Quality assessment on a conical taper part based on the minimum zone definition using genetic algorithms

Abstract

Traditionally, the least-squares method is applied on modern measurement systems for quality assurance. The least-squares method can only provide an approximate solution to the real values of conical parts. It may also result in a possible overestimation of tolerance errors on various engineering surfaces. Thus, a heuristic approach is presented in this paper to efficiently evaluate the cone type error based on the minimum zone definition using genetic algorithms (GAs). GAs show more flexibility in evaluating the engineering surfaces via adjusting the genetic parameters. Two fitness functions were also developed for improving the speed of convergence of GAs. Numerical results indicate that the genetic algorithm provides better accuracy and efficiency and can thus be used to solve difficult product attitude evaluation problems in engineering metrology.