Smart generation control based on multi-agent reinforcement learning with the idea of the time tunnel

Xi, Lei; Chen, Jianfeng; Huang, Yuehua; Xu, Yanchun; Liü, Lang; Zhou, Yimin; Li, Yudan

doi:10.1016/j.energy.2018.04.042

Cited by 92 publications

(41 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In recent years, scholars have attempted to combine TL with RL to form TRL; to combine TL with heuristic SI algorithms such as ABC to form TBO; to combine TL with DRL to form a DTRL algorithm; to combine DL with typical RL algorithms such as Q‐learning to form a new DQL algorithm; to combine DL with another typical RL algorithm, the ADP, to form a new D‐ADP algorithm; and to combine TL with extreme learning to form a new extreme TL algorithm. Among these, the theoretical framework for a new D‐ADP algorithm proposed in Yin et al contains three networks, prediction network, evaluation network, and execution network, which are improved by adopting the DQNs.…”

Section: Hybrid Learningmentioning

confidence: 99%

“…Among these, the theoretical framework for a new D‐ADP algorithm proposed in Yin et al contains three networks, prediction network, evaluation network, and execution network, which are improved by adopting the DQNs. These above‐mentioned HL algorithms have been preliminarily used to good effect in the SG and EI fields …”

Section: Hybrid Learningmentioning

confidence: 99%

“…Recently, researchers have cascaded DL, RL, TL, heuristic SI algorithms, strategic‐search intelligent algorithms, and other methods to construct a variety of advanced HL algorithms to optimize and control SG and EI. For instance, a win‐or‐learn fast policy hill‐climbing (WoLF‐PHC) algorithm was combined with the idea of the time tunnel–based RL algorithm (ie, the SARSA algorithm) to construct an integrated WoLF‐PHC(λ) algorithm in Xi et al This algorithm was used to acquire the equilibrium solution of multiregional smart generation control to solve the problem of stochastic disturbances caused by the integration of large‐scale and high‐penetration new energy resources that traditional centralized AGC methods cannot solve. Based on this, Xi et al further combined DL with RL to form a new DRL capable of action self‐optimization to solve the random disturbances caused by the large‐scale integration of new energy and distributed energy into an interconnected power grid.…”

Section: Hybrid Learningmentioning

confidence: 99%

“…These above-mentioned HL algorithms have been preliminarily used to good effect in the SG and EI fields. 12,[59][60][61]65,68…”

Section: Basic Principle Of Hlmentioning

confidence: 99%

See 3 more Smart Citations

A new generation of AI: A review and perspective on machine learning technologies applied to smart energy and electric power systems

Cheng

2019

Int J Energy Res

233

View full text Add to dashboard Cite

Summary The new generation of artificial intelligence (AI), called AI 2.0, has recently become a research focus. Data‐driven AI 2.0 will accelerate the development of smart energy and electric power system (Smart EEPS). In AI 2.0, machine learning (ML) forms a typical representative algorithm category used to achieve predictions and judgments by analyzing and learning from massive amounts of historical and synthetic data to help people make optimal decisions. ML has preliminarily been applied to the Smart Grid (SG) and Energy Internet (EI) fields, which are important Smart EEPS representatives. AI 2.0, especially ML, is undergoing a critical period of rapid development worldwide and will play an essential role in Smart EEPS. In this context, this study, combined with the emerging SG and EI technologies, takes the typical representative of AI 2.0—ML—as the research objective and reviews its research status in the operation, optimization, control, dispatching, and management of SG and EI. The paper focuses on introducing and summarizing the mainstream uses of seven representative ML methods, including reinforcement learning, deep learning, transfer learning, parallel learning, hybrid learning, adversarial learning, and ensemble learning, in the SG and EI fields. In this survey, we begin with an introduction to these seven types of ML methods and then systematically review their applications in Smart EEPS. Finally, we discuss ML development under the big data thinking and offer a prospect for the future development of AI 2.0 and ML in Smart EEPS. We conduct this survey intended to arouse the interest and excitement of experts and scholars in the EEPS industry and to look ahead to efforts that jointly promote the rapid development of AI 2.0 in the Smart EEPS field.

