Article
Engineering, Electrical & Electronic
Qiushi Cui, Yang Weng
Summary: This paper proposes an environment-adaptive protection scheme (E-APS) to solve the protection coordination issue using reinforcement learning for relay settings adjustment, proving its effectiveness. The scheme shows higher adaptability in different system operation scenarios compared to other optimization-based protection schemes, achieving high performance in protection coordination.
INTERNATIONAL JOURNAL OF ELECTRICAL POWER & ENERGY SYSTEMS
(2021)
Article
Green & Sustainable Science & Technology
Van-Hai Bui, Wencong Su
Summary: This study develops a dynamic network reconfiguration strategy to minimize operation cost and load shedding by using deep reinforcement learning for determining the optimal reconfiguration and set-points of distributed generators in real-time operation. The proposed method improves the stability and service reliability of distribution networks by quickly finding the optimal configuration and real-time set-points.
SUSTAINABLE ENERGY TECHNOLOGIES AND ASSESSMENTS
(2022)
Article
Green & Sustainable Science & Technology
A. T. D. Perera, Parameswaran Kamalaruban
Summary: Energy systems are transitioning to accommodate renewable energy technologies and face challenges in controlling energy flows. Research has shown potential for reinforcement learning in improving performance, but certain complexities remain unresolved, indicating room for further development.
RENEWABLE & SUSTAINABLE ENERGY REVIEWS
(2021)
Article
Energy & Fuels
M. Elgamal, Nikolay Korovkin, A. Abdel Menaem, Akram Elmitwally
Summary: This paper proposes a new operation management scheme (OMS) for complex power scheduling in a reconfigurable microgrid. The OMS aims to minimize the operation cost, unmet load, and curtailed renewable power, and has been evaluated using a case study system.
Article
Energy & Fuels
Rendong Shen, Shengyuan Zhong, Xin Wen, Qingsong An, Ruifan Zheng, Yang Li, Jun Zhao
Summary: Under the backdrop of high global building energy consumption, utilizing renewable energy to meet the increasing demand of building energy systems can promote clean energy transformation and carbon neutrality. However, the complexity of BES control is increased with the introduction of renewable energy, and addressing the mismatch between supply and demand sides is challenging. The proposed multi-agent deep reinforcement learning framework optimized energy management in buildings, improving device control efficiency and renewable energy utilization in BES.
Article
Computer Science, Artificial Intelligence
Ling-Ling Li, Jun-Lin Xiong, Ming-Lang Tseng, Zhou Yan, Ming K. Lim
Summary: This study establishes a dynamic reconfiguration integrated optimization model for active distribution network (ADN) and proposes a novel solving approach using multi-objective sparrow search algorithm. By considering distributed generation and time-varying load, the study aims to improve the power quality, economic benefits, and energy benefits of ADN. Experimental results show that the proposed method effectively reduces power loss and node voltage deviation.
EXPERT SYSTEMS WITH APPLICATIONS
(2022)
Article
Engineering, Electrical & Electronic
Xi Zhang, Weiqi Hua, Youbo Liu, Jiajun Duan, Zhiyuan Tang, Junyong Liu
Summary: This paper proposes a performance-oriented method for active distribution network planning, which improves the performance of the planning through dynamically updating the solution space using deep neural networks.
INTERNATIONAL JOURNAL OF ELECTRICAL POWER & ENERGY SYSTEMS
(2022)
Article
Energy & Fuels
Zonggen Yi, Yusheng Luo, Tyler Westover, Sravya Katikaneni, Binaka Ponkiya, Suba Sah, Sadab Mahmud, David Raker, Ahmad Javaid, Michael J. Heben, Raghav Khanna
Summary: This paper introduces a deep reinforcement learning (DRL)-based framework to address the complex decision-making tasks for nuclear-renewable integrated energy systems (NR-IES). Comparisons with a conventional control approach demonstrate the superiority of DRL in controlling NR-IES.
Article
Automation & Control Systems
Zhigang Ye, Chen Chen, Bo Chen, Kai Wu
Summary: This article presents an integrated optimization model for unbalanced distribution system restoration after extreme events, which coordinates control actions of various distributed energy resources and considers topology flexibility. Numerical results validate the effectiveness of the model and highlight the necessity of coordinating different flexible resources.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2021)
Article
Automation & Control Systems
Xuesong Wang, Tianyi Li, Yuhu Cheng
Summary: The novel Proximal Parameter Distribution Optimization (PPDO) algorithm enhances the exploration ability of reinforcement learning agents by transforming neural network parameters and setting two groups of parameters. By limiting the amplitude of two consecutive parameter updates, PPDO reduces the influence of bias and variance on the value function approximation, thus improving the stability of parameter distribution optimization.
IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS
(2021)
Article
Construction & Building Technology
Rendong Shen, Shengyuan Zhong, Ruifan Zheng, Dongfang Yang, Bingqing Xu, Yang Li, Jun Zhao
Summary: In order to reduce energy costs and carbon emissions, HVAC systems need to be based on renewable energy utilization to improve efficiency. This study proposes a multi-agent cooperative optimization control framework based on deep reinforcement learning, which achieves control of the regenerative electric heating system by optimizing the match between supply and demand. By introducing feasible action screening mechanism, prioritized experience replay mechanism, and regulation mechanism based on occupant behavior, the stability, efficiency, and flexibility of the optimization control framework are further improved. Simulation results show that compared to the baseline model, the multi-agent cooperative optimization framework reduces thermal discomfort duration by 84.86%, unconsumed renewable energy by 70.79%, and energy costs by 16.08%.
ENERGY AND BUILDINGS
(2023)
Article
Engineering, Chemical
Naiwei Tu, Zuhao Fan
Summary: A dynamic reconfiguration method based on the improved multi-objective dung beetle optimizer (IMODBO) is proposed to reduce the operating cost of the distribution network with distributed generation (DG) and ensure the quality of the power supply. It also minimizes the number of switch operations during dynamic reconfiguration. The method includes establishing a multi-objective model, applying the K-means++ clustering algorithm, and using the IMODBO algorithm for reconstruction. The results demonstrate the effectiveness of the proposed method in addressing the multi-objective distribution network dynamic reconfiguration problem.
Article
Engineering, Electrical & Electronic
Mostafa Ghasemi, Ahad Kazemi, Andrea Mazza, Ettore Bompard
Summary: This study introduces a novel three-stage stochastic planning model to maximize the resilience of distribution systems by deciding on line hardening and Distributed Generation (DG) placement. It then forms provisional microgrids (MG) based on line outage uncertainty and minimizes load shedding costs through demand-side management programs.
IET GENERATION TRANSMISSION & DISTRIBUTION
(2021)
Article
Engineering, Electrical & Electronic
Ziyang Yin, Shouxiang Wang, Qianyu Zhao
Summary: This paper proposes a real-time reconfiguration method for the integration of distributed generation and distribution network using deep reinforcement learning. By constructing a Markov decision process-based reconfiguration model and a soft open point optimization model, the decision-making can be achieved in milliseconds. The proposed method effectively reduces the operation cost and solves the overvoltage problem caused by high photovoltaic integration.
JOURNAL OF MODERN POWER SYSTEMS AND CLEAN ENERGY
(2023)
Article
Computer Science, Information Systems
Abhijeet Sahu, Venkatesh Venkatraman, Richard Macwan
Summary: Recently, there has been a lot of research on data-driven approaches using machine learning techniques to control an electric grid. Reinforcement learning (RL) provides a viable alternative to conventional solvers when there is uncertainty in the environment. Efficiently training an agent requires a lot of interaction with the environment. This paper focuses on developing and validating a mixed-domain RL environment and presents the results of co-simulation and training RL agents for a cyber-physical network reconfiguration and Volt-Var control problem in a power distribution feeder.
Article
Engineering, Multidisciplinary
Yuanzheng Li, Zhixian Ni, Tianyang Zhao, Minghui Yu, Yun Liu, Lei Wu, Yong Zhao
IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS
(2020)
Article
Engineering, Multidisciplinary
Yuanzheng Li, Zhixian Ni, Tianyang Zhao, Tianwei Zhong, Yun Liu, Lei Wu, Yong Zhao
IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS
(2020)
Article
Engineering, Multidisciplinary
Yuanzheng Li, Zhixian Ni, Yun Liu, Xiaomeng Ai, Qingyong Zhang, Jiawei Yang, Xi Li
Summary: This paper proposes a multi-network framework to consider the coupled flow constraints in HVSEF planning. By using a modified maximum covering location method and a multi-objective optimization model, the problem of HVSEF planning is successfully addressed.
IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS
(2022)
Proceedings Paper
Energy & Fuels
Yuanzheng Li, Zhixian Ni, Tianyang Zhao, Yun Liu, Lei Wu, Yong Zhao
Summary: This paper presents a coordinated operation method for electric vehicle charging stations and a distribution power network, considering integrated energy and reserve regulation. The proposed method includes energy-reserve integrated decision formulations, shared energy and reserve scheduling model based on coalition game, and economic benefit allocation study. Case study verifies the effectiveness of the method by comparing operation costs in coordinated and non-coordinated modes for EVCSs and DPN.
2021 IEEE/IAS 57TH INDUSTRIAL AND COMMERCIAL POWER SYSTEMS TECHNICAL CONFERENCE (I&CPS)
(2021)
Proceedings Paper
Engineering, Industrial
Yuanzheng Li, Zhixian Ni, Tianyang Zhao, Minghui Yu, Yun Liu, Lei Wu
2019 IEEE INDUSTRY APPLICATIONS SOCIETY ANNUAL MEETING
(2019)
