Wenqi Cui

Assistant Professor

New York University

I am currently a tenure-track assistant professor in the Department of Electrical and Computer Engineering at New York Univerisity. My research interests lie broadly in the intersection of control, machine/reinforcement learning, and optimization for energy systems, with applications extending to general cyber-physical systems. My research group aims to develop fundamental theories and algorithms to tackle challenges in power and energy systems, and provide safe and robust AI-based solutions for real-world applications.

Before joining NYU, I was a postdoctoral fellow in the Department of Computing + Mathematical Sciences at California Institute of Technology (Caltech), hosted by Prof. Steven Low and Prof. Adam Wierman. I am honored to be a recipient of the Pioneer Postdoctoral Fellowship and PIMCO Postdoctoral Fellowship at Caltech. I obtained my Ph.D. in Electrical and Computer Engineering from the University of Washington, advised by Prof. Baosen Zhang.

I am looking for self-motivated PhD students starting in Fall 2027. If you are interested, please apply to the PhD program in the Department of Electrical and Computer Engineering via NYU’s official application system https://engineering.nyu.edu/admissions/graduate before December 1st. To ensure that I notice your application, please also email me at wenqicui@nyu.edu with your CV and academic transcripts.

Download my CV.

Interests

Control and Optimization
Machine Learning
Cyber-Physical Energy Systems

Education

PhD in Electrical and Computer Engineering, 2019-2024

University of Washington
MS in Electrical Engineering, 2016-2019

Zhejiang University
BEng in Electrical Engineering and Automation, 2012-2016

Southeast University

Recent News

All news»

[2026/09] New paper on Decentralized Gain Learning for Voltage Control via Local Trajectory Convolution.

[2026/09] New paper on Analytical Power-Aware Provisioning for Prefill-Decode Disaggregated AI Inference.

[2026/09] New paper on Measured Joules, Learned Routes: Learning to Route for Energy-Efficient LLM Serving.

[2026/03] We are organizing a workshop on AI Foundations for Power Grids at NeurIPS 2026. Welcome to submit your contributions and join us in advancing AI methodologies for reliable, intelligent, and resilient power grids!

[2026/08] New paper on Synthesizing Voltage Ride-Through Controllers for Data Centers.

Recent Publications

Quickly discover relevant content by filtering publications.

Wenqi Cui, Yan Jiang, Baosen Zhang, Yuanyuan Shi (2023). Structured Neural-PI Control with End-to-End Stability and Output Tracking Guarantees. NeurIPS 2023.

PDF Code Project

Wenqi Cui, Yan Jiang, Baosen Zhang (2023). Reinforcement Learning for Optimal Frequency Control: A Lyapunov Approach. IEEE Transactions on Power Systems.

PDF Cite Code Project

Wenqi Cui, Linbin Huang, Weiwei Yang, Baosen Zhang (2023). Efficient Reinforcement Learning Through Trajectory Generation. Learning for Dynamics and Control Conference.

PDF Code Project

Wenqi Cui, Baosen Zhang (2022). Equilibrium-Independent Stability Analysis for Distribution Systems with Lossy Transmission Lines. IEEE Control Systems Letters (L-CSS).

PDF Project

Yan Jiang, Wenqi Cui, Baosen Zhang, Jorge Cortés (2022). Stable Reinforcement Learning for Optimal Frequency Control: A Distributed Averaging-Based Integral Approach. IEEE Open Journal of Control Systems.

PDF Project

See all publications

Projects

Structured Neural-PI Control with End-to-End Stability and Output Tracking Guarantees

We study the optimal control of multiple-input and multiple-output dynamical systems via the design of neural network-based controllers with stability and output tracking guarantees. While neural network-based nonlinear controllers have shown superior performance in various applications, their lack of provable guarantees has restricted their adoption in high-stake real-world applications.

Efficient Reinforcement Learning Through Trajectory Generation

A key barrier to using reinforcement learning (RL) in many real-world applications is the requirement of a large number of system interactions to learn a good control policy. Off-policy and Offline RL methods have been proposed to reduce the number of interactions with the physical environment by learning control policies from historical data.

