The digital and high-precision error detection of complex freeform surface

The exploitation and modification of automobile die attach great importance to the automobile exploitation. It should be realized in the process of modifying mould-obtaining 3D model by reverse engineering-testing errors-confirming model. This thesis applies the method of ATOS Optical Scanner and CMM (Three-Coordinate Measuring Machine) to obtain the complete high-precision point cloud data because of the complicated free form surface of automobile model. It is difficult to have error evaluation because the CAD model and the point cloud data are impossible to make automatic alignment. Therefore, on the basis of the surface reconstruction of reverse engineering software Imageware and the comparative functions between surface and point cloud data, we propose a new perspective of detecting digital error, i.e. the contour adjustment of surface automatically by the way of golden section method. Under the condition of satisfying the least square method, this method, which can make anastomosis for the actual and theoretical surface, reduces and even eliminates the position error caused by the inconsistency between the actual reference and CAD model reference, laying a foundation to measure shape error by the way of direct alignment method and reducing time as well as improving the error detection precision. Being simplistic, effective and practical, this method has been successfully utilized in the female die design of interior plastic trimming for a certain type of Shanghai Volkswagen car, having great reference value in terms of the automatization, flexibility and digitalization of products on-line detection.