Rapid Transient Thermal Analysis of Three-Dimensional Interconnect Structures Using Proper Orthogonal Decomposition Method

Author(s):  
Banafsheh Barabadi ◽  
Satish Kumar ◽  
Yogendra K. Joshi

The increase in the integration of interconnect wiring, as well as the high level of current densities are resulting in increased concerns about hot spot formation due to Joule heating in the metal lines of microprocessors. This temperature rise poses a major challenge in maintaining the quality and reliability of future devices, requiring a focus on physics based approaches for rapid and accurate thermal analysis of interconnect architectures. This work investigates the problem of transient Joule heating in a three-dimensional array of copper interconnects embedded in dielectric layers of SiO2 and Si3N4 using Proper Orthogonal Decomposition (POD) as the reduced order modeling approach. The case of natural convection was assumed on the boundaries. For validation, the results were compared with a three-dimensional finite volume model developed in Fluent and good agreements models were observed. While the Fluent model required hours of computational time, the POD based model predictions were achieved within seconds.

Author(s):  
Elizabeth H. Krath ◽  
Forrest L. Carpenter ◽  
Paul G. A. Cizmas ◽  
David A. Johnston

Abstract This paper presents a novel, more efficient reduced-order model based on the proper orthogonal decomposition (POD) for the prediction of flows in turbomachinery. To further reduce the computational time, the governing equations were written as a function of specific volume instead of density. This allowed for the pre-computation of the coefficients of the system of ordinary differential equations that describe the reduced-order model. A penalty method was developed to implement time-dependent boundary conditions and achieve a stable solution for the reduced-order model. Rotor 67 was used as a validation case for the reduced-order model, which was tested for both on- and off-reference conditions. This reduced-order model was shown to be more than 10,000 times faster than the full-order model.


Author(s):  
Mahdi Hosseinali ◽  
Stephen Wilkins ◽  
Lhendup Namgyal ◽  
Joseph Hall

In this paper, classic Proper Orthogonal Decomposition (POD) on a polar coordinate and snapshot POD on a Cartesian grid will be applied separately in the near field of a turbulent wall jet. Three-component stereoscopic PIV measurements are performed in the transverse plane of a wall jet formed using a round contoured nozzle with a Reynolds number of 250,000. Eigenfunctions and energy distributions of the two methods are compared. Reconstructions using same number of modes and same content of energy have been compared. The effect of grid resolution on the energy content of the classic method has also been studied.


2019 ◽  
Vol 20 (K9) ◽  
pp. 5-14
Author(s):  
Nguyen Ngoc Minh ◽  
Nguyen Thanh Nha ◽  
Truong Tich Thien ◽  
Bui Quoc Tinh

The consecutive-interpolation technique has been introduced as a tool enhanced into traditional finite element procedure to provide higher accurate solution. Furthermore, the gradient fields obtained by the proposed approach, namely consecutive-interpolation finite element method (CFEM), are smooth, instead of being discontinuous across nodes as in FEM. In this paper, the technique is applied to analyze transient heat transfer problems. In order increase time efficiency, a model- reduction technique, namely the proper orthogonal decomposition (POD), is employed. The idea is that a given large-size problem is projected into a small-size one which can be solved faster but still maintain the required accuracy. The optimal POD basis for projection is determined by mathematical operations. With the combination of the two novel techniques, i.e. consecutive-interpolation and proper orthogonal decomposition, the advantages of numerical solution obtained by CFEM are expected to be maintained, while computational time can be significantly saved.


2004 ◽  
Vol 126 (3) ◽  
pp. 416-421 ◽  
Author(s):  
Sean A. Mortara ◽  
Joseph Slater ◽  
Philip Beran

The nonlinear panel flutter problem solved by Dowell in 1966 is used to investigate the new application of the proper orthogonal decomposition model reduction technique to aeroelastic analysis. Emphasis is placed on the nonlinear structural dynamic equations with nonconservative forcing modeled assuming a supersonic, inviscid flow. Here the aeroelastic coupled equation is presented in discrete form using a finite difference approach, and subsequently in state space form, to be integrated as a set of first order differential equations. In this paper, a POD approach is developed for generalized second-order differential equations; however, the application of POD to the governing equations in state space form is also discussed. This study compares the results and effectiveness of the model reduction technique for integration of the full set of degrees of freedom. The solution is compared to Dowell’s classic results which forms the base reference for the model reduction study. The reduced order model is then created from the full simulation model. Accuracy of the solution, reduced computational time, limits of stability, and the strengths and weaknesses of the model reduction are investigated.


Sign in / Sign up

Export Citation Format

Share Document