Runtime Stress Estimation for Three-dimensional IC Reliability Management Using Artificial Neural Network

Abstract An artificial neural network (ANN) is a mathematical or computational model that simulates the structure and function of biological neural networks used to evaluate or approximate functions at given points. After developing the training algorithm, the resulting model will be used to solve image recognition problems, control problems, optimization, etc. In the process of ANN training, the algorithm of backpropagation is used in the case of convex optimization functions. The article is analyzed test functions for experiments and also study the effect of the number of ANN layers on the quality of approximation in cases one-, two- and three-dimensional. The backpropagation method is improved during the experiments with the help of adaptive gradient, as a result of which more accurate approximations of the functions are obtained. This article also presents the numerical results of test functions.

Download Full-text

Utilization of a double-cross-linked amino-functionalized three-dimensional graphene networks as a monolithic adsorbent for methyl orange removal: Equilibrium, kinetics, thermodynamics and artificial neural network modeling

Environmental Research ◽

10.1016/j.envres.2021.112156 ◽

2021 ◽

pp. 112156

Author(s):

Ceren Karaman ◽

Onur Karaman ◽

Pau-Loke Show ◽

Yasin Orooji ◽

Hassan Karimi-Maleh

Keyword(s):

Neural Network ◽

Artificial Neural Network ◽

Methyl Orange ◽

Network Modeling ◽

Three Dimensional ◽

Neural Network Modeling ◽

Artificial Neural Network Modeling ◽

Artificial Neural ◽

Double Cross

Download Full-text

An Artificial Neural Network Based Solution Scheme for Periodic Computational Homogenization of Electrostatic Problems

Mathematical and Computational Applications ◽

10.3390/mca24020040 ◽

2019 ◽

Vol 24 (2) ◽

pp. 40 ◽

Cited By ~ 2

Author(s):

Felix Selim Göküzüm ◽

Lu Trong Khiem Nguyen ◽

Marc-André Keip

Keyword(s):

Neural Network ◽

Artificial Neural Network ◽

Trial Function ◽

A Priori ◽

Three Dimensional ◽

Internal Parameters ◽

Artificial Neural ◽

Artificial Neural Network Ann ◽

Core Idea ◽

Electrostatic Problems

The present work addresses a solution algorithm for homogenization problems based on an artificial neural network (ANN) discretization. The core idea is the construction of trial functions through ANNs that fulfill a priori the periodic boundary conditions of the microscopic problem. A global potential serves as an objective function, which by construction of the trial function can be optimized without constraints. The aim of the new approach is to reduce the number of unknowns as ANNs are able to fit complicated functions with a relatively small number of internal parameters. We investigate the viability of the scheme on the basis of one-, two- and three-dimensional microstructure problems. Further, global and piecewise-defined approaches for constructing the trial function are discussed and compared to finite element (FE) and fast Fourier transform (FFT) based simulations.

Download Full-text