基于Gocator视觉传感器的轨头参数计算

Calculation of Rail Head Parameters Based on Gocator Vision Sensor

摘要:

轮轨间紧密接触使轨头轮廓产生不规则变化，传统钢轨检测方法主要测量钢轨的垂直磨耗和水平磨耗，不能全面反映钢轨横截面的轮廓信息。基于Gocator视觉传感器采集、拼接得到的完整轨头轮廓数据及标准钢轨轨头轮廓曲线解析式，提出计算轨头剩余面积、轨头45°角磨耗和轨头角度参数的计算方法。该方法通过传感器拼接得到的离散点数据，采用优化分段三次的拟合方法进行多项式拟合，依次对多项式积分得到整个轨头剩余面积；通过轨头45°角所在直线与测量轮廓的交点计算其磨耗；通过计算轨头圆弧相交点处切线斜率得到轨头角度。实验数据表明，在钢轨同一截面计算得到的参数误差小，精度高，计算速度快，这些参数的有效测量值为钢轨自动化打磨维护提供指导意义。

Abstract:

The close contact between the wheel and rail makes the irregularities of the head of the rail, and the traditional rail detection method mainly measures the vertical wear and horizontal wear of the rail, and can not fully reflect the contour information of the rail cross section. Based on the Gocator visual sensor acquisition and stitching the complete rail head contour data and the standard rail track profile curve, the calculation of the residual area of the head, the head 45 ° angle wear and the head angle parameters of the calculation method. In this method, the discrete points data obtained by the sensor splicing are used to optimize the sub-cubic fitting method. The polynomial is obtained by integrating the polynomials in turn. The intersection of the straight line and the measurement profile is calculated by the wear of the 45 ° angle ; calculate the head of the arc at the intersection of tangent slope to get the head angle. The experimental data show that the parameters with the same cross section are small, the precision is high and the calculation speed is fast. The effective measurement of these parameters can provide guidance for the automatic grinding and maintenance of rail.