전산유체역학과 최적설계기술의 융합에 의한 열 시스템 최적설계

Abstract

The components of circuits on silicon chips such as Central Processing Unit (CPU) of computers have been integrated twice every 5 years according to the so-called "Moor's law". By enhancing the integration rate the Moor's law continues unbroken. The high-integration trend inevitably brought the increase of heat generation per unit area of the chips. The cooling technology is, therefore, becoming important along with the development of electronic devices for the steady operation of the core components. In order to obtain the thermal stability of the core electronic devices, pin-fin or plate-fin type heat sinks are widely used. Due to the compactness of the electronic devices as well as the core components, it is getting difficulties to have enough spaces available for the heat exchange from the components. Since the exchangers such as heat sinks should be installed in the limited spaces, many factors such as the heat transfer, the size of the space available for the installment and the pressure drop are to be simultaneously considered and designed. The optimal design is, therefore, should be performed to obtain the optimal shape or structure of the thermal systems. In this paper, the optimal solutions of the design variables for the thermal systems such as heat sinks and heat exchangers are obtained numerically using the Computational Fluid dynamics (CFD) and the Computer Aided Optimization (CAO)(that is, BFGS method, sequential quadratic programming (SQP) method, genetic algoritm(GA) method, successive approximation optimization (SAO) method, and Kriging method). In the pin-fins heat sink, the optimum design variables for fin height (h), fin width (w), and fan-to-heat sink distance (c) can be achieved when the thermal resistance () at the junction and the overall pressure drop () are minimized simultaneously. To complete the optimization, the finite volume method for calculating the objective functions, the BFGS method for solving the unconstrained non-linear optimization problem, and the weighting method for predicting the multi-objective problem are used. For the shape optimal design of the plate-fin type heat sink with vortex generator, global approximate optimization techniques have been introduced into the optimization of fluid/thermal systems. In this study, Kriging method is used to obtain the optimal solutions associated with the computational fluid dynamics (CFD). In constrained nonlinear optimization problems of thermal/fluid systems, three fundamental difficulties such as high computational cost for function evaluations (i.e., pressure drop and thermal resistance), the absence of design sensitivity information, and the occurrence of numerical noise are commonly confronted. Thus, a sequential approximate optimization (SAO) algorithm has been introduced because it is very hard to obtain the optimal solutions of fluid/thermal systems by means of gradient-based optimization techniques. In this study, the progressive quadratic response surface method (PQRSM) based on the trust region algorithm, which is one of sequential approximate optimization algorithms, is used for optimization and the heat sink is optimize by means of combination it with the computational fluid dynamics (CFD). Shape optimization of internally finned circular tube has been performed for three-dimensional periodically fully developed turbulent flow and heat transfer. The physical domain considered in this study is very complicated due to periodic boundary conditions both streamwise and circumferential directions. Therefore, Pareto frontier sets of a heat exchanger can be acquired by coupling the CFD and the multi-objective genetic algorithm(GA), which is a global optimization technique.2