Home

ORCID

0000-0002-4031-8457

ResearcherID

C-3012-2013

Google Scholar

XiKya3AAAAAJ

Twitter

@PNadukandi

GitHub

nadukandi

Contact:
Advanced Mathematics
Repsol Technology Lab
Paseo de Extremadura, Km 18
Móstoles, Madrid, Spain
p[DOT]nadukandi[AT]repsol[DOT]com

NEWS: New open access paper "Computing the Wave-Kernel Matrix Functions" (with Nicholas J. Higham) published in the SIAM Journal on Scientific Computing. The MATLAB software and the raw data are available at my GitHub repository.

I am a Technical Advisor in Advanced Mathematics at the Repsol Technology Lab, Madrid, Spain. I contribute as a specialist in numerical methods for scientific computing. Currently, I work at the interface of computational mechanics and machine learning for applications in the oil and energy industry. Additionally, I lead the communication activities of the Advanced Mathematics discipline.

Before joining Repsol I worked nearly a decade in academics. From Sep-2016 to Aug-2018 I held a Marie Skłodowska-Curie Individual Fellowship at the School of Mathematics, University of Manchester, UK. Prior to that, I was an Assistant Research Professor at CIMNE Barcelona, Spain.

I graduated with a B.Tech. degree in civil engineering from IIT Guwahati, India and a Ph.D. degree in computational mechanics from UPC (BarcelonaTech), Spain. I am a native of Kerala state in southern India.

My MSCA project (FastFlowSim) focused on particle-based exponential integrators for the simulation of low viscosity incompressible flows and its efficient implementation in software. My approach is distinct from peers/colleagues in the sense that I employ analytical solutions to compute the trajectories of the particles driven by velocity and acceleration fields approximated by piecewise linear interpolants. The analytical solutions are expressed using the wave-kernel matrix functions—the numerically stable and efficient computation of which is a specific goal.

The solution to systems of second-order ordinary differential equations can be written succinctly using the wave-kernel matrix functions. These matrix functions provide a starting point from which to develop trigonometric time integrators for stiff second-order systems, to compute waves on graphs, and to analyze and control asymmetric systems prone to non-decaying vibrations.

In the past I have proposed, among other works, novel Petrov–Galerkin FEM for the numerical simulation of the convection–diffusion–reaction equation and the Helmholtz equation.

The following poem, which I learnt at school, has motivated the approach taken in my activities till date:

Climb every mountain, ford every stream,
Follow every rainbow, till you find your dream.
A dream that will take all the love that you can give,
All the days of your life as long as you live.