A simulated annealing approach for curve fitting in automated manufacturing systems

Abstract

Purpose This research aims to develop an effective and efficient algorithm for solving the curve fitting problem arising in automated manufacturing systems. Design/methodology/approach This paper takes curve fitting as an optimization problem of a set of data points. Expressing the data as a function will be very effective to the data analysis and application. This paper will develop the stochastic optimization method to apply to curve fitting. The proposed method is a combination optimization method based on pattern search (PS) and simulated annealing algorithm (SA). Findings The proposed method is used to solve a nonlinear optimization problem and then to implement it to solve three circular arc‐fitting problems of curve fitting. Based on the analysis performed in the experimental study, the proposed algorithm has been found to be suitable for curve fitting. Practical implications Curve fitting is one of the basic form errors encountered in circular features. The proposed algorithm is tested and implemented by using nonlinear problem and circular data to determine the circular parameters. Originality/value The developed machine vision‐based approach can be an online tool for measurement of circular components in automated manufacturing systems.