My story
I am a data scientist with expertise in medical imaging and data processing, artificial intelligence, and computational fluid dynamics. I have a track record of successfully developing algorithms, conducting research, and communicating findings to technical and non-technical audiences.
Video resume
My Expertise
Hard Skills
Python
100%Matlab
100%COMSOL Multiphysics
90%Ansys Fluent
80%Solid Works
70%Microsoft Office
100%Publications
1. Estimating Aerodynamic Data via Supervised Learning
Supervised learning extracts a relationship between the input and the output from a training dataset. We consider four models – Support Vector Machine, Random Forest, Gradient Boost, and K-Nearest Neighbor – and employ them on data pertaining to airfoils in two different cases. First, given data about several different airfoil configurations, our objective is to predict the aerodynamic coefficients of a new airfoil at different angles of attack. Second, we seek to investigate how the coefficients can be estimated for a specific airfoil if the Reynolds number dramatically changes. It is our finding that the Random Forest and the Gradient Boost show promising performance in both the scenarios.
2. Airfoil Generation using GAN and Optimization using Model Order Reduction (In preperation)
The use of computational simulation to scan many alternative designs is valuable, but the number of possible design variations is too large for complete evaluation. To find the true optimum shape, the ultimate goal of computational simulation methods should not just be the analysis of prescribed shapes, but the determination of the best design. The design of aircraft wings is a highly iterative optimization process that requires a large volume of computations and the use of shape and surface parameterization schemes. Wing Shape Optimization (WSO) is constructed through Computational Fluid Dynamics (CFD) where the change of the geometric shape largely depends on Free Form Deformation (FFD). The complex system demands huge computational resources and produces a large system of equations. Model Order Reduction (MOR) technique is used to reduce computational time and complexities of the system. This project proposes using different MOR techniques to optimize the wing's shape efficiently and increase the Lift-to-Drag (L/D) ratio, while examining the effects of having winglets.
Projects
1. Model Order Reduction for Aircraft Wing Shape Optimization (CTRG-21/SEPS/15)
Air transportation is one of the globally popular ways of traveling whose reputation is gradually increasing among the mass people of developing countries like Bangladesh due to the high congestion and accident-prone in the roadway and inefficiency and sluggishness of the railway and waterway. It has been quoted in [1] that in the last 3 years, the increment of the usage of air traffic for traveling by the passengers of our country is an average 8% whereas the phenomenal growth is observed about 13% in the cargo sector. As a result, the necessity of the enhancement of the efficiency of the air vehicles is becoming demandable day by day what is largely dependent on the optimization of the air wings for ensuring smooth movement through the air layer working against aerodynamic drag. Since our proposal is directed at prescribing a optimized shape of the aircraft wings, the aircraft efficiency must be increased and may have an impact on the air transportation system more lucrative to the futuristic passengers.
2. Development of the Efficient Algorithms to optimize of the Solar Thermal state of the Photovoltaic Panel by analyzing the generated dynamic mathematical model
The usage of solar electric energy is increasing, and Photo-Voltaic (PV) panels are becoming popular. However, like other semiconductor devices, PV cells are sensitive to temperature. The overheating of PV panels causes a decrease in voltage and energy storage capacity. Various techniques have been proposed for thermal management, but they have limitations. This project proposes a mathematical model of the PV panel based on the state-space model, which focuses on identifying and minimizing hot spots in the panel. The model will be modified to preserve stability and optimize efficiency. The best arrangements will be selected after minimizing hot spots of the model, whose work efficiency will be less deteriorated in high temperature comparing with other existing procedures.
Accomplishments
1. Dutch-Bangla Bank Scholarship
The Dutch-Bangla Bank Scholarship program aims to offer financial aid to deserving and talented students who have exhibited exceptional academic performance in their Secondary School Certificate (SSC) exams. In 2013, I was selected as a recipient of this scholarship, acknowledging my outstanding academic achievements and potential to excel in higher education. The scholarship provided me with the opportunity to pursue my education without financial constraints for a period of two years. This program is a testament to the Dutch-Bangla Bank's commitment to recognizing and supporting meritorious students in need and helping them realize their academic aspirations..
2. North South University Scholarship
In 2017, while I was pursuing my undergraduate degree, I was granted a scholarship which covered 50% of my tuition fees. Then, in 2019, I was able to secure an extension to the scholarship, which ultimately covered the entire cost of my tuition fees, equivalent to a 100% scholarship.