Michele profile picture

Michele (MP91)

Tutoring in
Computer Science
Engineering
High School Subjects
Languages
Mathematics

Master of Engineering | Average response time: 2 minutes

Master of Engineering Average response time: 2 minutes

Michele (MP91)

Tutoring in
Computer Science
Engineering
High School Subjects
Languages
Mathematics

Master of Engineering| Average response time: 2 minutes

About Michele

I got my Bachelor's and Master's degrees at Politecnico di Milano, with a Master's thesis at the University of California Irvine. In this first research experience, I worked with Professor Elghobashi, in order to develop an open and close channel flow direct numerical simulation code, starting from their homogeneous isotropic turbulence solver. In order to do so, I started implementing a constant pressure gradient as a forcing term and the boundary conditions. The choice of the initial condition was taken considering the method employed by de Villiers (2006). The flow statistics were compared with Kim et al. (1987), showing a perfect collapse. After the flow validation, the particles were introduced through an immersed boundary method based on Uhlmann (2005). Details
of the method can be found in Lucci et al. (2010). The simulations with different flow roughness were compared with Chan-Braun et al. (2011), showing small differences, mainly due to the employment of different forcing terms for the Navier-Stokes equations in the paper and in our code.

After my Master’s graduation, I started my PhD as a part of the project ”Micro-organisms and turbulence”. The focus of my PhD is the direct numerical simulation of scalar transport in open channel flow. My analysis focused on pollutant and gas transfer. Since atmospheric gases and bacteria usually have very low diffusivities in water, they will tend to create thin, elongated structures in which very steep gradients can be found, requiring extremely fine grid resolutions that made previous direct numerical simulations unfeasible. My research group overcome this problem by employing a fifth-order accurate weighted non-oscillatory (WENO) scheme for scalar convection, combined with a fourth-order accurate central method for scalar diffusion (Kubrak et al. (2013)). Our in-house code is able to perform simulations with parameters closer to reality, at an accessible computational cost. The numerical code uses a central fourth-order finite-difference scheme to solve the flow field and a three-stage Runge-Kutta scheme for the time integration.
The transfer of gases across a gas-liquid interface is a fundamental process in several research fields, from civil engineering to biology. The number of gases exchanged between water and the atmosphere is a key factor for marine life and the balance of greenhouse gases present on our planet. In the last decades, the inadequacy of technologies underlined the inability of experimental apparatus to assess the main factors playing a role in mass transfer phenomena, hindering the development of a unified model, even in simplified experimental set-ups. In contrast, the direct numerical simulation did not allow to reach Schmidt (𝑆𝑐 = 𝜈/𝐷, where 𝜈 is the kinematic viscosity and 𝐷 the molecular diffusivity) and Reynolds numbers (π‘…π‘’πœ = π‘’πœ 𝐻/𝜈, π‘’πœ the friction velocity and 𝐻 the height of the open channel) found in real rivers. In fact, previous direct numerical simulation studies of mass transfer across the air-water interface of open channel flows were limited to small domain sizes and low Schmidt or Reynolds numbers that hinder the formation and meandering of very large scale motions. I performed DNS of low to high diffusivity (4 ≀ 𝑆𝑐 ≀ 200) mass transfer across a clean surface driven by low to moderate turbulent intensity open channel flow (2875 ≀ 𝑅𝑒𝑏 ≀ 12000) in a domain sizeable to capture very large scale motions (up to 24𝐻 ×𝐻 Γ—6𝐻). In order to properly resolve the velocity field, up to 5.10Γ—106 grid points were employed, while, to fully resolve the highest Schmidt and Reynolds number gas transport simulations 1.2 Γ— 10^10 grid points
where needed. The results support the validity of previous scaling laws and existing experimental data obtained for high Reynolds numbers and different flow typologies. The most important models developed were found to work quite well also in open-channel flow with some major limitations. Moreover, the employment of different domain sizes, from 3𝐻 ×𝐻 Γ—3𝐻 to 24𝐻 ×𝐻 Γ—6𝐻, and moderate Reynolds numbers allowed for the analysis of the effects of very large scale motions (VLSM) on interfacial mass transfer.

During my Bachelor's and Master's years, I always tutored students at Casa Dello Studente in Italy. I had to manage several students at the same time and this allowed me to develop a working method to teach and help students of any age, from primary school to high school. After my Master's graduation, I started my PhD where I had to teach the exercise part of Informatics 2, a course on C++ language. During the course, I had to prepare the exercises for every week and the final exams for a class of around 30/40 students. The rating of the classes has always reached the maximum score, showing the satisfaction of the students.

Degrees

Master of Engineering

Aeronautical Engineering

2017

Aerodynamics

Bachelor of Engineering

Aerospace Engineering

2014

Fluid mechanics

Policies

Cancellation Policy: Please provide 24 hour notice if you are cancelling your lesson.
No monthly subscription required.

No recent reviews.
Live Chats