Advancements in Industry 4.0 and smart manufacturing have increased the demand for precise and intricate part surface geometries, making the analysis of surface morphology essential for ensuring assembly precision and product quality. This study presents an innovative fitness landscape-based methodology for extreme point analysis of part surface morphology, effectively addressing the limitations of existing techniques in accurately identifying and analyzing extremum points. The proposed approach integrates adaptive Fitness-Distance Correlation (FDC) with a roughness index to dynamically determine the number and spatial distribution of initial points within the pattern search algorithm, based on variations in surface roughness. By partitioning the feasible domain into subregions according to FDC values, the algorithm significantly reduces optimization complexity. Regions with high ruggedness are further subdivided, facilitating the parallel implementation of the pattern search algorithm within each subregion. This adaptive strategy ensures that areas with intricate surface features are allocated a greater number of initial points, thereby enhancing the probability of locating both regional and global extremum points. To validate the effectiveness and robustness of the proposed method, extensive testing was conducted using five diverse test functions treated as black-box functions. The results demonstrate the method’s capability to accurately locate extremum points across varying surface profiles. Additionally, the proposed method was applied to flatness error evaluation. The results indicate that, compared to using only the raw measurement data, the flatness error increases by approximately 3% when extremum points are taken into account.
Read full abstract