5 August 2025 in Building and Environment
1 December 2016 in AIAA Journal
This study develops a jet-noise-prediction model for chevrons and microjets. A novel equation is proposed to express the amplitude of the fourth-order space-time velocity cross correlations, which represent the sources of noise emanated from unheated jets, in terms of mean-flow parameters and turbulence statistics, such as streamwise circulation, axial velocity, and turbulent kinetic energy. The cross correlations based on a Reynolds-averaged Navier-Stokes flowfield showed a good agreement with those based on a large-eddy-simulation flowfield. With the novel acoustic-source description, there is a good agreement between the model’s jet-noise predictions and the experimental data for unheated jets for a wide range of frequencies and observer angles for both chevrons and microjets. As the model provides quick and accurate jet-noise predictions, a parametric study is performed to understand the impact of chevrons and microjets on jet noise. Chevron penetration is the underpinning factor for jet-noise reduction, and its optimum is found to be around one-seventh of the nozzle diameter. The number of chevrons has a considerable effect on jet noise, and six is found to be an optimum number of chevrons. The injected mass-flow rate of a system of microjets has a noticeable impact on jet noise, and for 18 microjets, its optimum is found to be around 0.0072 of the main-jet mass-flow rate. There is a good agreement between predicted and measured optimum values. This establishes that the model is indeed capable of assessing and optimizing jet-noise-reduction concepts, and could contribute toward the development of quieter nozzles for future aircraft.
26 February 2015 in Experiments in Fluids
1 January 2024 in Environmental Data Science
This article addresses the challenges of assessing pedestrian-level wind conditions in urban environments using a deep learning approach. The influence of large buildings on urban wind patterns has significant implications for thermal comfort, pollutant transport, pedestrian safety, and energy usage. Traditional methods, such as wind tunnel testing, are time-consuming and costly, leading to a growing interest in computational methods like computational fluid dynamics (CFD) simulations. However, CFD still requires a significant time investment for such studies, limiting the available time for design modification prior to lockdown. This study proposes a deep learning surrogate model based on a MLP-mixer architecture to predict mean flow conditions for complex arrays of buildings. The model is trained on a diverse dataset of synthetic geometries and corresponding CFD simulations, demonstrating its effectiveness in capturing intricate wind dynamics. The article discusses the model architecture and data preparation and evaluates its performance qualitatively and quantitatively. Results show promising capabilities in replicating key wind features with a mean error of 0.3 m/s and rarely exceeding 0.75 m/s, making the proposed model a valuable tool for early-stage urban wind modelling.