Author name cluster

Minghui Yan

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

2 papers

1 author row

EAAI Journal 2026 Journal Article

A look-ahead dispatch method via evolution strategies embedded with domain knowledge

Yuchen Dai
Weiran Jiao
Yi Tang
Minghui Yan
Feng Xue
Jianfeng Zhao

Modern power systems require proactive look-ahead dispatch strategies to address various uncertainties. Traditional decision-making methods based on physical models often suffer from slow processing speeds and struggle to handle multiple uncertain scenarios. Meanwhile, reinforcement learning methods face challenges such as hyperparameter sensitivity and a tendency to converge to local optima. To overcome these limitations, a look-ahead dispatch method via evolution strategies embedded with domain knowledge is proposed. First, a knowledge-embedded Markov decision process model of look-ahead dispatch is developed. This model encodes critical physical knowledge into the action space without computational burden. Second, a decision-making approach based on evolution strategies and physical models is introduced. This method enhances parallel exploration efficiency and reduce communication burden by leveraging synchronous random seeds and mirror perturbation techniques. Then, physical models are used to fine-tune agents in new scenarios with limited data. Finally, case studies based on the IEEE 118 system show that the proposed method significantly improves decision-making efficiency without sacrificing accuracy. Compared to deep reinforcement learning, the evolution strategies algorithm offers superior training efficiency and performance, effectively addressing the high-dimensional uncertainties and complexities of modern power systems. This establishes the proposed method as an effective solution for complex decision-making tasks in power system operations.

Details DOI

EAAI Journal 2025 Journal Article

Deep reinforcement learning explanation-assisted integer variable reduction method for security-constrained unit commitment

Yuchen Dai
Wei Xu
Minghui Yan
Feng Xue
Jianfeng Zhao

The large-scale security-constrained unit commitment (SCUC) is pivotal for ensuring the secure and economical operation of modern power systems. Formulated as a mixed-integer nonlinear programming problem, mathematical model-based methods struggle to balance computation efficiency and solution accuracy. While artificial intelligence methods offer promising potential, they face several obstacles, including limited interpretability and generalizability constraints. In light of these challenges, this paper proposes an interpretation method for deep reinforcement learning models that is used to reduce integer variables for large-scale SCUC problem. This method employs a Gaussian Mixture Model to cluster the decision outcomes of the agents and utilizes an improved decision tree to interpret the clustering results. We analyze the physical implications behind the phenomenon of unit output distributions exhibiting multiple independent Gaussian distributions. Then, these interpretations are applied to identify active integer variables, thereby simplifying the complexity of the SCUC problem and enhancing solution efficiency. Furthermore, an improved Markov decision process model with domain knowledge pertinent of power systems is constructed to enhance the interpretability and reliability of the agents. A distinctive feature of this model is the incorporation of a bidirectional mapping of unsafe and safe actions. The case studies on the SG-126 system demonstrate that the proposed method achieves a significant increase in solution speed without loss of accuracy. The identified active integer variables are proven to be accurate and effective, contributing to improve computation efficiency of unit commitment. The proposed method also provides a novel explainable artificial intelligence-assisted method for complex decision-making problems in other fields.

Details DOI