A Ph.D. student in fluid mechanics

I completed my master's degree in Mechanical engineering at Sari Azad University. Then I was accepted in the Ph.D. entrance exam in mechanical engineering at Babol Nooshirvani technical university. Afterwards, I started studying microclimate in urban canyon streets numerically and tried to use my background knowledge about wind catchers and turbulence for enhancing natural ventilation in urban canyons. During that time, I started writing three papers which one of them was prepared to submit but as I was omitted from the research group, I couldn't complete it. It was about passive cooling scenarios of a microclimate in NYC/ USA. "Passive cooling '' here defined as using no electrical power and just by the means of vegetation and its evaporative effect, we want to quantify how vegetation and its arrangement can affect the humidity and temperature of the microclimate. It was an interesting experience because I learned about DoE (design of experiment) and also used software such as Envi-Met (atmospheric modeling), grasshopper ladybug (shadow calculation), and Minitab (DOE). The two other paper has similar subjects, one of them was about the cooling effect of parks in microclimates and the effect of prevailing wind speed, and the second was about how a city (or even miniclimate) should be developed to can gain thermal comfort condition. I believe that as outdoor temperature decreases, energy demand for cooling indoor areas decreases too (several papers have confirmed it and quantified it through field experiments). In other words, the purpose of these papers was to reduce the greenhouse effect and effort for standing against global warming. I attended ICEM-2018 in Istanbul/Turkey as an oral presenter of my paper titled: "Evaluating how turbulent flow affects heat transfer efficiency in a wind catcher". That was a great experience for me to get familiar with fluid mechanic professors from all over the world. Some of them encouraged me to continue researching in this field and pushed me to study Ph.D. It was inspiring to hear something and it empowered my passion for becoming a CFD specialist. Becoming a CFD specialist in this field is my long-term goal and I’ll do whatever is needed to achieve that. After that, I tried to read any book I found about CFD and turbulence. Finite volume and finite difference methods are my favorite approach in CFD and I wish to develop an efficient DNS (Direct numerical simulation) code for parallel computing wake flow of buildings array. I think using the efficient DNS approach in cooperation with some reduced-order models (such as POD) could help decrease the computation cost of such simulations. Since I always want to progress day and day, I have tried to take part in some of the related courses, workshops, and conferences with CFD. For example, in 2019, I attended an online course about "Open-Foam", "Large Eddy simulation fundamentals" and "POD (proper orthogonal decomposition) ", then I made a huge effort to learn Python and C++ on my own for developing CFD codes. Also, I'm interested in Machine Learning and I think my knowledge in Python would be helpful in the implantation of Machine learning codes. After that, I started a DL (deep learning) course online which is in progress. In the field of applied mechanics, in my opinion, the most interesting topics are those that can create a connection between new theories or approaches with demanding applications. For example, the appropriate orthogonal decomposition method, or any other effective order reduction method, which can reduce the computational costs of turbulent flow, can be used for large-scale simulations such as atmospheric incompressible flow (in the cities). The central theme of POD is actually on the technical treatment for effective methods that can generate numerical solutions for time-dependent PDEs involving only a small set of data, but yield decent solutions that are accurate and suitable for applications. It reduces data storage, CPU time, and, especially, computational complexity – several orders of magnitude. The key idea and methodology in proper orthogonal decomposition (POD), from properties of eigensolutions to a problem involving a large data set. In such cases, the Navier-Stokes equations can be discretized with the POD approach, and whether the finite volume or finite difference method is used, a direct numerical simulation can be performed at this scale. It is also possible to develop an open-source code using Python or C++ so that it can be used for other cases. I think the subject of turbulence is a beautiful and nebulous phenomenon thus this is the reason why I am very interested in it. One of my goals is to develop an efficient DNS-based turbulence model to provide more accurate predictions of incompressible flow.