Abstract
Supply and demand balance is required to assure the normal operation of microgrids. When unintentional islanding occurs in a microgrid, a load shedding (LS) control scheme is necessary to achieve the power balance and frequency stability of an islanded microgrid due to its limited power generation. However, the LS control of a microgrid faces two key challenges, i.e., how to determine the appropriate LS amount and LS objects in an islanded microgrid. To solve these problems, this paper proposes a coordinated LS control scheme based on Double-Q learning for an islanded microgrid. In the proposed coordinated LS control scheme, aiming at the LS amount problem in the process of unintentional islanding, this paper considers the relationship between the active power and frequency deviation of each distributed energy resource (DER) and uses the frequency deviation measured locally to calculate the LS amount and guide LS actions. Aiming at the problem of LS object selection in the process of unintentional islanding, this paper describes the LS control problem as a Markov decision process (MDP). The load priority is considered in a reward value function to ensure the uninterrupted power supply of critical loads in a microgrid. The two value functions of Double-Q learning are used to make the two Q value tables measure each other, which can reduce the overestimation of Q values and improve the accuracy of decision-making. The performance of the proposed coordinated LS control scheme is verified by a microgrid model based on a modified IEEE 13-bus system. The test results verify the feasibility and effectiveness of the proposed coordinated LS control scheme.
Highlights
Microgrids can operate in the grid-connected and islanded mode to ensure the continuous power supply [1]-[2]
To achieve the supply and demand balance and frequency stability in an islanded microgrid, a coordinated load shedding (LS) control scheme based on Double-Q learning is proposed
Compared with the IELS control scheme and the contribution-based load shedding (CBLS) scheme, the proposed control scheme can reduce the range of frequency fluctuations and ensure the power supply reliability of critical loads
Summary
Microgrids can operate in the grid-connected and islanded mode to ensure the continuous power supply [1]-[2]. The study in [15] proposed a centralized LS scheme for managing the power balance of an islanded microgrid This scheme uses a nonlinear model predictive control to achieve the automatic shedding of non-critical loads. The study in [22] proposed an intelligent multi-microgrid energy management method based on the model-free reinforcement learning technique This method realizes the maximization profit of selling power and the minimization of the peak-to-average ratio. Considering the operating environment information of a microgrid, the proposed LS scheme based on Double-Q learning is implemented to quickly eliminate the power deficit in a microgrid. The main contributions of this paper are as follows: 1) To solve the power deficit in an islanded microgrid, an LS scheme based on Double-Q learning is proposed.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
