About Me
I am Sahar Hemmati, a researcher and engineer with expertise in control systems, machine learning, and optimization. My academic journey includes a Master’s in Modeling, Simulation and Control, where I focused on developing data-driven models and exploring Physics-Informed Neural Networks (PINNs) for complex systems. With hands-on experience in process design and CFD simulations, I am passionate about bridging theory and practice to create innovative solutions for industrial challenges. My research aims to contribute meaningfully to sustainable technologies and interdisciplinary advancements. From developing efficient algorithms to designing physics-based models, I enjoy unraveling complex problems and finding elegant, practical solutions. Specially, My love for mathematics, linear algebra, and numerical methods has driven me to explore innovative solutions at the intersection of science, engineering, and computation.
This website is a space to share my journey, insights, and projects in fields like computational science and engineering design. Whether you’re a fellow enthusiast or just curious about these topics, I hope you find something inspiring here!
Education
2018-2021
M.Sc. in Modeling, Simulation, and Control, Sharif University of Technology, Tehran, Iran
• Thesis topic: Development of low-order model/controllers for non-linear systems
Keywords: Data-Driven Modeling, Model Order Reduction (MOR), Machine Learning
2012-2016
B.Sc. in Chemical Engineering (Process design), Azad University South Tehran branch (Top Branch), Tehran, Iran. GPA: 18.25/20
• Project topic: Study on how to produce Li-ion Paper battery
Keywords: CNT(Carbon NanoTube), Degradation, time-dependent simulation
Research Interests
• Machine Learning and Data Analysis in Industrial Processes
• Analysis, Identification, and Control of Nonlinear Processes
• Modeling, Simulation, and Optimization of Novel Processes
• Applications of Model Order Reduction in Modeling and Simulation and Control
• Designing Neural Network Models and Controllers
• Calculations of Environmental Processes and Fault diagnostics
• Applications of KPI for Industrial Unit Management
Academic Projects
• Rigorous Dynamic Simulation and Control of Distillation Column
using MATLAB for implementation and Aspen Plus Dynamics to verify the results.
• Design, Simulation, and Evaluation of Different CFD Models (based on Bird and Deen Textbook) in COMSOL Multiphysicssing
• Designed an Expert System based on SQL Database (Microsoft SQL) with GUI Implemented in C# Programming Language by Windows Form
The application helps users to design and perform fault diagnosis for two-phase separators by asking minimum required questions.
• Designed a Fuzzy Controller and Globally Linearizing Control (GLC) to Control a CSTR
• Adaptive Control of Pure-Feedback systems with full state constraint
Barrier Lyapunov function-based command filtered output feedback control for full state-constrained nonlinear systems
• Train a Neural Network as a Surrogate Model for MPC Controller for a CSTR
• Stabilization of a CSTR with three arbitrarily switching modes using Modal State Feedback Linearization (FF control) and Robust Control
• Design and Computational modeling of the Multicomponent process
• Cost Evaluation and Feasibility Study (FS) of Process Plants
• Creation of a platform for Key Parameter Indicator (KPI) on real-time data
Publications
Data-Driven PINN for Thermal Dynamics Modeling: Applications in Heating Rod Temperature Control and Complex System Analysis
Applications of Data-Driven Model Order Reduction in Modeling, Optimization, and Control
Honours & Awards
“Ranked 1st in the Chemical Engineering B.Sc. program among 75 students, averaging 18.25/20 (A+).”
“Graduated from the First-Ranked University of Iran in Engineering, Sharif University of Technology (Top 150 worldwide in Engineering)”
Skills
Programming:
• Professional in MATLAB/Simulink/App Designer
• C/C++, C#, Windows Form, Python, Fortran
• PLC S7 1200, TIA Portal
Database:
• SQL-Based Database (MySQL, Microsoft SQL)
Software:
• Aspen Plus/Dynamics, Aspen HYSYS, Aspen EDR
• COMSOL Multiphysics
• AVEVA diagrams, AutoCad
Contact
Posts
-
Deep Generative Modeling
Artificial intelligence (AI) involves more than just decision-making; it encompasses understanding and generating data. Traditional discriminative models, tasked with classifying data into distinct categories, face challenges when noise is present,…
-
Multiphase Flow Patterns Identification
Introduction: Identifying multiphase flow patterns is crucial across various industries, such as pipeline transport, reactor safety, electronics cooling, and medical diagnostics. Multiphase flows consist of distinct phases: liquid, gas, or…