Affiliation:
1. Vellore Institute of Technology, Chennai, India
Abstract
The concept of microgrid (MG) is based on the notion of small-scale power systems that can operate independently or in conjunction with the larger power grid. MGs are generally made up of renewable energy resources, such as solar panels, wind turbines, and energy storage devices (batteries). Overuse of non-renewable resources causes depletion of the ozone layer and eventually leads to global warming. The classical techniques are not sufficient to solve the problem and require modern solutions like machine learning (ML) algorithms—a subset of artificial intelligence, and deep learning -a subset of ML algorithms. Though MGs have many advantages, they also have issues like high costs, complex management, and the need for better energy storage. ML can predict energy demand, optimize power flow to save money, improve energy storage management, enhances cybersecurity, and protects MGs from hackers. The chapter presented here provides a review of different ML techniques that can be implemented on MGs, their existing problems, and some improvised solutions to overcome the grid issues.
Reference83 articles.
1. Abbasi, M., & Tousi, B. (2018). A novel controller based on single-phase instantaneous pq power theory for a cascaded PWM transformerless statcom for voltage regulation. Journal of Operational and Automation Power Engineering, 6, 80–88.
2. Adams, L., et al. (2021). MG Performance Enhancement for Limited Budgets: A Case Study in Optimization. International Journal of Energy Management, 12(3), 211-224.
3. Microgrid energy management using deep Q-network reinforcement learning
4. AlpaydinE. (2020). Introduction to Machine Learning (3rd ed.). MIT Press.
5. Economic evaluation of microgrids