Hybrid Policy-Based Reinforcement Learning of Adaptive Energy Management for the Energy Transmission-Constrained Island Group

Yang, Lingxiao; Li, Xiaofeng; Sun, Mengwei; Sun, Changyin

doi:10.1109/tii.2023.3241682

Cited by 68 publications

(29 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…An HPRL method that can deal with discrete-continuous mixing behavior is proposed to solve the energy management problem of island groups. For complex models with discrete-continuous mixed actions, it is better to avoid simplifying the model to obtain an optimization strategy (Yang et al, 2023). The methods for solving MG optimization problems are shown in Supplementary Table S1.…”

Section: Related Workmentioning

confidence: 99%

A bi-layer optimization method of the grid-connected microgrid based on the multi-strategy of the beluga whale algorithm

Zhong,

Sun,

2024

Front. Energy Res.

View full text Add to dashboard Cite

Endeavoring to enhance the penetration rate of renewable energy sources, concurrently ensuring economic and operational stability, this study proposes a novel bi-layer optimization method of the wind–solar-storage AC/DC microgrid (MG). First, by incorporating a superordinate electric and seasonal hydrogen hybrid energy storage system (E&SHESS), the topology structure of the microgrid is established. Subsequently, to rectify the intrinsic limitations of the conventional beluga whale optimization (BWO) algorithm, this paper proposes a multi-strategy hybrid improvement to BWO (MHIBWO). This innovative improvement integrates an MTent strategy, a step size adjustment mechanism, and a crisscross strategy. Then, constructing a bi-layer iterative model based on the topology, annual net income and grid-connected friendliness are introduced as optimization objectives for the outer and inner layers, respectively, utilizing MHIBWO and CPLEX for resolution. Through a nested iteration of the two layers, the model outputs the capacity scheme with the best performance of economy and stability. Finally, the simulation unequivocally demonstrated the superiority of MHIBWO and the model proposed. In addition, based on the real data of the Elia power station, the validity of the method in operation is tested using the fuzzy C-means algorithm (FCMA) to extract and aggregate typical days, thereby presenting a sophisticated solution for the field of microgrids optimization configuration.

show abstract

Section: Related Workmentioning

confidence: 99%

A bi-layer optimization method of the grid-connected microgrid based on the multi-strategy of the beluga whale algorithm

Zhong,

Sun,

2024

Front. Energy Res.

View full text Add to dashboard Cite

show abstract

“…In specific scenarios, such a system can play a crucial role in scheduling distributed resources. For instance, in isolated islands where the inverse distribution of resources and limited energy transmission are the factors, adaptive energy management can be accomplished through reinforcement learning based on hybrid strategies to govern the energy of island clusters (Yang et al, 2023). Therefore, with a panoramic perception of distributed resources, the regulation of renewable energy can be more precise and effective, facilitating the interconnection and interaction between sources, networks, loads, and storage.…”

Section: Distributed Smart Sensing Systemmentioning

confidence: 99%

Dynamic planning of edge sensing terminals in distribution network supporting distributed resources observable and controllable

Ji,

Liu,

Yan

et al. 2024

Front. Energy Res.

View full text Add to dashboard Cite

With the advancement of low-carbon distribution networks, the heightened stochasticity introduced by a multitude of renewable energy sources in the power grid has significantly augmented the regulatory challenges faced by the power grid. Dispatching distributed resources emerges as an effective solution to this issue. However, these resources often lack observability and controllability, hindering their participation in power regulation services. To establish a reliable interaction between distributed resources and power grids, the deployment of numerous edge sensing terminals becomes essential, albeit incurring high costs. In light of this, our paper proposes a dynamic network planning method for edge sensing terminals based on node differentiation and resource observability criteria, aiming to facilitate real-time and dependable observation of distributed resources. Initially, the node weight, a metric to gauge the disparity among nodes, is computed, considering communication quality deviation, resource development synergy, and the distribution of distributed resources. Subsequently, an optimal configuration method is introduced, accounting for the terminal’s reliability under faults. Lastly, a method for dynamic terminal networking planning is presented, gradually reducing the depth of unobservable resources. An enhanced genetic algorithm is employed to address this challenge. This method was validated using an IEEE 33 node system and a 91 node actual system, demonstrating significant effectiveness in reducing terminal configuration costs.

show abstract

“…Moreover, it is also used to avoid DC over voltages due to motors (PMSG) during line faults and slow-changing power surges in the system. The microgrid can switch to the islanded mode of operation if power flow fluctuation approaches its limits; therefore, an RES-based islanded microgrid is controlled through adaptive EMSs using the transmission line constraints as control parameters in [88]. However, this kind of system is particularly not implementable in real life because it needs multi-objective and multi-constraint EMSs.…”

Section: Energy Management Systemsmentioning

confidence: 99%

Enhancing Microgrid Stability and Energy Management: Techniques, Challenges, and Future Directions

Safder,

Sanjari,

Hamza

et al. 2023

Energies

View full text Add to dashboard Cite

Microgrid technology offers a new practical approach to harnessing the benefits of distributed energy resources in grid-connected and island environments. There are several significant advantages associated with this technology, including cost-effectiveness, reliability, safety, and improved energy efficiency. However, the adoption of renewable energy generation and electric vehicles in modern microgrids has led to issues related to stability, energy management, and protection. This paper aims to discuss and analyze the latest techniques developed to address these issues, with an emphasis on microgrid stability and energy management schemes based on both traditional and distinct approaches. A comprehensive analysis of various schemes, potential issues, and challenges is conducted, along with an identification of research gaps and suggestions for future microgrid development. This paper provides an overview of the current state of the field and proposes potential areas of future research.

show abstract

Hybrid Policy-Based Reinforcement Learning of Adaptive Energy Management for the Energy Transmission-Constrained Island Group

Cited by 68 publications

References 29 publications

A bi-layer optimization method of the grid-connected microgrid based on the multi-strategy of the beluga whale algorithm

A bi-layer optimization method of the grid-connected microgrid based on the multi-strategy of the beluga whale algorithm

Dynamic planning of edge sensing terminals in distribution network supporting distributed resources observable and controllable

Enhancing Microgrid Stability and Energy Management: Techniques, Challenges, and Future Directions

Contact Info

Product

Resources

About