Venue: 2022 South Thayer Building
The MICDE PhD Student Seminar Series showcases the research of students in the Ph.D. in Scientific Computing. These events are open to the public, but we ask that you register to attend the seminar. If you have any questions, please email email@example.com.
Urban air mobility (UAM) vehicles have taken form as advanced rotorcraft with sets of wings, rotors, canards, and other appendages. Noise generation is an important technical barrier that must be addressed to prevent these vehicles from causing excessive disturbance to the communities they are intended to service. To understand the noise these vehicles generate, and to develop designs that can minimize disturbance, there is a need for analysis and optimization tools specifically for the conceptual design and sizing phase of urban air mobility vehicle development. Such tools must be computationally efficient to allow for the repeated analyses needed for design optimization. This presentation will review the work being carried out at the University of Michigan, coupling aerodynamic, structural, and aeroacoustic disciplines within the multidisciplinary gradient-based design optimization framework OpenMDAO. While aerostructural optimization has been performed previously, coupling with aeroacoustics is challenging given the requirement for time accurate simulations and the associated computational cost of such analyses. By leveraging multiple model fidelities and utilizing efficient gradient calculation techniques, such as the adjoint method and algorithmic differentiation, these disciplines can be formulated into an optimization framework that can be applied to UAM vehicle designs. This presentation will review the work completed to date, including preliminary results, and expand on the future goals of the project, working towards a broader optimization framework for rotorcraft vehicle design optimization.
Bernardo Pacini, Ph.D. candidate in Mechanical Engineering and Scientific Computing
Bernardo Pacini is a Ph.D. Candidate at the University of Michigan focusing his research on aerodynamic, structural, and aeroacoustic modeling for multidisciplinary design optimization of urban air mobility vehicles. He is a member of the Multidisciplinary Design Optimization Laboratory led by Professor Joaquim R. R. A. Martins and of the Computational Aerosciences Laboratory led by Professor Karthik Duraisamy. Bernardo’s work to date is on developing an aero-structural-acoustic analysis framework that can be implemented within the multidisciplinary design optimization process for rotorcraft and urban air mobility vehicle design.
With a growing interest in probabilistic performance assessments of building systems subjected to wind loads, there is a demand for accurately representing building-specific wind loads, considering their non-Gaussian and non-stationary features. While typical wind tunnel data collected for a set of discrete wind directions provide a single realization of stationary pressures, there is currently no experimentally validated model for the stochastic simulation of non-Gaussian and non-stationary wind pressures that can be calibrated to wind tunnel datasets. Such a model is essential for simulating building aerodynamics, especially the stochastic, path-dependent responses associated with time-varying wind speed and direction experienced by a building during hurricane wind events. This talk will review a recently developed theoretical formulation for generating these stochastic pressures. Through carefully designed tests conducted at the University of Florida wind tunnel facility, the formulation was extensively validated with respect to its ability to capture trends over time, occurrences of peaks, and time-varying frequency content.
Srinivasan Arunachalam, Ph.D. candidate in Civil and Environmental Engineering and Scientific Computing
Srinivasan Arunachalam is a PhD candidate in Civil and Environmental Engineering. His research interests lie in uncertainty quantification and understanding the inelastic behavior of wind-excited structures. He is excited about the algorithmic developments that enable efficient reliability assessments, as well as the evolving insights into the physics of extreme responses and their implications for structural design.