Laser cutting provides various advantages such as high flexibility in terms of process parameters and cut material type, as well as the possibility to obtain complex geometry in different dimensions with high precision. From industrial point of view, the two more competitive laser cutting technologies are based on the use of CO2 and active fiber sources, which produce samples visually different, with non-uniform surface and different depth of the striations. The quality assessment between the two laser systems within the industry is commonly based on standard ISO 9013; that covers several aspects of quality, the most used are the surface roughness and edge perpendicularity; however 2D profilometers adopted for measures are not able to analyze the complex 3D surface topography of the cutting edge. As a result, despite the fact that the differences are visually appreciated, measured 2D roughness values of different CO2 and fiber laser cutting conditions are very similar. Recently, a greater diffusion of 3D surface profilometry devices is present. These devices allow areal surface roughness parameters to be defined, which are potentially suitable to better quantify the laser cut quality. This work points out the use of a focus-variation microscopy to acquire 3D surfaces and evaluate analytically the surface quality of laser cut edges using areal surface roughness parameters. In particular, the purpose is to define a simple and repeatable method to identify the type of cutting process analyzed through the reconstruction of surface characteristics and quality of the cut-edge. As a case study, two stainless steel samples with the same geometry obtained with different laser sources, CO2 and active, fiber is presented. For comparison purposes the cutting conditions were fixed to represent the state of the art of respective laser cutting technologies, which actually show distinct cutting edge characteristics.
This work proposes a model for suggesting optimal process configuration in plunge centreless grinding operations. Seven different approaches were implemented and compared: first principles model, neural network model with one hidden layer, support vector regression model with polynomial kernel function, Gaussian process regression model and hybrid versions of those three models. The first approach is based on an enhancement of the well-known numerical process simulation of geometrical instability. The model takes into account raw workpiece profile and possible wheel-workpiece loss of contact, which introduces an inherent limitation on the resulting profile waviness. Physical models, because of epistemic errors due to neglected or oversimplified functional relationships, can be too approximated for being considered in industrial applications. Moreover, in deterministic models, uncertainties affecting the various parameters are not explicitly considered. Complexity in centreless grinding models arises from phenomena like contact length dependency on local compliance, contact force and grinding wheel roughness, unpredicted material properties of the grinding wheel and workpiece, precision of the manual setup done by the operator, wheel wear and nature of wheel wear. In order to improve the overall model prediction accuracy and allow automated continuous learning, several machine learning techniques have been investigated: a Bayesian regularized neural network, an SVR model and a GPR model. To exploit the a priori knowledge embedded in physical models, hybrid models are proposed, where neural network, SVR and GPR models are fed by the nominal process parameters enriched with the roundness predicted by the first principle model. Those hybrid models result in an improved prediction capability.
The paper presents a novel geometrical stability analysis of centerless grinding that takes into account the nonlinearity associated to wheel-workpiece detachment during lobes formation. Even though the rounding mechanism in centerless grinding has been studied since more than fifty years, stability analysis has been carried out applying stability criteria for linear systems (e.g., Nyquist) on a process model that neglects actual removal “clipping” due to wheel-workpiece detachment. This model limitation is usually overcome by considering only an integer number of lobes, supporting the restriction by the claim that a non-integral number of waves is less likely to build up since the waviness must be constantly removed and replaced by a succeeding wave, which is constantly moving around the workpiece. In this work, the nonlinearity entailed by removal clipping is explicitly taken into account and, by harmonic linearization, represented by a double input describing function (DIDF). Applying the Nyquist criterion on the resulting equivalent delayed system, the paramount instability associated to a quasi-integer number of lobes emerges naturally, without requiring additional assumptions. Moreover, it is shown that the nonlinearity due to wheel-workpiece detachment does not produce a limit cycle in a reasonable operation time. The results delivered by the proposed approach are verified by numeric simulations and positively compared to the relevant literature. The proposed formulation can be easily extended to consider also machine structure dynamics, thus increasing, even in this case, the accuracy of the stability analysis provided by the standard approach.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.