论文部分内容阅读
机电设备运行状态的监测对保障系统稳定可靠运行、预防重大事故发生有重要意义。针对传统诊断方法由于故障信息不足导致的诊断精确度差,提出了一种基于主特征模式识别的故障诊断方法。基于多源特征信息融合,研究了基于多传感器系统的特征融合故障诊断模型,讨论了反映系统运动状态特征的指标体系及故障诊断算法。文中以滚动轴承系统故障诊断为例,首先计算了各传感器获取信号的时域特征参数,然后,借助主特征模式对特征信息进行融合与降维处理,实验测试数据显示出与传统诊断方法相比较该算法有更好的故障诊断性能。研究结果表明了该方法在重型机电设备故障诊断中应用的可行性与合理性。
The monitoring of mechanical and electrical equipment operating status is of great importance to ensure stable and reliable operation of the system and prevent major accidents. Aiming at the poor diagnostic accuracy of the traditional diagnosis method due to insufficient fault information, a fault diagnosis method based on the main characteristic pattern recognition is proposed. Based on multi-source feature information fusion, a fault diagnosis model of feature fusion based on multi-sensor system is studied. The index system and fault diagnosis algorithm reflecting the characteristics of system motion are discussed. Taking the fault diagnosis of rolling bearing system as an example, the time-domain characteristic parameters of each sensor are calculated firstly. Then, the feature information is fused and dimension-reduced by using the main characteristic mode. The experimental test data shows that compared with the traditional diagnosis method Algorithm has better fault diagnosis performance. The results show the feasibility and rationality of this method in the fault diagnosis of heavy mechanical and electrical equipment.