Design Optimization of Latent Heat Thermal Energy Storage System Using Computational Fluid Dynamics, Response Surface Methodology and Genetic Algorithm

Abstract

Product development processes involve computational methods like Computational Fluid Dynamics to simulate real world fluid flow and heat transfer phenomena numerically using computers. Also, traditional heuristic design methods require enormous numerical simulations to obtain optimum designs which makes these methods inefficient and highly expensive. This thesis presents a design of experiments-based approach to develop an optimization tool that can predict the optimum performance of a Thermal Energy Storage (TES) system using Computational Fluid Dynamics (CFD) [1], Response Surface Methodology (RSM) [2] [3], and Genetic Algorithm [4]. A cylindrical capsule filled with Phase Change Material (PCM) is used as a Thermal Energy Storage (TES) system or Latent Heat Thermal Energy Storage (LHTES) system, which stores and releases heat as the PCM melts or solidifies respectively. Based of literature review and previous research, Outer Diameter and Wall thickness of the capsule were considered as the possible influencing parameters (i.e. design variables), whereas Melt time of PCM and Weight of cylindrical capsule were considered as the objective functions (output responses). The effects of design variables on the objective functions was studied using the results of CFD simulations and Central Composite Design as the Design of Experiments type. Single and Multi-Objective optimization was performed to find best possible results for the above two objectives in 6 different cases. A robust optimization tool for a thermal battery design was developed which in theory could be applied to any type of heat exchanger design to reduce the overall product development time.