Evaluation of Geometric Deviations in Sculptured Surfaces Using Probability Density Estimation

摘要

The number and location of samples acquired during coordinate measurement are among the most important parameters that affect the accuracy of evaluating the actual geometric deviations and comparing them with tolerance specifications. A method based on iterative sampling of discrete 3D points is presented for evaluating actual geometry deviations of machined sculptured surfaces and for assisting the important accept/reject decisions making process. The strategy relies on estimating the density function of the geometric deviations between the actual and substitute surfaces, using the Parzen windows technique. The computed error density function is used to identify portions of the surface that require further iterative sampling of measurement points until the desired accuracy is achieved. This method may be used for assessing any geometry deviations due to systematic or non-systematic error sources. No prior assumptions about the fundamental form of the underlying probability density function of errors are required. Examples are presented to both illustrate the proposed methodology and validate its performance. The use of this method reduces inspection cost as well as the cost of rejecting good parts or accepting bad parts.