Development of a Markov-Chain-Based Solar Generation Model for Smart Microgrid Energy Management System

The growth of smart power grids has complicated the balancing of supply and demand, the control and management of power outages, and the reduction of grid costs. In order to strengthen power grid reliability and reduce the risk of failure in production resources, demand-side management has gained particular importance. Microgrids are dramatically expanding due to the potential benefits of an increasingly reliable, sustainable, efficient, and environmentally friendly energy supply from renewable energy sources. In recent years, the air pollution crisis has attracted much attention. Among the factors that aggravate this crisis, pollution from fossil fuels has been playing a significantly greater role. As a result, the trend toward renewable energies, such as solar power, has grown considerably. As a gradually introduced alternative to fossil fuels, solar energy presents many benefits as well as numerous challenges. One of the most critical issues is the unreliability of this energy source. The management solution is to define an appropriate model for predicting how solar energy works and how to accordingly manage its storage. This paper proposes a solar generation model based on the Markov Chain, by which the formation of a cluster of chains predicts the power generation of solar cells. In order to reduce the prediction error, Markov chains are continuously trained by historical data. Evaluation results confirm the accuracy of the proposed model.