Sequential topology recovery of complex power systems based on reinforcement learning

Wu, Jiajing; Fang, Biaoyan; Fang, Junyuan; Chen, Xi; Tse, Chi K.

doi:10.1016/j.physa.2019.122487

Cited by 35 publications

(12 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In fact, the MDP provides a mathematical structure for modeling decision making in situations where outcomes are partly random and partly under the control of an agent [41]. Therefore, to employ RL in control theory, it is necessary to model the system under control as a finite MDP [24,42,43]. Given that the strategy of this paper is to control the frequency and voltage of the MG, in the following the control problem is modeled as a finite MDP and then solved using RL.…”

Section: Methodsmentioning

confidence: 99%

Assessing the Use of Reinforcement Learning for Integrated Voltage/Frequency Control in AC Microgrids

2020

View full text Add to dashboard Cite

The main purpose of this paper is to present a novel algorithmic reinforcement learning (RL) method for damping the voltage and frequency oscillations in a micro-grid (MG) with penetration of wind turbine generators (WTG). First, the continuous-time environment of the system is discretized to a definite number of states to form the Markov decision process (MDP). To solve the modeled discrete RL-based problem, Q-learning method, which is a model-free and simple iterative solution mechanism is used. Therefore, the presented control strategy is adaptive and it is suitable for the realistic power systems with high nonlinearities. The proposed adaptive RL controller has a supervisory nature that can improve the performance of any kind of controllers by adding an offset signal to the output control signal of them. Here, a part of Denmark distribution system is considered and the dynamic performance of the suggested control mechanism is evaluated and compared with fuzzy-proportional integral derivative (PID) and classical PID controllers. Simulations are carried out in two realistic and challenging scenarios considering system parameters changing. Results indicate that the proposed control strategy has an excellent dynamic response compared to fuzzy-PID and traditional PID controllers for damping the voltage and frequency oscillations.

show abstract

Section: Methodsmentioning

confidence: 99%

Assessing the Use of Reinforcement Learning for Integrated Voltage/Frequency Control in AC Microgrids

2020

View full text Add to dashboard Cite

show abstract

“…Being a typical complex network, the power network's resilience has also attracted the interest of many scholars [7,8]. The resilience of the power network refers to the ability of the power grid to resist interference and restore power promptly under inclement weather such as typhoons, heavy rain/snow, and earthquakes, or artificial attacks such as graphite bombs, high-altitude nuclear magnetic explosions, and computer viruses [9,10]. Reference [11] comprehensively reviewed the evolution and current status of the U.S. power network, and exploreed ways to improve the resilience of the power network.…”

Section: Backgroundsmentioning

confidence: 99%

A Power Dispatch Optimization Method to Enhance the Resilience of Renewable Energy Penetrated Power Networks

Huang

Zhang

et al. 2021

Front. Phys.

View full text Add to dashboard Cite

With the wide deployment of renewable energies, future power grids become more vulnerable to extreme environments. This paper investigates enhancing the resilience of power systems with high penetrations of renewable energies under emergencies. The resilience enhancement firstly is defined as maintaining as much electric energy to critical loads in a fixed number of post-disaster periods by properly coordinating the available resources. Then, an optimal decision-making method is proposed to maximize the power supply of critical loads and to minimize the instability risks due to the randomness of the output power of renewable energies. The power consumption of loads, charging/discharging power of power storage plants, power generation of generators, and spinning reserve ratios of the renewable energy at each period are taken as decision variables. Constraints include spinning reserve, power flow constraints, and power consumption/generation limits. The interior-point algorithm is used to solve the formulated optimization problem. Numerical simulations verified the effectiveness and superiority of the proposed optimization method in boosting grid resilience after disasters. It is also found that a balance should be sought between decreasing stability risks and increasing the power supply benefit in extreme environments.

show abstract

“…Its identification is similar to vital node and link identification work (Lu et al, 2016, Du et al, 2017. However, it focuses on minimising the damage of vital node , Buldyrev et al, 2010, Wu et al, 2019, Brummitt et al, 2012, Cai et al, 2016.…”

Section: Related Workmentioning

confidence: 99%

Identifying and mitigating security risks for secure and robust NGI networks

Lever

Kifayat

2020

Sustainable Cities and Society

View full text Add to dashboard Cite

show abstract

Sequential topology recovery of complex power systems based on reinforcement learning

Cited by 35 publications

References 28 publications

Assessing the Use of Reinforcement Learning for Integrated Voltage/Frequency Control in AC Microgrids

Assessing the Use of Reinforcement Learning for Integrated Voltage/Frequency Control in AC Microgrids

A Power Dispatch Optimization Method to Enhance the Resilience of Renewable Energy Penetrated Power Networks

Identifying and mitigating security risks for secure and robust NGI networks

Contact Info

Product

Resources

About