Indian Institute of Technology Madras

About: Indian Institute of Technology Madras is a facility organization based out in Chennai, Tamil Nadu, India. It is known for research contribution in the topics: Catalysis & Heat transfer. The organization has 20118 authors who have published 36499 publications receiving 590447 citations.

Heuristic algorithms for continuous flow-shop problem

Abstract: In this article we present two heuristic algorithms for scheduling in the constrained or continuous flow shop to minimize total flow time of jobs. Two heuristic preference relations are used as the basis for job insertion to build up a schedule by the heuristics. When evaluated over a large number of problems of various sizes, the heuristics are found to be very effective in yielding near-optimal solutions.

Evaluation of circularity from coordinate and form data using computational geometric techniques

Abstract: Data for evaluating circularity error can be obtained from coordinate measuring machines or form measuring instruments. In this article, appropriate methods based on computational geometric techniques have been developed to deal with coordinate measurement data and form data. The computational geometric concepts of convex hulls are used, and a new heuristic algorithm is suggested to arrive at the inner hull. Equi-Distant (Voronoi) and newly proposed Equi-Angular diagrams are employed for establishing the assessment features under different conditions. The algorithms developed in this article are implemented and validated with the simulated data and the data available in the literature.

Embedding Deep Learning in Inverse Scattering Problems

Abstract: In this paper, we introduce a deep-learning-based framework to solve electromagnetic inverse scattering problems. This framework builds on and extends the capabilities of existing physics-based inversion algorithms. These algorithms, such as the contrast source inversion, subspace-optimization method, and their variants face a problem of getting trapped in false local minima when recovering objects with high permittivity. We propose a novel convolutional neural network architecture, termed the contrast source network, that learns the noise space components of the radiation operator. Together with the signal space components directly estimated from the data, we iteratively refine the solution and show convergence to the correct solution in cases where traditional techniques fail without any significant increase in computational time. We also propose a novel multiresolution strategy that helps in producing high resolution solutions without any significant increase in computational costs. Through extensive numerical experiments, we demonstrate the ability to recover high permittivity objects that include homogeneous, heterogeneous, and lossy scatterers.