Finally, the new distribution is verified with two real data sets as examples to demonstrate its capability.展开更多
文摘基于数据驱动的故障诊断方法已被广泛应用于旋转机械零部件故障诊断领域。目前,大多数诊断方法主要依赖于定长数据分割产生的大量数据,但分割的数据通常为短周期的小片段信号,而实际长周期冗余信号由于数据尺度不匹配,无法直接作为测试样本进行故障识别。针对以上不足,提出了一种新的基于数据概率密度与一维卷积神经网络(Data Probability Density and One-Dimensional Convolutional Neural Network,DPD-1DCNN)的故障诊断方法,其具有两个特点:①提取信号的密度特征可抵抗数据的冗余;②适应不同长度的冗余信号可作为诊断模型的输入。该方法采用DDS试验台产生的行星齿轮箱故障数据进行了验证;其在保证高诊断精度的同时,又增强了诊断模型的适应性。
基金supported by the National Natural Science Foundation of China under Grant Nos.11461051and 11361036the Natural Science Foundation of Inner Mongolia under Grant No.2014MS0112
文摘A new distribution for the fluctuation of materials＇ lifetime cumulative hazard rate is firstly proposed. The new distribution is extended from the Weibull distribution by adding a sine function. After that, the properties of its hazard rate function, cumulative hazard rate function, probability density function and cumulative distribution function are studied. The analysis result shows this distribution can well model the lifetime with variable and periodic hazard rate. Finally, the new distribution is verified with two real data sets as examples to demonstrate its capability.
