Intelligent data analysis for sustainable smart grids using hybrid classification by genetic algorithm based discretization

Smart grids, or intelligent electricity grids that utilize modern IT/communication/control technologies, become a global trend nowadays. Smart Grids which enable two-way communication and monitoring between service providers and end-users need novel computational intelligent algorithms for supporting generation of power from wide range of sources, efficient energy distribution, and sustainable consumption. Sustainability is of great importance due to increasing demands and limited resources. Many problem classes in sustainable energy systems are data mining, optimization, and control tasks. The aim of this paper is to focus on the existing electricity generation infrastructure, electricity consumption behavior of the consumers and the need for Smart Grid. The various methods that have been concentrated on are that of machine learning and data mining techniques that can be mapped to these smart grid environments. We use publicly available smart grid datasets such as: Residential Electricity consumption survey (RECS) dataset conducted in US; US SMART Home Microgrid dataset; Reference Energy Disaggregation dataset (REDD) and Almanac of Minutely Power (AMPds) aggregation Dataset in our analysis in order to optimize the energy consumption for sustainability. We utilize Gaussian process regression with Radial basis function (RBF) kernel, Best First Tree (BFTree) and Ordered weighted average fuzzy-rough K-nearest neighbor (OWAKNN) with equal width(EWD) and Genetic algorithm based Discretization (GAD) in our approach to predict and forecast the consumer behavior in electricity consumption. The result obtained in terms of errors will be an ingredient to make effective decisions for developing a sustainable smart grid infrastructure.