Evaluation of axis straightness error based on modified dung beetle optimization algorithm

Abstract

The straightness error of axis components is a key parameter for assessing the quality of modern equipment manufacturing. Traditional error assessment methods, such as those used for shaft and hole components, often suffer from limitations including low robustness, low efficiency, and insufficient accuracy. To address these issues, this paper proposes a Modified Dung Beetle Optimization (MDBO) algorithm for evaluating axis straightness errors. Firstly, the mathematical model for evaluating axis straightness using the minimum zone method is formulated as the objective function. Secondly, the principles of the Dung Beetle Optimization (DBO) algorithm are analyzed, and three enhancement strategies are introduced to improve its performance: a good point set strategy to address uneven population distribution during initialization, a spiral search strategy to refine position updates during dung beetle reproduction and foraging behaviors, and a Lévy flight strategy to improve position updates in the stealing behavior, which enhances the algorithm’s ability to escape local optima. Then, the performance of MDBO is tested through four practical engineering design problems, showing satisfactory results. Finally, the MDBO achieves axial straightness errors of 0.015697 mm (inner surface) and 0.016175 mm (outer surface) for tubular components, improving accuracy by 20.99% and 20.59%, respectively, compared to the Least Squares Method (LSM). Validation through engineering design problems and experimental measurements confirms MDBO’s robustness and competitiveness in precision manufacturing applications.