show abstract

Section: Hybrid Learningmentioning

confidence: 99%

Section: Hybrid Learningmentioning

confidence: 99%

Section: Hybrid Learningmentioning

confidence: 99%

“…These above-mentioned HL algorithms have been preliminarily used to good effect in the SG and EI fields. 12,[59][60][61]65,68…”

Section: Basic Principle Of Hlmentioning

confidence: 99%

See 2 more Smart Citations

A new generation of AI: A review and perspective on machine learning technologies applied to smart energy and electric power systems

Cheng

2019

Int J Energy Res

233

View full text Add to dashboard Cite

show abstract

“…It can be found that when a fault occurred, both of its upstream LPs and downstream LPs will be affected to some degree conditionally. The degree relies on the operations state of the involving protective devices and AS [24,25]. In other words, considering a particular component, the effect of its downstream failures on its upstream LPs will more or less depend on its own operation state.…”

Section: General Model For Distribution Systemmentioning

confidence: 99%

Recursive Method for Distribution System Reliability Evaluation

Wang

Zhang

Li³

et al. 2018

Energies

View full text Add to dashboard Cite

This paper proposes a novel hybrid recursive method for distribution system reliability evaluation to deal with the computational limit and low-efficiency problem which exist in previously developed techniques as the system becomes larger. This method includes a bottom-up process and a top-down process, which are developed on the basis of a recursive principle, and the synthesis of both processes yield the reliability performance of each bus of the system. The bottom-up process considers the effects of downstream failures on upstream customers, and the top-down process considers the effects of upstream failures on downstream customers. In addition, a novel switch zone concept is defined and introduced into the bottom-up recursive process to save the computation cost. Besides, section technique (ST) and shortest path method (SPM) are employed to effectively simplify the recursive path and thus, the computation efficiency can be improved. The most significant feature of the proposed method over ST, SPM, failure mode and effect analysis (FMEA) is that it provides a more generalized equivalent approach to maximally simplify the network for reliable evaluation irrespective of the network topology. The effectiveness of the proposed method has been validated through comprehensive tests on Roy Billinton test system (RBTS) bus 6 and a practical-sized distribution system in China.

show abstract

Deep learning and reinforcement learning approach on microgrid

Kumar¹,

Kandasamy

Ramanathan

2020

Int Trans Electr Energ Syst

View full text Add to dashboard Cite

Microgrid is a new era in the power system and it has more scope of investigation on research. Due to an increase in demand and future expansion of the power system, analyzing the complexities of the network becomes a challenging task. Artificial intelligence plays a vital role in resolving such issues in a microgrid in various aspects. Owing to the rapid growth of periodical update in computational cost reduction, enhanced data analysis-based algorithm artificial intelligence enters into new epoch Artificial Intelligence AI 2.0. Based on such approach, machine learning has been evolved as AI 2.0 initially. Now, it develops branches like deep learning, reinforcement learning, and a combination of both deep reinforcement learning algorithms. These algorithms are precise to attain higher priority in decision-making under a complex network. This paper deals with numerous challenges of the above algorithm to state the significance of AI 2.0 and summarization of their application toward microgrid is useful to analyze the power system. K E Y W O R D S deep learning, deep reinforcement learning, microgrid, reinforcement learning 1 | INTRODUCTION Microgrid is a cluster of distributed generators (DG), stored energy system, local loads along with protecting and monitoring devices. Microgrid has its own specialty in operating under two modes of operation: autonomous and grid-connected mode, mode selection attained by the use of static transfer switch that bridges operation selection in microgrid. The presence of multiple DGs creates a distribution access problem in multipoint is transformed into a concentrated

show abstract

Smart generation control based on multi-agent reinforcement learning with the idea of the time tunnel

Cited by 92 publications

References 17 publications

A new generation of AI: A review and perspective on machine learning technologies applied to smart energy and electric power systems

A new generation of AI: A review and perspective on machine learning technologies applied to smart energy and electric power systems

Recursive Method for Distribution System Reliability Evaluation

Deep learning and reinforcement learning approach on microgrid

Contact Info

Product

Resources

About