⚡ Power Systems · 🤖 Machine Learning & Intelligence · 🕸 Networked Dynamical Systems
I am Nima Taghipour Bazargani, an electrical engineer and researcher with a focus on power systems, machine learning, and networked dynamical systems. My work is motivated by a simple question: how do large infrastructures behave when viewed as interacting physical, informational, and economic layers, and how can we design tools that make these systems more observable, understandable, and resilient in practice?
Professional Experience
Over the past years I have worked on real-time event identification, wide-area measurement analytics, forecasting frameworks for system strength and inertia, and large-scale electricity market studies. This path has taken me through academic research labs, industry roles, and international collaborations, and it has shaped a perspective that connects theoretical models with operational realities in power and energy systems.
Most recently, I worked as a Power System Engineer at Electric Power Group (EPG) in Pasadena, California. In this role, I developed a real-time inertia and system-strength forecasting framework that integrates synchrophasor data, physics-informed features, and continual-learning techniques. The work involved building a robust Python pipeline for streaming data ingestion, adaptive model updates, and multi-horizon forecasting, with results incorporated into operator-facing visualization tools that enhance situational awareness for utilities and reliability coordinators.
Before joining EPG, I served as a Network Model Analyst at CWP Energy, collaborating remotely with teams in Montreal. I conducted large-scale security-constrained unit commitment and security-constrained economic dispatch simulations across major U.S. energy markets—including MISO, SPP, PJM, and ERCOT. I developed and maintained a Python- and Google Cloud–based simulation framework for long-term scenario evaluation, congestion analysis, and strategy assessment, strengthening my understanding of how network topology and constraints shape pricing, risk, and market behavior.
As a Data Scientist Intern at Powerline (gigElev), I contributed to the development and refinement of software tools for predicting electricity-grid and e-mobility features. My work focused on Python-based integrations with third-party APIs, supporting automated tariff and rate calculations, and contributing to data preprocessing, feature analysis, and machine-learning components within Powerline’s product ecosystem.
I also contributed to collaborative R&D through the U.S.–Israel Energy Center (BIRD Foundation), where I helped develop an open-source event-identification package for synchrophasor data. My work included feature-extraction modules, data-generation workflows, and semi-supervised learning components for disturbance classification.
Earlier in my career, I worked at TU Darmstadt (MAKI CRC 1053) in Germany on virtual network embedding for cyber-physical infrastructures, developing stochastic mixed-integer optimization models focused on resilience and efficient resource allocation.
Academic Journey
I completed my Ph.D. in Electrical Engineering at Arizona State University (ASU), specializing in the real-time identification of power system events using high-dimensional spatio-temporal PMU data, under the guidance of Professor Oliver Kosut and Professor Lalitha Sankar.
During my doctoral studies, I contributed to several large research efforts, including NSF- and PSERC-funded projects on high-dimensional data science, information-theoretic monitoring, and event identification, as well as the U.S.–Israel Energy Center program on ML-based frameworks for PMU event identification.
Before ASU, I earned my M.Sc. in Power Systems from K.N. Toosi University of Technology, graduating second in my program. My research focused on microgrid resiliency and optimal sizing of battery storage systems. I completed my B.Sc. in Electrical Engineering at Shahid Beheshti University, where I studied transmission and distribution systems along with transformer testing infrastructure.
Research and Technical Interests
My research interests are centered on networked dynamical systems in power and energy contexts. I am particularly interested in:
- how graph structure and network topology influence system dynamics and risk;
- how physical, cyber, and market layers interact in real operating environments;
- how measurements and models can be combined into robust monitoring and decision tools;
- how learning algorithms can adapt to evolving system conditions and rare events.
These themes have gradually evolved into a broader direction that I refer to as Graph Dynamical Flows, which aims to organize thinking about flows, constraints, and feedback across interconnected physical and informational networks.
Personal Interests
Beyond engineering, my interests center on how complex systems learn, coordinate, and fail. I engage with neuroscience, psychology, history, and philosophy, view poker as a compact setting for reasoning about uncertainty and strategy, follow football as a large-scale coordination problem, and enjoy hiking, camping, off-road travel, solving mechanical and three-dimensional puzzles, and writing.