基于点云处理的差速器行星齿轮齿面球形半径测量方法

Measurement method of differential planetary gear tooth surface spherical radius based on point cloud processing

  • 摘要: 为了满足工业领域对差速器行星齿轮齿面球形半径在高精度和快速测量方面的需求,提出了一种基于点云处理的非接触式测量方法。首先,对行星齿轮点云数据进行预处理,并提取齿面点云。其次,在特定位置生成空间切平面,对齿面点云进行旋转抽样,并在每次抽样后提取齿面与空间切平面交线上的轮廓点。再次,对各交线上的轮廓点进行空间圆拟合,并构建半径特性曲线,通过设定容忍值进一步筛选拟合点。最后,利用改进的粒子群算法对拟合点进行空间球体拟合,从而实现齿面球形半径的非接触精确测量。试验结果表明,所提方法的最大绝对误差为0.029 mm,最大重复性标准差为0.015 mm,算法平均测量耗时为4.56 s,具有较高的测量精度与速度,并且整个测量过程完全自动化,满足企业的批量化自动测量要求。

     

    Abstract: To address the industrial demand for high-precision and rapid measurement of the spherical radius of differential planetary gear tooth surfaces, a non-contact measurement method based on point cloud processing is proposed.Firstly, the point cloud data of the planetary gear is preprocessed and the point cloud of tooth surface is extracted. Secondly, spatial cutting planes are generated at specific locations, rotational sampling of the tooth surface point cloud is performed, and the contour points on the intersection line between the tooth surface and the spatial tangent plane are extracted after each sampling. Thirdly, the spatial circle fitting of contour points on each intersection line is carried out, and the radius characteristic curve is constructed. The fitting points are further screened by setting tolerance values. Finally, an improved particle swarm optimization algorithm is employed to perform spatial sphere fitting on the fitted points, achieving precise non-contact measurement of the spherical radius of the tooth surface. Experimental results show that the maximum absolute error of the proposed method is 0.029 mm, the maximum repeatability standard deviation is 0.015 mm, and the average measurement time of the algorithm is 4.56 s, which has high measurement accuracy and speed, and the whole measurement process is fully automated, which meets the requirements of enterprises for mass automated measurements.

     

/

返回文章
返回