When: Feb 07 2020 @ 4:00 PM
Where: 213 Hodson Hall
213 Hodson Hall

4:10 pm Presentation
“Locating Scalar Sources from Remote Sensors using Eigen-Responses in Turbulent Environments”
Presented by QI WANG (Adviser: Prof. Zaki)
The ability to identify the location and intensity of a scalar source in turbulent environment from remote measurements is obfuscated by the stochastic dispersion from turbulent eddies and by molecular diffusion. An algorithm is proposed to solve this inverse problem, which relies on estimating the left and right singular vectors of the scalar impulse-response system, or its eigen-sources and eigen-measurements. The projection of the true source onto an eigen-source is proportional to the projection of the sensor signal onto the corresponding eigen-measurement, and the corresponding singular value gives the proportionality. When only the sensor signal is available, the unknown source is identified by minimizing its deviation from this proportionality. A pre-requisite of the algorithm is the knowledge of the eigen-spectrum of the system. Previous eigen-sources can be obtained from historical data. The eigen-sources for the current time horizon can be then approximated using proper orthogonal decomposition of the observation matrix from simulations using previous eigen-sources as trial sources. Furthermore, this approach utilizes only forward simulations and can be easily applied with expanding time horizon of the measurement. We demonstrate that using only five ensemble members, the source location and intensity are predicted with less than 10% error. Furthermore, we quantify the effect of sensor noise by evaluating the standard deviation of the reconstructed source parameters when Gaussian noise is added to the measurement.

4:35 pm Presentation
“RNL Model of Flow over Riblets”
Presented by XIAOWEI ZHU (Adviser: Prof. Gayme)
The restricted nonlinear (RNL) model has been shown to accurately capture the low order statistics, spanwise energy spectra and energy transport in moderate Reynolds number smooth wall-turbulence despite its simplified representation of the dynamics. Here we extend the RNL modeling paradigm to perform computationally efficient simulations of flow over spanwise heterogeneous surfaces, such as riblets. Riblets are spanwise-varying grooves that have been shown to reduce drag by modifying the near-wall flow structure. We first demonstrate that RNL simulations at Re_τ=180 and Re_τ=350 produce results consistent with DNS, specifically that drag reduction can be achieved for riblets with certain peak-to-peak spacings (S^+ < 20). In addition, the RNL simulations are shown to accurately predict the formation and distributions of the secondary flow which causes the ‘breaks down’ of the drag reduction when S^+ > 20. RNL is thus a promising approach to investigate the turbulent drag reduction over riblets (which is usually prohibitively expensive) with vastly reduced computational resources.