Search

The time-domain dynamic process model is used to generate data and guides the stability criteria for machine learning, saving the experimental costs for a number of required data for the metal process. Fourier transformation of vibration data simulated using a dynamic process model generates the feature lists including multiple frequencies and amplitudes at each process condition. The feature lists for milling stability are analyzed for training the machine learning algorithm. The amplitude and frequency distributions may change according to the dynamic pattern of the machining stability. The vibration patterns are grouped into stable, chatter, and boundary conditions by performing data training using support vector machines and gradient tree boosting. In the high-speed milling of Al6061-T6 with 6000 to 18,000 RPM and variations of axial and radial depths of cuts, 2400 data sets of the time domain data were trained and tested. Actual experimental tests are carried out for new process conditions with the range of 9890 to 28,470 RPM and 989 to 2847 mm/min. The experimental stability outcomes are compared with predictions from the algorithms. Stability is accurately predicted over new conditions with around 0.9 prediction accuracy, which means the methodology can be used to predict, categorize, and monitor stability in end milling processes.

In liquid crystal (LC) displays, deriving an optimum resistance level of an LC alignment polyimide (PI) layer is important because of the trade-off between the voltage holding and surface-discharging properties. In particular, to apply a power-saving low-frequency operation scheme to fringe-field switching (FFS) LC modes with negative dielectric LC (n-LC), delicate material engineering is required to avoid surface-charge-dependent image flickering and sticking problems, which severely degrade with lowering operation frequency. Therefore, this paper proposes a photocontrolled variable-resistivity PI layer in order to systematically investigate the voltage holding and discharging properties of the FFS n-LC modes, according to the PI resistivity (ρ) levels. By doping fullerene into the high-ρ PI as the photoexcited charge-generating nanoparticles, the ρ levels of the PI were continuously controllable with a wide tunable range (0.95 × 1015 Ω∙cm to 5.36 × 1013 Ω∙cm) through Ar laser irradiation under the same LC and LC alignment conditions. The frequency-dependent voltage holding and discharge behaviors were analyzed with photocontrolled ρ variation. Thus, the proposed experimental scheme is a feasible approach in PI engineering for a power-saving low-frequency FFS n-LC mode without the image flickering and image sticking issues.

In this article, a digitized stress function-based feed rate scheduling algorithm is formulated for the prevention of tool breakage while having an optimum material removal rate in mesoscale rough milling of hardened steel. Instead of setting limits to the cutting forces and material removal rates, the presented method regulates the tool’s stresses. A 3D coupled Eulerian-Lagrangian finite element method (FEM) model is used to simulate a 3D chip flow-based stress according to the mesoscale tool’s rotation during cutting of hardened steel. Maximum uncut chip thickness and tool engaging angle of the uncut chip is identified as the fundamental driving factors of tool breakage in down milling configuration. Furthermore, a multiple linear regression model is formed to digitize the stress with two major factors for digitized feed scheduling. The optimum feed rates for each segment along the tool path can be obtained through finite element models and a multiple linear regression model. The feed rate scheduling method is validated through cutting experiments with tool paths of linear and arc segments. In a series of experimental validations, the algorithm demonstrated the capability of reducing the machining time while eliminating cutting tool breakages.