Shape Optimization of Fluid Flow Ducting Using a Shape Function, a Search Method and CFD: Some Early Results

Professor Sreenivas Jayanti, IIT Madras

Professor Geoff Hewitt Lecture, 2015 PhD Research Symposium

Abstract

Computational fluid dynamics (CFD) provides an excellent platform for flow analysis but it fares poorly as a design too. A case in point is flow branching or flow split at a junction which is often encountered in many engineering and industrial applications. A naïve approach of CFD treats this as a rating problem, that is, specify the position of a guide plate and find the flow distribution for a given pressure drop. Finding the optimal shape of the junction which would automatically divide the flow into the required streams is a much tougher design problem. We present a shape optimization method which can arrive at the optimal shape of the junction which will ensure the desired flow split without the need for a guide vane or a valve. The method uses Bézier curves to describe the bounding surfaces of the junction and Box’s complex method as a search engine to arrive at the right shape of the junction. Extension of the method to the design of a manifold is discussed along with some improvements to speed up the convergence of the algorithm.

Biography

Professor Sreenivas Jayanti studied mechanical engineering at IIT-BHU, Varanasi, India; nuclear engineering at Ohio State University, Columbus, Ohio, USA; fluid mechanics at INPG, Grenoble, France and obtained his PhD from the Department of Chemical Engineering at Imperial College, London, UK in 1990. After a post-doctoral fellowship at Imperial College, he joined IIT Madras as a visiting faculty in 1994. He is currently a professor in the department of chemical engineering at IIT Madras. His main research interests include computational fluid dynamics, combustion and fuel cells.

This is the final seminar in the 2015 Department of Chemical Engineering Distinguished Seminar Series. It is the Professor Geoff Hewitt Lecture and the opening lecture of the 2015 Chemical Engineering PhD Symposium.