26 years old
Minnesota, US
Research Assistant
University of Minnesota
I, go by Niloo , was born and raised in Zabol, a small city in eastern Iran. In 2021, I earned my Bachelor's degree in Mechanical Engineering with a CGPA of 3.58/4 (16.78/20) from the University of Tehran under the supervision of Dr. Farzad A. Shirazi. During my undergraduate years, I actively engaged in different research groups, exploring various disciplines and expanding my knowledge beyond the core curriculum.
As a Master's student at the University of Tehran, I focused on control theory and data-driven modeling for mechatronic systems, earning a CGPA of 4.00/4.00 (18.30/20). Under the supervision of Dr. Ali Sadighi, my research explored advanced control methods, integrating machine learning techniques to enhance system performance.
Currently, I am a 2nd-year Ph.D. student at the University of Minnesota, where we design and analyze high-speed rotating machinery. Under the guidance of Prof. Eric Severson, my research focuses on the development of bearingless motors and electrodynamic bearings to enhance system efficiency and reduce power electronics dependency, particularly for flywheel energy storage applications.
GPA: 3.85/4
Selected Courses:
Topics in Controls: Design of Rotating Electric Machines (-), Instructor: Prof. E. Severson
Optimal Filtering & Estimation (A), Instructor: Prof. K. Strandjord
Electric Drives (A), Instructor: Prof. S. Raju
GPA: 18.3/20 (4/4)
Selected Courses:
Advanced Control (18/20), Instructor: Dr. F. A. Shirazi;
Digital Control (17/20), Instructor: Dr. A. Sadighi;
Adaptive Control (16.65/20), Instructor: Dr. M. Ayati;
Neural Network (18.68/20), Instructor: Dr. A. Kalhor;
More details: I was ranked 92nd among more than 10,000 participants in the nationwide M. Sc. university entrance exam and was offered a full scholarship.
GPA: 16.78/20 (3.58/4), Last two years GPA: 18.06/20 (3.96/4)
Selected Courses:
Statics (20/20), Instructor: Dr. M. Nazari;
Thermodynamics I (19.55/20), Instructor: Dr. M. Saffaripour;
Measurement Systems and Instrumentation (18/20), Instructor: Dr. F. A. Shirazi;
Circuit and Electric Machines (18.5/20), Instructor: Dr. M. Jahangiri;
Principles of Electronic (20/20), Instructor: Dr. A. Shaeshaani;
Applied Finite Element Method (17.7/20), Instructor: Dr. M. Mahnama;
Optimization of Mechanical Systems (16.52/20), Instructor: Dr. F. Kosari;
Automatic Control (17.3/20), Instructor: Dr. A. Yousefi Koma;
More details: I was ranked 326th among more than 160,000 participants in the nationwide B.Sc. university entrance exam and was offered a full scholarship.
Python for Everybody Specialization, UMICH
Deep Learning Specialization, Deeplearning.AI
Machine Learning, SU
IBM AI Engineering Professional Certificate, IBM
Design of Rotating Electric Machines, UMN
Power Electronics Course, UMN
Identification, Estimation, and Learning, MIT
Control Bootcamp, University of Washington,
Data Driven Modeling and Dynamical Systems, University of Washington
Oct. 2022
Score: 105 (R: 29, L: 30, S: 20, W: 26)
My research focuses on the development of bearingless motors (BM) to enhance system efficiency and optimize the dynamic performance of high-speed rotating machinery. I conduct this research under the supervision of Prof. Eric Severson within the eLev Lab. With a strong background in magnetic bearing technology, particularly in electrodynamic bearings (EDB), my work involves modeling, simulation, and experimental validation of advanced electromechanical systems. As part of this research, I will be working with the Advanced Motor Drive Controller (AMDC) to implement real-time control strategies and enhance system performance.
My M.Sc. thesis under the supervision of Dr. Ali Sadighi used Dynamic Mode Decomposition (DMD) method to identify the underlying equations of nonlinear systems, detect any changes, and use adaptive strategies to control the system's motion. In the first part of this project, I identified the system using the DMDc method, an extension of DMD for systems with actuation. It is the base of the second part and has great significance. In the second section, I used the developed model to track system changes, apply appropriate control commands and improve the system's performance.
