Minimum zone sphericity evaluation based on a modified cuckoo search algorithm with fuzzy logic

Abstract

This study proposes an intelligent algorithm that combines a modified cuckoo search (MCS) algorithm with an adaptive fuzzy logic controller (MCSF) to evaluate minimum zone sphericity with high speed and accuracy. To accelerate the convergence of the iteration procedure, the proposed MCS algorithm introduces an optimized initial solutions distribution model and a modified biased/selective mechanism to the original cuckoo search. Moreover, an adaptive fuzzy logic controller is designed to ensure global search ability in the early stage of the search process and strong convergence performance in the later stage of iteration. This is achieved by adaptively adjusting the control coefficient α of step size and the probability pa of discovering an invasive cuckoo’s egg in MCS. Simulation and comparison results indicate that the proposed method achieves excellent convergence behavior and reaches the global optima faster than conventional heuristic methods. The MCSF method runs 60 iterations in ~0.4 s, and its results for the standard deviation of sphericity are 1/8–1/2 of those of the conventional heuristic methods. The proposed method can easily be extended to evaluate minimum zone roundness, cylindricity, and straightness.

Micro-Scale Spherical and Cylindrical Surface Modeling via Metaheuristic Algorithms and Micro Laser Line Projection

“…The micro laser line projected on the metallic surface is shown in Figure 8b. Thus, the metallic surface is scanned in the x-axis via the micro laser line to compute the coordinates (x i,j , y i,j ) by means of Equations ( 26) and (27). Then, the coordinates (x i,j , y i,j ) are substituted in Equations ( 22) and (23) to compute the micro-scale surface (z i,j , y i,j ).…”

“…To do so, the microscope vision system scans the surface to perform the micro-scale surface modeling. During the scanning, the CCD camera captures the micro laser line to compute the line coordinates (x i,j , y i,j ) via Equations ( 26) and (27). Then, the coordinates (x i,j , y i,j ) are substituted in Equations ( 22) and ( 23) to compute the surface coordinates (z i,j , y i,j ).…”

“…The radial distortion is determined via the micro laser line position. The micro laser line position (x i,j , y i,j ,) is computed via Equations ( 26) and (27). Thus, the radial distortion is deduced by means of the line position x i,j = x i,j + δx i and y i,j = y i,j + δy j , where (x i,j , y i,j ) are the distorted coordinates and (δx i , δy j ) are the distortion.…”

“…Therefore, artificial intelligence algorithms have been implemented to accurately construct surface models. For instance, an algorithm with an adaptive fuzzy logic controller has been implemented to construct spherical surface models [27]. This algorithm minimizes an error function by employing surface data to obtain the spherical surface model parameters.…”