Mechanical and Civil Engineering Seminar

Mechanical and Civil Engineering Seminar Series

Title: Advances in Decomposition Techniques for Turbulent Jet Analysis

Abstract: Turbulent flows are high-dimensional systems characterized by instabilities, nonlinearity, and randomness. Within the chaotic motions, these flows exhibit structures that are coherent in space and time. Further insights into coherent structures can aid our understanding of turbulence and help develop reduced-order models for flow control. My work involves developing data-driven techniques for identifying the dynamically relevant structures and using these techniques to investigate the flow physics of turbulent jets.

The first part of my talk will encompass our contributions to spectral modal decomposition techniques. Spectral proper orthogonal decomposition (SPOD) is the frequency domain variant of principal component analysis, and it computes modes that are coherent in space and time. In recent years, SPOD has emerged as a major tool for analyzing turbulent flows. We extend SPOD for low-rank reconstruction and frequency time analysis. A convolution-based strategy is proposed for frequency-time analysis that characterizes the intermittency of spatially coherent flow structures. Next, bispectral mode decomposition (BMD), a technique that extracts flow structures associated with nonlinear triadic interactions by optimizing the third-order statistics, is presented. These techniques are applied to high-fidelity numerical simulations in the second part of the talk. Large-eddy simulations of two unforced and four forced jets at a Reynolds number of 50,000 were performed and validated against companion experiments. In a transitional jet, the emphasis of our analysis is on characterizing vortex pairing and the associated energy transfer. The SPOD-based spectral energy budget reveals that the energy flows from the fundamental to its subharmonic, resulting in the growth of the subharmonic frequency. Finally, forced turbulent jets are investigated using BMD to analyze the triads created due to the sum and difference interactions between the fundamental forcing and its harmonics.

Bio: Akhil Nekkanti is a postdoctoral scholar at Caltech, working with Prof. Tim Colonius on multiphase and cavitating flows. He received his Ph.D. from the University of California San Diego, under the guidance of Prof. Oliver Schmidt, with a focus on the hydrodynamics and noise-generating mechanisms of turbulent jets. The motivation of his work is to uncover the physical phenomena of turbulent flows. His research involves conducting high-fidelity numerical simulations and developing modal decomposition techniques. He received the ‘Outstanding Graduate Student in MAE, 2023' award at UCSD, and his paper was selected as one of the best papers at the 12th International Symposium on Turbulent Shear Flow Phenomena.

NOTE: At this time, in-person Mechanical and Civil Engineering Lectures are open to all Caltech students/staff/faculty/visitors.