Performance Evaluation of Machine Learning Algorithms for Stability Prediction in Smart Grids

A smart grid represents an advanced electrical network that incorporates modern technology to efficiently manage electricity generation, distribution, and consumption. Majorly, a smart grid’s stability is affected by voltage fluctuations and frequency deviations due to power demand. Due to this, there is instability in the grid, fluctuations in the energy supply, and the need for real-time adjustments and also, due to the combination of renewable energy sources. Therefore, the real-time monitoring of the smart grid as required. Thus, this paper aims to focus on the primary parameters and factors that significantly influence the stability of smart grids and these parameters are used to apply the neural networks to forecast the stability of the smart grid. In this project, machine learning algorithms are used for Decentralized Smart Grid Control (DSGC) systems, to forecast smart grid stability. This paper compares different machine learning algorithms that provide better accuracy values.