Zhisen Jiang

ETH Zürich Ph.D. Student, ETH Zürich (2025-Now)

I am a Ph.D. student at the Automatic Control Laboratory (IfA) at ETH Zürich, working with Prof. Florian Dörfler and Dr. Saverio Bolognani. I received my B.Eng. degree (with honors) in Electrical Engineering from North China Electric Power University and M.S. degree (with honors) in Electrical Engineering from Tsinghua University, China, in 2022 and 2025, respectively.

I am interested in applying game theory and control theory to power systems. In my previous work, I designed peer-to-peer energy trading mechanisms and developed allocation methods for Dynamic Operating Envelopes in distribution networks. Currently, I am working on co-designing incentives and automation strategies for transmission and distribution network coordination. In my free time, I enjoy traveling and watching basketball games.

📢 I am currently conducting theoretical research on zeroth-order methods for bilevel optimization, with a focus on nonsmooth analysis and feedback optimization. My research interests remain centered on power systems, particularly distribution network operation and energy markets. If you are interested in pursuing a thesis or semester project in these areas, feel free to get in touch.

Google Scholar LinkedIn
Education
  • ETH Zürich

    ETH Zürich

    Ph.D. in Electrical Engineering Sep. 2025 - Now

  • Tsinghua University

    Tsinghua University

    M.S. in Electrical Engineering (with honors) Sep. 2022 - Jul. 2025

    Rank 1/1215 during master's program, GPA: 4.0/4.0

  • North China Electric Power University

    North China Electric Power University

    B.Eng. in Electrical Engineering and its Automation (with honors) Sep. 2018 - Jul. 2022

    Rank 1/46 in undergraduate studies, GPA: 94.48/100

Honors & Awards
  • 🏆Outstanding Graduate of Beijing (Master) 2025
  • 🏆Tsinghua University Comprehensive Excellence Scholarship (First Class) 2024
  • 🎓NCCR Automation Fellowship 2024
  • 🏆Outstanding Graduate of Beijing (Bachelor) 2022
  • 🥇3x National Scholarship (2019, 2020, 2021) 2019-2021
Experience
Service
News
2026
🎤 Gave a talk at Prof. Le Xie's group at Harvard. I talked about TSO-DSO coordination in Switzerland and about our PowerUp work on incentive-based load curtailment with bilevel zeroth-order learning.
Sep 25
🏛️ Started visiting Prof. Gioele Zardini's lab at MIT. During the visit, I presented our recent results on Project MESTRO and our bilevel algorithm to Prof. Saurabh Amin's group and to the energy group in the MIT Climate & Sustainability Consortium.
Sep 15
🎤 Presented our full paper Incentive-Based Load Curtailment with Limited Information: A Bilevel Zeroth-Order Learning Approach (arXiv) and a poster of a master thesis that I supervised at PowerUp 2026 in Boulder, Colorado. A small but intensive conference that was very informative and useful. I loved it!
Sep 09
📄 Our paper Bi-ZOL: Bilevel Zeroth-Order Learning with Nonsmooth Responses (arXiv) is online. It provides a "smarter" zeroth-order estimate of the undefined hypergradient for general bilevel optimization with a constrained lower level. The method handles the information asymmetry in Stackelberg games (black-box followers) and the nondifferentiable lower-level solution map at the same time.
Sep 07
📄 Our paper on Incentive-Based Load Curtailment with Limited Information: A Bilevel Zeroth-Order Learning Approach (arXiv) has been accepted to the 2026 PowerUp Conference in Boulder, CO. We propose a partial smoothing method to deal with the nonsmoothness in bilevel problems. You could check the method and convergence proof first! A more formal, general, and mathematically rigorous version is on the way!
Apr 10
📋 Participated in Champéry 2026, where I presented our joint work with Swissgrid titled 'OPFV 2.0: Re-Engineering Voltage Procurement via Online Bilevel Games' during the poster session. You can find the presentation below.
Feb 10
2025
📊 Will give a presentation about my CDC paper 'Voltage Support Procurement in Transmission Grids: Incentive Design via Online Bilevel Games' at Session GameTheory VFrB13, Room Oceania IX, Rio
Dec 12
🏢 Visited Swissgrid and gave a presentation 'OPF-V 2.0: Automatic incentive for voltage support procurement'
Nov 26
🎓 Started my Ph.D. studies at ETH Zürich in the Automatic Control Laboratory (IfA) Lab Website
Sep 15
Selected Publications (view all )
Bi-ZOL: Bilevel Zeroth-Order Learning with Nonsmooth Responses
Bi-ZOL: Bilevel Zeroth-Order Learning with Nonsmooth Responses

Zhisen Jiang, Saverio Bolognani

arXiv:2609.08021 2026 Preprint

This paper studies bilevel optimization with a constrained lower level, where the lower-level model is unavailable and the response mapping is locally Lipschitz but possibly nonsmooth, as in Stackelberg games with black-box followers. Bi-ZOL (Bilevel Zeroth-Order Learning) keeps the exact upper-level partial gradients and uses zeroth-order sampling only to estimate the response Jacobian. It converges to a stationary point with a quantified bias, and experiments on tracking problems show improved performance over standard zeroth-order smoothing.

