Digital Commons Network^™

Training Set Optimization In An Artificial Neural Network Constructed For High Bandwidth Interconnects Design, Bo Pu, Heegon Kim, Xiao Ding Cai, Bidyut Sen, Chunchun Sui, Jun Fan

Electrical and Computer Engineering Faculty Research & Creative Works

In this article, a novel training set optimization method in an artificial neural network (ANN) constructed for high bandwidth interconnects design is proposed based on rigorous probability analysis. In general, the accuracy of an ANN is enhanced by increasing training set size. However, generating large training sets is inevitably time-consuming and resource-demanding, and sometimes even impossible due to limited prototypes or measurement scenarios. Especially, when the number of channels in required design are huge such as graphics double data rate (GDDR) memory and high bandwidth memory (HBM). Therefore, optimizing the training set selection process is crucial to minimizing the training …

Hamiltonian-Driven Adaptive Dynamic Programming With Efficient Experience Replay, Yongliang Yang, Yongping Pan, Cheng Zhong Xu, Donald C. Wunsch

Electrical and Computer Engineering Faculty Research & Creative Works

This article presents a novel efficient experience-replay-based adaptive dynamic programming (ADP) for the optimal control problem of a class of nonlinear dynamical systems within the Hamiltonian-driven framework. The quasi-Hamiltonian is presented for the policy evaluation problem with an admissible policy. With the quasi-Hamiltonian, a novel composite critic learning mechanism is developed to combine the instantaneous data with the historical data. In addition, the pseudo-Hamiltonian is defined to deal with the performance optimization problem. Based on the pseudo-Hamiltonian, the conventional Hamilton–Jacobi–Bellman (HJB) equation can be represented in a filtered form, which can be implemented online. Theoretical analysis is investigated in terms …

A Methodical Approach For Pcb Pdn Decoupling Minimizing Overdesign With Genetic Algorithm Optimization, F. De Paulis, Y. Ding, M. Cocchini, Chulsoon Hwang, S. Connor, M. Doyle, S. Scearce, W. D. Becker, Albert E. Ruehli, James L. Drewniak

Electrical and Computer Engineering Faculty Research & Creative Works

An optimization routine is applied for the decoupling capacitor placement on Power Distribution Networks to identify the limit beyond which the placement of additional decaps is no longer effective, thus leading to wasting layout area and components, and to a cost increase. A specific test example from a real design is used together with the required target impedance and frequency band of interest for the PDN design. The effectiveness of the decap placement while selecting different layers of the stack-up, and while moving the upper limit of the PDN design band is analyzed. Such analysis leads to helpful insights based …

Variational Data Assimilation For Two Interface Problems, Xuejian Li

Doctoral Dissertations

“Variational data assimilation (VDA) is a process that uses optimization techniques to determine an initial condition of a dynamical system such that its evolution best fits the observed data. In this dissertation, we develop and analyze the variational data assimilation method with finite element discretization for two interface problems, including the Parabolic Interface equation and the Stokes-Darcy equation with the Beavers-Joseph interface condition. By using Tikhonov regularization and formulating the VDA into an optimization problem, we establish the existence, uniqueness and stability of the optimal solution for each concerned case. Based on weak formulations of the Parabolic Interface equation and …

Using Computational Methods To Optimize High Heat Flux Component Thermal Performance In Magnetic Confinement Fusion Reactor Research, Monica Gehrig

Doctoral Dissertations

"Heat transfer enhancement by means of internally modified geometries in tubes and channels is an important mechanism to improve the survivability of components in extreme high-heat flux environments. Various features such as ribs and fins are studied using computational fluid dynamics in both uniform and one-sided heating in tubes and rectangular channels respectively to determine the most effective geometries across a variety of different flow and heating conditions. This work examines heat transfer enhancement and rib geometry optimization to support experimental research for nuclear fusion applications. The project begins by designing and analyzing test sections supporting a helium flow loop …

Particle Swarm Optimization For Critical Experiment Design, Cole Michael Kostelac

Masters Theses

“Critical experiments are used by nuclear data evaluators and criticality safety engineers to validate nuclear data and computational methods. Many of these experiments are designed to maximize the sensitivity to a certain nuclide-reaction pair in an energy range of interest. Traditionally, a parameter sweep is conducted over a set of experimental variables to find a configuration that is critical and maximally sensitive. As additional variables are added, the total number of configurations increases exponentially and quickly becomes prohibitively computationally expensive to calculate, especially using Monte Carlo methods.

This work presents the development of a particle swarm optimization algorithm to design …