Employers need engineers capable of leveraging CFD simulations to make intelligent design decisions, but undergraduate computational fluid dynamics (CFD) courses are not adequately preparing students for this type of work. CFD courses commonly familiarize students with topics, such as method derivation, domain creation, boundary conditions, mesh convergence, turbulence models, numerical convergence, and error analysis. This approach is an effective way to teach novices how CFD software works and how to prepare CFD analyses. However, it neglects development of higher level CFD skills and intuition important to engineering analysis and design, deferring this task to future study and training. This paper introduces the “Machine Learning Driven Interpretation of Fluid Dynamics Simulations to Develop Student Intuition” (MIFoS) software, a program designed to help CFD novices develop the high‐level skills and intuition that employers need in their engineers. A data‐driven approach was used to create the MIFoS software, which allows the submission of arbitrary geometries, automates an external flow simulation, and returns expert‐level graphical interpretation of simulation data. MIFoS's automated CFD simulation and feedback space allows novices to experiment with expert‐level suggestions on their own designs, enabling the skill and intuition development typically gained through years of study, practice, and expert guidance.

No abstract

Engineers use three tools for analyzing fluid flows: analytical, experimental, and numerical. Thorough analyses involve all three tools to inform design decisions or postmortem investigations. However, few undergraduate fluid mechanics lab experiences expose students to the simultaneous application of these tools to a common problem. In this paper, we discuss a set of exercises and lab activities spanning multiple weeks to study a cylinder in a crossflow, providing students with their first experience in the use and comparison of analytical, experimental, and numerical tools for studying a fluid flow. The analytical approach involves the solution derived for potential (inviscid) flow around a cylinder. While no novel examples or analytical approaches are used to teach this portion of the experience, this idealized flow solution is used as a comparison for experimental and numerical lab activities. The experimental approach was initially designed to leverage low-cost microcontrollers and MEMS-based pressure sensors while reinforcing mechatronics and related programming content used in previous courses. Over four years of use and iteration, it has evolved from instructors supplying only a C++ library for the pressure sensors, a wiring diagram, and a rough programming framework to a simplified MATLAB interface that abstracts away many of the low-level commands required to complete the lab exercise. This range of interfaces and experiences allows instructors flexibility in their own implementation and lab goals. The latest iteration includes the use of a hot-wire anemometer, which must be calibrated by students, and allows for data capture and analysis of the resulting vortex street formation and Strouhal number. The numerical approach uses a transient 2D planar analysis for flow around a cylinder and leverages a commercial computational fluid dynamics software, Ansys, for which there are free student versions. Students are provided with an instructor-built flow domain and mesh capable of providing a mesh-independent solution. Students choose solvers, set boundary conditions, run simulations, and process resulting data to compare with analytical and experimental results. A transient analysis is used to capture the effects of vortex formation on the coefficient of lift and to calculate and compare the system’s Strouhal number with that found from the physical experiment. Students are challenged throughout the experience, from completing classic problems such as the lift force on a Quonset hut, to automating sensor positioning and data collection, to simulating a transient fluid flow. A concerted analysis of this relatively simple geometry affords students the opportunity to use analytical, experimental, and numerical tools to compare drag forces, pressure distributions, and transient flow behaviors.