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

摘要

Smart grids, or intelligent electricity grids that utilize modernIT/communication/control technologies, become a global trend nowadays. SmartGrids which enable two-way communication and monitoring between serviceproviders and end-users neednovel computational intelligent algorithms for supportinggeneration of power from wide range of sources, efficient energydistribution, and sustainable consumption. Sustainability is of greatimportance due to increasing demands and limited resources. Many problemclasses in sustainable energy systems are data mining, optimization, andcontrol tasks. The aim of this paper is to focus on the existing electricitygeneration infrastructure, electricity consumption behavior of the consumersand the need for Smart Grid. The various methods that have been concentratedon are that of machine learning and data mining techniques that can bemapped to these smart grid environments. We use publicly available smartgrid datasets such as: Residential Electricity consumption survey (RECS)dataset conducted in US; US SMART Home Microgrid dataset; Reference EnergyDisaggregation dataset (REDD) and Almanac of Minutely Power (AMPds)aggregation Dataset in our analysis in order to optimize the energyconsumption for sustainability. We utilize Gaussian process regression withRadial basis function (RBF) kernel, Best First Tree (BFTree) and Orderedweighted average fuzzy-rough K-nearest neighbor (OWAKNN) with equal width(EWD) and Genetic algorithm based Discretization (GAD) in our approach topredict and forecast the consumer behavior in electricity consumption. Theresult obtained in terms of errors will be an ingredient to make effectivedecisions for developing a sustainable smart grid infrastructure.