Day-Ahead Economic Dispatch Strategy for Distribution Networks with Multi-Class Distributed Resources Based on Improved MAPPO Algorithm DOI Creative Commons

Juan Zuo,

Qian Ai, Wenbo Wang

и другие.

Mathematics, Год журнала: 2024, Номер 12(24), С. 3993 - 3993

Опубликована: Дек. 19, 2024

In the context of global response to climate change and promotion an energy transition, Internet Things (IoT), sensor technologies, big data analytics have been increasingly used in power systems, contributing rapid development distributed resources. The integration a large number resources has led issues, such as increased volatility uncertainty distribution networks, large-scale data, complexity challenges optimizing security economic dispatch strategies. To address these problems, this paper proposes day-ahead scheduling method for networks based on improved multi-agent proximal policy optimization (MAPPO) reinforcement learning algorithm. This achieves coordinated multiple types within network environment, promoting effective interactions between grid coordination among Firstly, operational framework principles proposed algorithm are described. avoid blind trial-and-error instability convergence process during learning, generalized advantage estimation (GAE) function is introduced improve algorithm, enhancing stability updates speed training. Secondly, model containing constructed, model, actions, states, reward designed. Finally, effectiveness solving problem grids verified using IEEE 30-bus system example.

Язык: Английский

Advancements in data-driven voltage control in active distribution networks: A Comprehensive review DOI Creative Commons
Sobhy M. Abdelkader, Sammy Kinga, Emmanuel Ebinyu

и другие.

Results in Engineering, Год журнала: 2024, Номер 23, С. 102741 - 102741

Опубликована: Авг. 19, 2024

Distribution systems are integrating a growing number of distributed energy resources and converter-interfaced generators to form active distribution networks (ADNs). Numerous studies have been undertaken mitigate various challenges in ADNs. However, voltage deviation reactive power control still requires more attention from researchers system engineers. The Volt/VAr (VVC) concept has developed improve the quality, minimize losses, maintain profile deployed utility-owned legacy mechanisms such as on-load tap changers, capacitor banks, automatic regulators operate discrete, slow timescales unidirectionally, rendering them insufficient for optimal regulation Owing increasing use smart meters, inverters (SIs), sensors, data analytics tools, improved communication networks, become an important resource. Data-driven approaches, particularly reinforcement learning (RL)-based, therefore gained recent years effectively solving VVC decision-making problem. This comprehensive review presents detailed analysis advanced approaches used address It includes general overview problem formulation, frameworks, basic notations, well comparisons existing recently proposed methods. study focuses on data-driven especially RL-based algorithms. Some open research experienced application these algorithms safety, data, scalability, problems, interpretability cybersecurity threats presented alongside future perspectives Internet Things (IoT), Transfer Learning (TL), hybrid human-in-the-loop AI approaches.

Язык: Английский

Процитировано

10

A Hierarchical Voltage Control Strategy for Distribution Networks Using Distributed Energy Storage DOI Open Access
Chao Ma, Wenjie Xiong, Zhiyuan Tang

и другие.

Electronics, Год журнала: 2025, Номер 14(9), С. 1888 - 1888

Опубликована: Май 6, 2025

This paper presents a novel hierarchical voltage control framework for distribution networks to mitigate violations by coordinating distributed energy storage systems (DESSs). The establishes two-layer architecture that integrates centralized optimization with execution. In the upper layer, model predictive (MPC)-based controller computes optimal power dispatch trajectories critical buses, effectively decoupling slow-timescale from real-time adjustments. lower broadcast-based dispatches parameterized regulation signals, enabling autonomous active tracking DESS units. design explicitly addresses scalability limitations of conventional and cyber vulnerabilities peer-to-peer strategies. effectiveness proposed is verified on modified IEEE 34-bus 123-bus test feeder. results show method can average violation 93.7% robustness even under 60% communication loss condition.

Язык: Английский

Процитировано

0

Intelligent Control Strategy for Coal to Ethylene Glycol Wastewater Emission Reduction based on Dynamic Simulation and Reinforcement Learning DOI
Yang Sun, Zijian Liu, Zhe Li

и другие.

Process Safety and Environmental Protection, Год журнала: 2024, Номер unknown

Опубликована: Дек. 1, 2024

Язык: Английский

Процитировано

1

Distribution Network Anomaly Detection Based on Graph Contrastive Learning DOI

Mingjun Feng,

Caiyun Liu, Yan Sun

и другие.

Journal of Signal Processing Systems, Год журнала: 2024, Номер unknown

Опубликована: Дек. 11, 2024

Язык: Английский

Процитировано

0

Day-Ahead Economic Dispatch Strategy for Distribution Networks with Multi-Class Distributed Resources Based on Improved MAPPO Algorithm DOI Creative Commons

Juan Zuo,

Qian Ai, Wenbo Wang

и другие.

Mathematics, Год журнала: 2024, Номер 12(24), С. 3993 - 3993

Опубликована: Дек. 19, 2024

In the context of global response to climate change and promotion an energy transition, Internet Things (IoT), sensor technologies, big data analytics have been increasingly used in power systems, contributing rapid development distributed resources. The integration a large number resources has led issues, such as increased volatility uncertainty distribution networks, large-scale data, complexity challenges optimizing security economic dispatch strategies. To address these problems, this paper proposes day-ahead scheduling method for networks based on improved multi-agent proximal policy optimization (MAPPO) reinforcement learning algorithm. This achieves coordinated multiple types within network environment, promoting effective interactions between grid coordination among Firstly, operational framework principles proposed algorithm are described. avoid blind trial-and-error instability convergence process during learning, generalized advantage estimation (GAE) function is introduced improve algorithm, enhancing stability updates speed training. Secondly, model containing constructed, model, actions, states, reward designed. Finally, effectiveness solving problem grids verified using IEEE 30-bus system example.

Язык: Английский

Процитировано

0