Gradient-Free Aeroacoustic Shape Optimization Using Large Eddy Simulation

arxiv(2023)

引用 0|浏览1
暂无评分
摘要
We present an aeroacoustic shape optimization framework that relies on high-order Flux Reconstruction (FR) spatial discretization, the gradient-free Mesh Adaptive Direct Search (MADS) optimization algorithm, and Large Eddy Simulation (LES). Our parallel implementation ensures that the runtime of each optimization iteration remains consistent, irrespective of the number of design parameters involved in the optimization problem, provided sufficient resources are available. This eliminates the dependence of the runtime of gradient-free algorithms on the number of design variables. The objective is to minimize Sound Pressure Level (SPL) at a near-field observer by computing it directly from the flow field. We evaluate this framework across three different problems. First, an open deep cavity is considered at a free-stream Mach number of $M=0.15$ and Reynolds number of $Re=1500$ based on the cavity's depth, reducing the SPL by $12.86dB$. Next, we optimized tandem cylinders at $Re=1000$ and $M=0.2$, achieving over $11dB$ noise reduction by optimizing cylinder spacing and diameter ratio. Lastly, a baseline NACA0012 airfoil is optimized at $Re=23000$ and $M=0.2$. The shape of the airfoil is optimized to generate a new 4-digit NACA airfoil at an appropriate angle of attack to minimize the SPL while ensuring the baseline time-averaged lift coefficient is maintained and prevent any increase in the baseline time-averaged drag coefficient. The SPL is reduced by $5.66dB$ while the mean drag coefficient is reduced by more than $7\%$. These results highlight the feasibility and effectiveness of our aeroacoustic shape optimization framework.
更多
查看译文
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要