Smart Grid Stability Prediction with Machine Learning

Abstract

Smart grids refer to a grid system for electricity transmission, which allows the efficient use of electricity without affecting the environment. The stability estimation of this type of network is very important since the whole process is time-dependent. This paper aimed to identify the optimal machine learning technique to predict the stability of these networks. A free database of 60,000 observations with information from consumers and producers on 12 predictive characteristics (Reaction times, Power balances, and Price-Gamma elasticity coefficients) and an independent variable (Stable / Unstable) was used. This paper concludes that the Random Forests technique obtained the best performance, this information can help smart grid managers to make more accurate predictions so that they can implement strategies in time and avoid collapse or disruption of power supply.