PREDICTION OF OPTIMAL BATTERY CAPACITIES FOR SOLAR ENERGY SYSTEMS

Solar energy using photovoltaic cells is a highly feasible solution for modern renewable-powered residential buildings in terms of cost reduction of utility bills. The installation of solar PV systems and the (BESS) battery energy resource is the most popular energy cost minimization solution and will continue to increase rapidly. The size of the battery depends on the various energy profile factors and load profile factors. The energy profile factors include Annual energy consumption, Accumulated Energy, Energy Utilization from Grid, and Energy Injected into the grid. And the load profile factors are Average Load, Peak Load, Median Load, and Rated Photovoltaic power. The project aims at predicting the battery size by analyzing a dataset comprising the above-mentioned energy profile factors and load profile factors using machine learning techniques. Optimal sizing of the BESS is an essential prospect for nZEBs. Currently, the price of BESS is very high; therefore, in many countries, governments and grid operators offer several incentives to consumers. However, the cost of BESS is also expected to drop in the coming years. For classification, the project employs two distinct classifiers: Gaussian Naive Bayes (GNB) and Bernoulli Naive Bayes (BNB). By applying these classifiers within a supervised learning framework, the project leverages statistical methods to iteratively improve prediction accuracy KEYWORDS: Bernoulli Naive Bayes, Gaussian Naive Bayes, Prediction, Solar PV systems,