Reinforcement Learning for Optimal Frequency Control - A Lyapunov Approach

The increase in penetration of inverter-based resources provides us with more flexibility in frequency regulation compared to conventional linear droop controllers. Using their power electronic interfaces, inverter-based resources can realize almost arbitrary responses to frequency changes.

Equilibrium-Independent Stability Analysis for Distribution Systems with Lossy Transmission Lines

Power distribution systems are becoming much more active with increased penetration of distributed energy resources. Because of the intermittent nature of these resources, the stability of distribution systems under large disturbances and time-varying conditions is becoming a key issue in practical operations.

Structured Neural-PI Control for Networked Systems: Stability and Steady-State Optimality Guarantees

We study the control of networked systems with the goal of optimizing both transient and steady-state performances while providing stability guarantees. Linear Proportional-Integral (PI) controllers are almost always used in practice, but the linear parameterization of the controller fundamentally limits its performance.

Decentralized Safe Reinforcement Learning for Voltage Control

Inverter-based distributed energy resources provide the possibility for fast time-scale voltage control by quickly adjusting their reactive power. The power-electronic interfaces allow these resources to realize almost arbitrary control law, but designing these decentralized controllers is nontrivial.

Few-Shot Machine Learning Classification of Electron Microscopy Data

This is a UW-DIRECT capstone project cooperated with Pacific Northwest National Laboratory (PNNL) and mentored by Marjolein Oostrom, Sarah Akers and Dr. Steven R. Spurgeon. We proposed a few-Shot machine learning algorithum for classification of electron microscopy images and implemented a graphical-user interface for it.

Lyapunov-Regularized Reinforcement Learning for Power System Transient Stability

Transient stability of power systems is becoming increasingly important because of the growing integration of renewable resources. These resources lead to a reduction in mechanical inertia but also provide increased flexibility in frequency responses.

Green Seattle - Prediction and Visualization

Green Seattle is an inter-disciplinary project in Clearn Energy Insitute (CEI) collaborated by seven students majoring in Chemical Sciences, Political Science and Electrical Engineering. The first part of Green Seattle is focused on predicting Seattle Traffic Trends (codes are publicly available here), while the second is focused on Visualizing Seattle Traffic Trends, as well as outputs of a regression model seeking to predict future traffic patterns based on user inputs (codes are publicly available here).

Predicting Power System Dynamics and Transients: A Frequency Domain Approach

The dynamics of power grids are governed by a large number of nonlinear ordinary differential equations (ODEs). To safely operate the system, operators need to check that the states described by this set of ODEs stay within prescribed limits after various faults.

Biography

University of Washington

Ph.D. Student in Electrical and Computer Engineering

Sep 2019 – Present Seattle, WA

Research experience include:

Safe reinforcement learning with Lyapunov staility
Optimal frequency control and voltage control in power systems
Power system dynamic prediction under large disturbances
Inter-disciplinary research in Clearn Energy Institute, including traffic prediction and few-shot learning for Electron Microscopy Data

Microsoft Research

Research Intern

Jun 2022 – Sep 2022 Redmond, WA

I interned at Microsoft Research Special Projects at Remond Lab, mentored by Weiwei Yang. We proposed sample-efficient reinforcement learning algorithms for the control of largescale physical systems, including power systems and traffic networks. The proposed methods overcome the challenges of partial observability, sample complexity and the lack of real-time communication capability in real-world applications.

Microsoft Research

Research Intern

Jun 2021 – Sep 2021 Redmond, WA

I interned at Microsoft Research Special Projects at Remond Lab, mentored by Weiwei Yang. We worked on AI for sustainable energy systems. We proposed a novel framework for power system dynamic predictions by learning in the frequency domain. Compared with state-of-the art AI methods including Physics-Informed neural networks and generic deep neural network, the proposed method reduce MSE prediction error by more than 50% and vastly improves the detection of unstable dynamics. It also provides a computation speed up of more than 400 times compared to existing power system tools.

Zhejiang University

M.S. in Electrical Engineering

Sep 2016 – Jun 2019 Hangzhou, P.R. China

Research experience include:

Optimal dispatch of demand response resources in power systems
Reliability evaluation of power systems with distributed energy resources
Participate in projects to implement demand response programs in Jiangsu and Zhejiang Province