Energy management of Renewable Energy Sources based on Support Vector Machine

Abstract

In recent years, Renewable energy has covered a growing portion of the global electrical power demand. Microgrids are gaining popularity as a promising technology in order to include renewable energy sources in the distribution system. These resource integrations, which include solar arrays, wind turbines, diesel generators, and battery storage systems, combined with load demand. The efficient integration of these DGs in a microgrid faces several obstacles, including the accuracy of energy predictions for renewable energy sources such as wind and solar, energy management, and economic dispatch (ED). Based on the data of renewable energy output and load forecasting in microgrid achieving the optimization of microgrid dispatch. In this paper, a system based on the machine learning algorithm is implemented to forecast. support vector regression (SVR) is a regression model that has been used for optimization. SVR is a type of support vector machine that can learn regression functions and is an extension of the support vector machine classification approach. Enhance the precision of energy forecasts so that the electrical grid can run more efficiently. MIQP (mixed-integer quadratic programming) is used to define the whole problem, which can be solved quickly vai Gurobi Optimizer.