[Paper]

Bi-ZOL: Bilevel Zeroth-Order Learning with Nonsmooth Responses
Bi-ZOL: Bilevel Zeroth-Order Learning with Nonsmooth Responses

Zhisen Jiang, Saverio Bolognani

arXiv:2609.08021 2026 Preprint

This paper studies bilevel optimization with a constrained lower level, where the lower-level model is unavailable and the response mapping is locally Lipschitz but possibly nonsmooth, as in Stackelberg games with black-box followers. Bi-ZOL (Bilevel Zeroth-Order Learning) keeps the exact upper-level partial gradients and uses zeroth-order sampling only to estimate the response Jacobian. It converges to a stationary point with a quantified bias, and experiments on tracking problems show improved performance over standard zeroth-order smoothing.

[Paper]

Incentive-Based Load Curtailment with Limited Information: A Bilevel Zeroth-Order Learning Approach
Incentive-Based Load Curtailment with Limited Information: A Bilevel Zeroth-Order Learning Approach

Zhisen Jiang, Florian Dörfler, Saverio Bolognani

2026 PowerUp Conference, Boulder, CO 2026 Conference

This work formulates a lower-level constrained bilevel optimization to solve load curtailment problem and propose the Bi-ZOL (Bilevel Zeroth-Order Learning) algorithm. Unlike conventional black-box methods, Bi-ZOL exploits the bilevel structure to decompose the hypergradient, integrating the exact analytical information of the leader's objective with a zeroth-order estimate of the unknown response sensitivity.

[Paper]

Incentive-Based Load Curtailment with Limited Information: A Bilevel Zeroth-Order Learning Approach
Incentive-Based Load Curtailment with Limited Information: A Bilevel Zeroth-Order Learning Approach

Zhisen Jiang, Florian Dörfler, Saverio Bolognani

2026 PowerUp Conference, Boulder, CO 2026 Conference

This work formulates a lower-level constrained bilevel optimization to solve load curtailment problem and propose the Bi-ZOL (Bilevel Zeroth-Order Learning) algorithm. Unlike conventional black-box methods, Bi-ZOL exploits the bilevel structure to decompose the hypergradient, integrating the exact analytical information of the leader's objective with a zeroth-order estimate of the unknown response sensitivity.

[Paper]

Voltage Support Procurement in Transmission Grids: Incentive Design via Online Bilevel Games
Voltage Support Procurement in Transmission Grids: Incentive Design via Online Bilevel Games

Zhisen Jiang, Saverio Bolognani, Giuseppe Belgioioso

64th IEEE Conference on Decision and Control (CDC) 2025 Conference

This paper presents a novel approach for voltage support procurement in transmission grids using online bilevel games. The proposed incentive design mechanism addresses the voltage procurement challenges between transmission and distribution networks of Swissgrid.

[Paper]

Voltage Support Procurement in Transmission Grids: Incentive Design via Online Bilevel Games
Voltage Support Procurement in Transmission Grids: Incentive Design via Online Bilevel Games

Zhisen Jiang, Saverio Bolognani, Giuseppe Belgioioso

64th IEEE Conference on Decision and Control (CDC) 2025 Conference

This paper presents a novel approach for voltage support procurement in transmission grids using online bilevel games. The proposed incentive design mechanism addresses the voltage procurement challenges between transmission and distribution networks of Swissgrid.

[Paper]

Dynamic Operating Envelopes Embedded Peer-to-Peer-to-Grid Energy Trading
Dynamic Operating Envelopes Embedded Peer-to-Peer-to-Grid Energy Trading

Zhisen Jiang, Ye Guo, Jianxiao Wang

Applied Energy 2025 Journal

This paper introduces a comprehensive framework for peer-to-peer-to-grid energy trading embedded with Dynamic Operating Envelopes (DOEs). The proposed approach utilizes DOEs to facilitate the grid-friendly P2P trading of prosumers. Moreover, the so-called DOE price is proposed to allow prosumers purchase more network injection limits.

[Paper]

Dynamic Operating Envelopes Embedded Peer-to-Peer-to-Grid Energy Trading
Dynamic Operating Envelopes Embedded Peer-to-Peer-to-Grid Energy Trading

Zhisen Jiang, Ye Guo, Jianxiao Wang

Applied Energy 2025 Journal

This paper introduces a comprehensive framework for peer-to-peer-to-grid energy trading embedded with Dynamic Operating Envelopes (DOEs). The proposed approach utilizes DOEs to facilitate the grid-friendly P2P trading of prosumers. Moreover, the so-called DOE price is proposed to allow prosumers purchase more network injection limits.

[Paper]

Bargaining-based Approach for Dynamic Operating Envelope Allocation in Distribution Networks
Bargaining-based Approach for Dynamic Operating Envelope Allocation in Distribution Networks

Zhisen Jiang, Ye Guo

IEEE Transactions on Smart Grid 2025 Journal

This paper presents a novel bargaining-based approach for Dynamic Operating Envelope (DOE) allocation in distribution networks. The proposed method addresses the challenges of fair and efficient DOE allocation in modern power distribution systems.

[Paper]

Bargaining-based Approach for Dynamic Operating Envelope Allocation in Distribution Networks
Bargaining-based Approach for Dynamic Operating Envelope Allocation in Distribution Networks

Zhisen Jiang, Ye Guo

IEEE Transactions on Smart Grid 2025 Journal

This paper presents a novel bargaining-based approach for Dynamic Operating Envelope (DOE) allocation in distribution networks. The proposed method addresses the challenges of fair and efficient DOE allocation in modern power distribution systems.

[Paper]

All publications