A local chemical potential approach within the variable charge method formalism

A minimum principle of complementary energy is established for cable networks involving only the stress components as variables with geometrical nonlinearities and nonlinear elastic materials. The minimization problem of total potential energy is reformulated as a variational problem with a convex objective functional and an infinite number of second-order cone constraints; its Fenchel dual problem is shown to coincide with the minimization problem of the complementary energy. It is of interest to note that the obtained complementary energy attains always its minimum value at the equilibrium state irrespective of the stability of the cable networks, contrary to the fact that only stationary principles have been presented for elastic trusses and continua, even in the case of a stable equilibrium state.

Phase decomposition in binary Fe1−x Cu x is studied using Monte Carlo simulations. Initially, density functional theory calculations are utilized to determine reference energies of various Fe–Cu compounds that serve as input for a temperature and composition-dependent cluster expansion. On this basis, the thermodynamic properties of the bcc Fe–Cu system are predicted and used to simulate the equilibrium constitution of bcc Cu-rich precipitates in an Fe-rich solid solution at various temperatures and supersaturations. Complementarily, computationally efficient pair potentials are developed in the local chemical environment approach that are calibrated on the first principles-cluster expansion results. These are then utilized in large-scale simulations for analysis of the multi-particle precipitate evolution. It is concluded that both approaches provide comparable information in terms of the precipitate radius as well as interface constitution. Whereas the cluster expansion (‘full-information’) path is especially useful in predicting energies of various ground state configurations for small systems, the local chemical environment approach (‘fast-computation’) path is particularly useful in evaluation of cluster formation kinetics and evolution statistics.

In this work, we propose a deep learning approach for parallel magnetic resonance imaging (MRI) reconstruction, termed a variable splitting network (VS-Net), for an efficient, high-quality reconstruction of undersampled multi-coil MR data. We formulate the generalized parallel compressed sensing reconstruction as an energy minimization problem, for which a variable splitting optimization method is derived. Based on this formulation we propose a novel, end-to-end trainable deep neural network architecture by unrolling the resulting iterative process of such variable splitting scheme. VS-Net is evaluated on complex valued multi-coil knee images for 4-fold and 6-fold acceleration factors. We show that VS-Net outperforms state-of-the-art deep learning reconstruction algorithms, in terms of reconstruction accuracy and perceptual quality. Our code is publicly available at https://github.com/j-duan/VS-Net.

In this article, a numerical solution method for the finite-strain rate-independent Cosserat theory of crystal plasticity is developed. Based on a time-incremental minimization problem of the mechanical energy, a limited-memory Broyden–Fletcher–Goldfarb quasi-Newton method applied to a finite-difference discretization is proposed. First benchmark tests study the convergence to an analytic solution. Further simulations focus on the investigation of rotation localization zones, the bending of a rod, and a torsion experiment.

The Linear Ordering problem (LOP) is an NP-hard problem, which has been solved using different metaheuristic approaches. The best solution for this problem is a memetic algorithm, which uses the traditional approach of hybridizing a genetic algorithm with a single local search; on the contrary, in this paper we present a memetic solution hybridized with multiple local searches through all the memetic process. Experimental results show that using the best combination of local searches, instead of a single local search, the performance for XLOLIB instances is improved by 11.46% in terms of quality of the solution. For the UB-I instances, the proposed algorithm obtained a 0.12% average deviation from the best known solutions, achieving 17 new best known solutions. A Wilcoxon test was performed, ranking the proposed memetic algorithm as the second best solution of the state of the art for LOP. The results show that the multiple local searches approach can be more effective to get a better control in balancing intensification/diversification than the single local search approach.

A variable transformation method for solving linear absolute-value objective-function problems with linear constraints

High accuracy implicit variable mesh methods for numerical study of special types of fourth order non-linear parabolic equations

To meet the growing quest for enhanced network capacity, mobile network operators (MNOs) are deploying dense infrastructures of small cells. This, in turn, increases the power consumption of mobile networks, thus impacting the environment. As a result, we have seen a recent trend of powering mobile networks with harvested ambient energy to achieve both environmental and cost benefits. In this paper, we consider a network of virtualized small cells (vSCs) powered by energy harvesters and equipped with rechargeable batteries, which can opportunistically offload baseband (BB) functions to a grid-connected edge server depending on their energy availability. We formulate the corresponding grid energy and traffic drop rate minimization problem, and propose a distributed deep reinforcement learning (DDRL) solution. Coordination among vSCs is enabled via the exchange of battery state information. The evaluation of the network performance in terms of grid energy consumption and traffic drop rate confirms that enabling coordination among the vSCs via knowledge exchange achieves a performance close to the optimal. Numerical results also confirm that the proposed DDRL solution provides higher network performance, better adaptation to the changing environment, and higher cost savings with respect to a tabular multi-agent reinforcement learning (MRL) solution used as a benchmark.

Optical properties of Cd20Se80−xMx (M: Zn, In, and Sn) thin film alloys

<!-- *** Custom HTML *** --> Rescaled spike and slab models are a new Bayesian variable selection method for linear regression models. In high dimensional orthogonal settings such models have been shown to possess optimal model selection properties. We review background theory and discuss applications of rescaled spike and slab models to prediction problems involving orthogonal polynomials. We first consider global smoothing and discuss potential weaknesses. Some of these deficiencies are remedied by using local regression. The local regression approach relies on an intimate connection between local weighted regression and weighted generalized ridge regression. An important implication is that one can trace the effective degrees of freedom of a curve as a way to visualize and classify curvature. Several motivating examples are presented.

art, les deux ou trois images simples et grandes sur lesquelles le coeur, une premire fois, s'est ouvert.

This article deals with the parallel numerical solution of the telegraph line. Telegraph partial differential equation is replaced by the system of ordinary differential equations. The linear problem is represented using a system of ordinary differential equations. The numerical solution is obtained using the variable step-size and the variable order Modern Taylor Series Method, which is based on a recurrent calculation of the Taylor series terms for each time interval. The general form of parallelization for the linear systems is introduced.

The close relationship between the ’’bond-charge electronegativity’’ of Ref. 1 and the chemical potential approach to electronegativity is investigated. (AIP)

On density effects in point defect solutions under irradiation. II. Formation of interstitial dislocation loops

Fully-geometric mesh one-leg methods for the generalized pantograph equation: Approximating Lyapunov functional and asymptotic contractivity