I completed my B.Sc. thesis under Dr. Farzad A. Shirazi’s supervision. In this thesis, I developed a controller to reduce motion error and fine motion tracking of a compliant mechanism.
This mechanism acts non-linearly; therefore, it is critical to reducing the error of mobility and fine motion tracking. At first, the PSO algorithm was used to determine the optimal parameters of the hysteresis of piezoelectric based on Bouc-Wen's equations. Then, an inverse feedforward controller is utilized alongside a PI backward controller to improve the mechanism motion.
As the first author, I published the results of this project in the Journal of Mechanical Engineering University of Tabriz, 86 (41), 2021.
You can see related paper
here.
Abstract — This work investigates bearingless motors configured to function as radial electrodynamic bearings (EDBs), focusing on the feasibility and effectiveness of different stator winding configurations and bearingless motors (BMs). A theoretical framework based on short-circuiting the suspension winding is developed to calculate shaft forces during eccentricity. This model integrates all possible radial flux bearingless motor configurations with pₛ = p ± 1 or pₛ = 1 suspension pole-pairs in a unified manner that incorporates their electromechanical properties. The paper uses this model to analyze stability, investigate the systems’ performance potential, and inform design decisions. It is shown that each configuration (BM type, winding type, polepair combination) offers distinct features and shortcomings when evaluated as an EDB that should be considered in the design process.
You can find this paper here.
In this study, a novel approach based on neural network surrogate
modeling was developed to improve the design time of compliant mechanisms. The traditional design methods are very time-consuming and principally involve challenging FEM
simulations, which makes the design process challenging. In this approach, two neural networks
were trained to estimate the design’s ultimate goals based on the geometric characteristics of
the mechanism. The PSO algorithm is then used to find the best combination of geometric
parameters.
You can find this paper here.
In my senior year, with the advice of Dr. Maryam Mahnama, I examined an optimization process based on statistical analysis and numerical simulation and then explored an ideal design.
The completed model and its vibrational analysis were presented in a paper published at the 10th
Iranian Society of Acoustic and Vibration Conference (ISAV-1063), Tehran, 2021.
You can find this paper here.
This course deepened my understanding of systems and the related theories to work with them.
The prototype of a solar tracking system was the team's final project. A schematic of the mechanism was initially developed in SolidWorks based on the expected results, and the mechanism was then assembled together. The mechanism was tested to various strategies in the final step using an Arduino Uno, and the approach with the best performance was chosen.
The adaptive control course was thrilling course for me. Besides working with various control strategies, I got familiar with a variety of offline and online methods for identifying systems and their applications. I designed adaptive controllers like Deterministic self-tuning regulators, Stochastic and Predictive self-tuning regulators,
Model reference adaptive systems, and optimal controllers like MPC, LQR, and LQG for my mini-course projects in MATLAB. For my final project, I developed an MRAC for a hydraulic underwater manipulator using the MIT rule in MATLAB.
In this course, I performed various projects in Python. Some of projects are:
Supervised and Unsupervised Learning, Classification, CNN (Bird Recognition in the City of Peacetopia (Case Study)) , RNN (Jazz Improvisation with LSTM), GAN (Art Generation with Neural Style Transfer), YOLO (Car detection),
Anomaly Detection, Image Segmentation (Face Recognition), Autonomous Driving (Case Study), Transfer Learning (MobileNet), Image Segmentation (U-Net), Operations on Word Vectors - Debiasing, Emojify, Neural Machine Translation, Trigger Word Detection, Transformers Architecture with TensorFlow.
In my final project, I designed P and PI controller for the QUANSER 3-DoF helicopter. In the first section, the mathematic relationships of this helicopter were derived and then linearized. In the second step, various control gains were applied to reach the desired output, and their performances were compared.
I did two primary projects in this course. In the first project, a turbine cycle in MATLAB was modeled, then optimized using both Lagrange multiplier and Generalized Reduced Gradient Algorithms. In the second project, the mathematical model of a heating system was derived, then optimized for two objectives with the Non-Dominant Sorting Genetic Algorithm in MATLAB.
This course deepened my understanding of systems and the related theories to work with them. In my final project, I designed microcontroller-based modeling for a digital thermocouple using Arduino software.
Feel free to contact me if you have any questions or would like to discuss anything.