基于对称极坐标和图像处理的柴油机故障诊断研究

Diesel Engine Fault Diagnosis Based on Symmetrical Polar Coordinate and Image Processing

将对称极坐标方法和数字图像处理技术结合并应用到柴油机机械故障诊断中。根据对称极坐标方法把不同磨损工况的柴油机连杆轴承振动信号转换成镜面对称雪花图;提出改进面积开运算方法对雪花状图像降噪;分割降噪前后图像特征区域并提取特征参数以反映不同磨损故障程度;最后采用模糊C均值聚类方法评价降噪效果和实现故障分类。试验结果表明,对称极坐标方法能定量地描述柴油机机械故障,改进面积开运算降噪方法能提高和改善柴油机故障诊断的准确率和效果。

Both the symmetric polar coordinate method and digital image processing technology were applied to diesel engine fault diagnosis. By using the symmetric polar coordinate, the connecting rod hearing vibration signal under different wear conditions was converted into snowflakes image with mirror symmetry. Then an improved area opening operation was put forward for the image noise reduction. The image feature area before and after noise reduction was split and the feature parameters were extracted to reflect the different degree of wearing fault. Finally, the noise reduction effect was evaluated and the fault classifi cation was realized with the fuzzy C-means clustering method. The experimental results show that the symmetric polar coordi- nate can describe the mechanical failure of diesel engine quantitatively and the improved area opening operation to noise reduction can improve the accuracy and effect of diesel engine fault diagnosis.