论文部分内容阅读
针对工程中航空滚动轴承实时状态监测的需要,提出了基于标准化欧氏距离的多特征融合评估方法。首先,进行了航空滚动轴承故障模拟试验,引入了故障灵敏度的定量评价指标,对融合前后特征的故障灵敏度进行了分析;在此基础上,将所提方法与主分量分析、支持向量数据描述和支持向量分布估计方法相比较;最后,进行了轴承疲劳加速试验,将所提融合方法应用于航空滚动轴承状态监测。试验表明:相比于主分量分析、支持向量数据描述和支持向量分布估计,基于标准化欧氏距离的融合值的故障灵敏度更高;其对不同类型、不同阶段的航空滚动轴承故障更加灵敏,相比于有效值更适合作为航空滚动轴承状态监测的指标。
Aiming at the need of real-time condition monitoring of aerospace rolling bearing in engineering, a multi-feature fusion evaluation method based on standardized Euclidean distance is proposed. First of all, the flight rolling bearing fault simulation test was carried out, the quantitative evaluation index of fault sensitivity was introduced and the fault sensitivity before and after fusion was analyzed. Based on this, the proposed method and principal component analysis, support vector data description and support Vector distribution estimation method. Finally, bearing fatigue accelerated test was carried out, and the proposed fusion method was applied to the condition monitoring of rolling bearings. The experimental results show that the fusion value based on standardized Euclidean distance has higher fault sensitivity than the principal component analysis, support vector data description and support vector distribution estimation. It is more sensitive to different types and stages of aviation rolling bearing faults, The effective value is more suitable as an indicator of the status of aerospace rolling bearings.