Research

Multi-agent systems have attracted increasing attention due to the rapid development of Internet of Things (IoT) technologies. An important example is the modern power system with numerous distributed energy resources (DERs). In a multi-agent system, each agent seeks to minimize its own costs, with the outcome dependent not only on individual actions but also on the collective decisions of all agents. These interactions necessitate a game-theoretic framework to model agent behaviors and to strike a good balance between individual agent goals and the overall system performance. Such game-theoretic analysis must address two critical aspects: the steady state (e.g., the existence, uniqueness, and optimality of an equilibrium) and the dynamics (e.g., the algorithms to compute an equilibrium).

I have been working on coordination mechanism design for power networks. Peer-to-peer energy trading mechanisms for networked prosumers have been developed, under which all agents play a (generalized) Nash game. Through rigorous mathematical analyses, we have proven that the proposed mechanisms possess desired properties such as the existence of a unique, (nearly) socially optimal equilibrium, and the attainment of a Pareto improvement. Effective bidding algorithms have been developed to compute the energy trading equilibrium. Real-world constraints, such as communication delays and measurement errors, are also taken into account.

Featured publications:

• M. Yang, R. Xie, Y. Zhang, Y. Chen*. Robust microgrid dispatch with real-time energy sharing and endogenous uncertainty, IEEE Transactions on Smart Grid, 2025, 16(4): 3085 - 3098.

• D. Yan, M.-Y. Chow, Y. Chen*. Low-carbon operation of data centers with joint workload sharing and carbon allowance trading, IEEE Transactions on Cloud Computing, 2024, 12(2): 750 - 761.

• Y. Chen*, C. Zhao, S. H. Low, Adam Wierman. An energy sharing mechanism considering network constraints and market power limitation. IEEE Transactions on Smart Grid, 2023, 14(2): 1027-1041.

• D. Yan, Y. Chen*. Distributed coordination of charging stations with shared energy storage in a distribution network, IEEE Transactions on Smart Grid, 2023, 14(6): 4666 - 4682.

• Y. Chen*, C. Zhao, S. H. Low, et al. Approaching prosumer social optimum via energy sharing with proof of convergence. IEEE Transactions on Smart Grid, 2020, 12(3): 2484-2495.

Robust optimization (RO) is a powerful tool for addressing uncertainties and has wide applications in smart grids, supply chain management, and portfolio construction. Traditional research often assumes that uncertain factors vary within a predetermined uncertainty set, overlooking the influence of decisions on this uncertainty set. However, in practice, decisions may directly affect the size or dimension of the uncertainty set and may indirectly affect the uncertainty set by changing the probability distribution underlying the historical data. Ignoring such decision-dependent features of uncertainties could lead to inaccurate and unreliable solutions.

To address this challenge, I work on end-to-end robust optimization with decision-dependent uncertainty. The direct influence of decisions will be captured by using a decision-dependent uncertainty (DDU) set rather than a decision-independent uncertainty (DIU) set; the indirect influence will be addressed using an “end-to-end” (or “prediction-and-optimize”) framework integrating prediction and optimization processes, rather than the traditional “predict-then-optimize” framework. Effective solution algorithms have been developed to address the potential failure of traditional algorithms to guarantee optimality and convergence when dealing with RO with DDU.

Featured publications:

• R. Xie, Y. Chen*. Multistage robust optimization for time-decoupled power flexibility aggregation with energy storage. arXiv preprint arXiv:2508.14477, 2025. (submitted to European Journal of Operational Research)

• R. Xie, Y. Chen*, P. Pinson. Predict-and-optimize robust unit commitment with statistical guarantees via weight combination. IEEE Transactions on Power Systems, 2025, early access.

• K. Qu, Y. Chen*, C. Zhao. Distributionally robust energy and reserve dispatch with distributed predictions of renewable energy, IEEE Transactions on Power Systems, 2025, early access. 

• R. Xie, P. Pinson, Y. Xu, Y. Chen*. Robust generation dispatch with purchase of renewable power and load predictions. IEEE Transactions on Sustainable Energy, 2024, 15(3): 1486 - 1501.

• T. Tan, R. Xie, X. Xu, Y. Chen*. A robust optimization method for power systems with decision‐dependent uncertainty. Energy Conversion and Economics, 2024, 5(3): 133-145. (Best Paper Award)

• Y. Chen*, W. Wei. Robust generation dispatch with strategic renewable power curtailment and decision-dependent uncertainty. IEEE Transactions on Power Systems, 2023, 38(5): 4640 - 4654. 

Online decision-making is critical for adapting to dynamic environments with uncertain future states. A primary challenge in this field is ensuring both feasibility and optimality under both time-average and worst-case scenarios. Two prevalent methodologies, model predictive control (MPC) and reinforcement learning (RL), present distinct advantages and limitations. MPC, being data-independent, often yields overly conservative solutions, while RL's full reliance on historical data may compromise its reliability in unprecedented situations. I work on developing trustworthy online algorithms that consistently achieve the best of both worlds, balancing the robustness of MPC with the adaptability of safe RL. These algorithms aim to ensure performance guarantees across various scenarios.

Featured publications:

• S. Huang, D. Han, J. Z. F. Pang, Y. Chen*. Optimal real-time bidding strategy for EV aggregators in wholesale electricity markets, IEEE Transactions on Intelligent Transportation Systems, 2025, 26(4): 5538 - 5551.

• D. Yan, S. Huang, Y. Chen*. Real-time feedback based online aggregate EV power flexibility characterization. IEEE Transactions on Sustainable Energy, 2024, 15(1): 658 - 673.

• R. Xie, Y. Chen*. Real-time bidding strategy of energy storage in an energy market with carbon emission allocation based on Aumann-Shapley prices, IEEE Transactions on Energy Markets, Policy, and Regulation, 2024, 2(3): 350-367.

• D. Yan, Y. Chen*. A distributed online algorithm for promoting energy sharing between EV charging stations. IEEE Transactions on Smart Grid, 2023, 14(2): 1158-1172.

• T. Li, Y. Chen, B. Sun, et al. Information aggregation for constrained online control, ACM SIGMETRICS 2021, 5(2): 1-35.