Solar and wind power have recently become a potential option in power systems and act significantly to meet load penetration demands. The present growth of such renewable energy sources has shown an exponential increase. The high penetration of such system helps a grid effectively meet its load in an irregular demand but also creates some disturbances in the grid due to frequent additions and detachments of load or source. The way by which the renewable energy sources usually work in the onβgrid mode is to be attached to and cut down from the grids without creating disturbances in a stable grid. Another important requirement is effective load management with fewer transmission losses. This article presents a detailed review of a microgrid and enumerates the possible methods for the analysis of the system, feature extraction, control methods, and options for machine learning. This paper examines the factors affecting the operations in a power system, their nature, interdependability, and controllability. It also inspects the various machine learning algorithms, their feasibility, and possible applications in power systems. The major contribution of the paper is the elucidation of expert system control methods for the performance improvement of solar PV assisted DC microgrids. The major objective of the paper is to provide an overview on various algorithms intended for the microgrid systems pertaining to its accuracy, precision, classification, prediction and forecasting.