Assistant Electrical Engineering

Elaheh Noursadeghi Research Assistant, University of Massachusetts Lowell

**** ******** **, *** *, Lowell, MA, 01851 857-***-**** ******.***********@*****.*** Summary Eight years of experience in research and modeling of control of dynamic systems, estimation, ma- chine learning. Produced novel results in distributed fault diagnosis of nonlinear stochastic systems, distributed estimation, fault identi cation of multi-robot systems, and plume source localization. Technical Skills:

Fault Diagnosis and Prognosis

Estimation, Signal Processing, Particle Filtering, Kalman Filtering

Machine Learning, Active Learning (Arti cial Neural Networks, Deep Learning, Support Vector Machines, Clustering, Bayesian Networks, Reinforcement Learning)

Control (MIMO, Nonlinear, Adaptive, Robust, Model Predictive)

Autonomous and Distributed Systems (Consensus, Distributed Average Tracking, Distributed Sensor Fusion, Distributed Optimization, Graph Theory)

Mathematical Modeling, Analysis and Simulation of Real Systems/States Education University of Massachusetts Lowell Expected May 2017 Ph.D., Mechanical Engineering, GPA: 3.825

Amirkabir University of Technology 2012

M.S., Electrical Engineering GPA: 3.7

Isfahan University of Technology 2009

B.S., Electrical Engineering GPA: 3.7

Computer Programming: MATLAB, Simulink, C/C++, Arduino, PLC, PCS7, familiarity with Python and Skills Java

Tools and Libraries: OrCAD, PSpice, LaTeX, Microsoft O ce Operating Systems: Windows, MacOS, Linux

Experiences

Boston College

Visiting Researcher, Computer Science Department May 2016-Sept. 2016

Working on active learning for dynamic systems. In active learning, labeling all samples is ex- pensive and time-consuming. The objectives were: i) how we label as few samples as possible without hurting performance and ii) if it matters waiting for longer or not. University of Massachusetts Lowell

Research Assistant, Autonomous Robotics Systems Laboratory Sept. 2013-Present

Distributed Full-order Fault Diagnosis of Nonlinear Complex Stochastic Systems: Develop a full-order distributed particle ltering-based algorithm where the entire state vector and failure modes are estimated at each detector node based only on its local observation. After computing the state estimates and the probability of failure modes locally, the local estimates and probabilities are fused through interactions between local neighborhoods in a distributed fashion to form the global fault decision.

Distributed Reduced-order Fault diagnosis of Large-scale Physical Systems: Develop a distributed reduced-order particle ltering-based algorithm. In this approach, the large system is decomposed into several interconnected subsystems. A subset of state variables and probabilities of failure modes is computed at each diagnoser node based on the local measurements and the information transmitted from the neighboring nodes.

Distributed Fault Detection and Identi cation Filter Design for a Network of Multi- agent Systems:

1

Development and design of fault detection and identi cation lters for a network of heterogeneous multiagent systems with di erent dynamics and order from one another. The proposed approach is capable of detecting and estimating simultaneous faults. Moreover, each agent can detect and isolate not only its own faults but also the faults of its nearest neighbors.

Plume Source Localization:

Develop an algorithm for detecting and tracking chemical plumes in distributed sensor networks. Given unknown sources released at unknown times, the sensor network must estimate the location of sources.

Teaching Assistant Sept. 2013-Present

Prepared and documented materials for \Dynamics", \Robotics", \Fundamental Solar Energy", and \Green Energy" courses.

Amirkabir University of Technology

Research Assistant, Real-time Systems Laboratory Sept. 2009-Feb. 2012

Defect Root Cause Analysis in Production Lines based on Bayesian Network: Design an expert system for defect root cause analysis of vehicle body surfaces in paint shop of car factory using image processing and Bayesian networks.

Prepared benchmark for large-scale system class with title \Control problem for seismic response of cable-stayed bridges".

Teaching Assistant Sept. 2010-Sept. 2011

Coordinated and documented materials for \Model Predictive Control" course to be used as a guidance for future teaching assistants.

Conducted and organized \Linear Control Systems Laboratory" for approximately 45 students.

Conceptualized Real-Time Workshop Toolbox in MATLAB.

Designed PID and lead-lag controllers in S, Z, and W planes and validated the designed controllers in MATLAB/Simulink, LabVIEW.

Applied the control algorithms such as PID to di erent systems including a robotic arm, a DC servo motor, and an inverted pendulum to improve their transient and steady state responses. Isfahan University of Technology

Internship, Digital Control Laboratory Summer 2009

Research on standards of industrial Ethernet, Pro net and Pro bus. Research Assistant Sept. 2008-June 2009

Designed an adaptive fuzzy-based controller for a tra c lights network to manage tra c on an urban area.

Publications

E. Noursadeghi and I.A. Raptis. A Particle-Filtering Based Approach for Distributed Fault Diag- nosis of Large-Scale Interconnected Nonlinear Systems. IEEE Transactions on Systems, Man, and Cybernetics, Submitted.

E. Noursadeghi and I.A. Raptis. A Particle Filtering-based Approach for Distributed Fault Diag- nosis and Estimation of Multi-Robot Systems. in Proceedings of the ASME Dynamics Systems and Control Conference, Minneapolis, Minnesota, October 2016, in Print. E. Noursadeghi and I.A. Raptis. Full-Order Distributed Fault Diagnosis for Large-Scale Nonlinear Stochastic Systems. in Proceedings of the ASME Dynamics Systems and Control Conference, Colum- bus, Ohio, October 2015.

2

E. Noursadeghi and I.A. Raptis. Distributed Fault Detection of Nonlinear Large-Scale Dynamic Systems. in Proceedings of the ACM/IEEE Sixth International Conference on Cyber-Physical Sys- tems, pp. 51-59, Seattle, WA, April 2015.

E. Noursadeghi, P. Kamani, A. Afshar. Defect Root Cause Analysis in Production Lines based on Hierarchical Baysian Network. in Proceeding of the ASME Computer and Computational Intelligence Conference, Bangkok, Thailand, September 2011.

E. Noursadeghi, P. Kamani, A. Afshar. Defect Root Cause Analysis in Large Scale Manufacturing System. in Proceeding of the ASME Computer and Computational Intelligence Conference, Bangkok, Thailand, September 2011.

P. Kamani, E. Noursadeghi, A. Afshar, F. Towhidkhah. Car Body Paint Defect Detection and Classi cation. in Proceeding of the ASME Computer and Computational Intelligence Conference, Bangkok, Thailand, September 2011.

P. Kamani, E. Noursadeghi, A. Afshar, F. Towhidkhah. Automatic Paint Defect Detection and Classi cation of Car Body. in Proceeding of the 7th IEEE Iranian Conference on Machine Vision and Image Processing, Tehran, Iran, November 2011. Honors and

Awards

Awarded Travel Grant, ASME Dynamic Systems and Control Conference Oct. 2015,2016

Awarded Travel Grant, IEEE Cyber Security Conference Sept. 2016

Top student award, Isfahan University of Technology Sept. 2009

Undergraduate scholarship prize, Isfahan University of Technology Sept. 2005-Sept. 2009

Ranked 564 among more than 450,000 participants in Iran’s nationwide university entrance exam Sept. 2005

Top sudent city council award Aug. 2004

Professional

Activities

Reviewer of American Control Conference Sept. 2015

Member of institute of Electrical and Electronics Engineers (IEEE) Sept. 2013-Present

Member of American Society of Mechanical Engineers (ASME) Sept. 2013-Present

Member of Association for Women in Science (AWIS) Sept. 2014-Sept. 2015 3

Contact this candidate