Search

Efficient and precise prediction of plasticity by data-driven models relies on appropriate data preparation and a well-designed model. Here we introduce an unsupervised machine learning-based data preparation method to maximize the trainability of crystal orientation evolution data during deformation. For Taylor model crystal plasticity data, the preconditioning procedure improves the test score of an artificial neural network from 0.831 to 0.999, while decreasing the training iterations by an order of magnitude. The efficacy of the approach was further improved with a recurrent neural network. Electron backscattered (EBSD) lab measurements of crystal rotation during rolling were compared with the results of the surrogate model, and despite error introduced by Taylor model simplifying assumptions, very reasonable agreement between the surrogate model and experiment was observed. Our method is foundational for further data-driven studies, enabling the efficient and precise prediction of texture evolution from experimental and simulated crystal plasticity results., Comment: 15 pages, 6 figures

The GPS is vulnerable to GPS spoofing attack (GSA), which leads to disorder in time and position results of the GPS receiver. In power grids, phasor measurement units (PMUs) use GPS to build time-tagged measurements, so they are susceptible to this attack. As a result of this attack, sampling time and phase angle of the PMU measurements change. In this paper, a neural network GPS spoofing detection (NNGSD) with employing PMU data from the dynamic power system is presented to detect GSAs. Numerical results in different conditions show the real-time performance of the proposed detection method.

In this work, the effect of building direction on the microstructure evolution of laser-powder bed fusion (LPBF) processed AlSi10Mg alloy was investigated. The building direction, as shown in experimentally fabricated parts, can influence the solidification behavior and promote morphological transitions in cellular dendritic microstructures. We develop a thermal model to systemically address the impact of laser processing conditions, and building direction on the thermal characteristics of the molten pool during laser processing of AlSi10Mg alloy. We then employ a multi-order parameter phase field model to study the microstructure evolution of LPBF-AlSi10Mg in the dilute limit, using the underlying thermal conditions for horizontal and vertical building directions as input. The phase field model employed here is designed to simulate solidification using heterogeneous nucleation from inoculant particles allowing to take into account morphological phenomena including the columnar-to-equiaxed transition (CET). The phase field model is first validated against the predictions of the previously developed steady-state CET theory of Hunt \cite{hunt1984steady}. It is then used under transient conditions to study microstructure evolution, revealing that the nucleation rate is noticeably higher in the horizontally built samples due to larger constitutional undercooling, which is consistent with experimental observations. We further quantify the effect of building direction on the local cooling conditions, and consequently on the grain morphology., Comment: 14 pages, 10 figures, two tables

Quantifying extra functions, herein referred to as outcome functions, over optimal solutions of an optimization problem can provide decision makers with additional information on a system. This bears more importance when the optimization problem is subject to uncertainty in input parameters. In this paper, we consider linear programming problems in which input parameters are described by real-valued intervals, and we address the outcome range problem which is the problem of finding the range of an outcome function over all possible optimal solutions of a linear program with interval data. We give a general definition of the problem and then focus on a special class of it where uncertainty occurs only in the right-hand side of the underlying linear program. We show that our problem is computationally hard to solve and also study some of its theoretical properties. We then develop two approximation methods to solve it: a local search algorithm and a super-set based method. We test the methods on a set of randomly generated instances. We also provide a real case study on healthcare access measurement to show the relevance of our problem for reliable decision making.

Flexibility in shape and scale of Burr XII distribution can make close approximation of numerous well-known probability density functions. Due to these capabilities, the usages of Burr XII distribution are applied in risk analysis, lifetime data analysis and process capability estimation. In this paper the Cross-Entropy (CE) method is further developed in terms of Maximum Likelihood Estimation (MLE) to estimate the parameters of Burr XII distribution for the complete data or in the presence of multiple censoring. A simulation study is conducted to evaluate the performance of the MLE by means of CE method for different parameter settings and sample sizes. The results are compared to other existing methods in both uncensored and censored situations.