A comprehensive overview on demand side energy management towards smart grids: challenges, solutions, and future direction

Energy Informatics

Table 8 Application of machine learning in DSM

References	Approach	Objectives
		Forecasting		Power consumption	Consumer comfort	Appliance control	Cost reduction
		Price	Load	Power consumption	Consumer comfort	Appliance control	Cost reduction
Giovanelli et al. (2017), Pal and Kumar (2016), Yang et al. (2018)	Support Vector Regression (SVR)	√	√
Weng and Rajagopal (2015), Weng et al. (2018)	Gaussian process regression		√	√
Tang et al. (2018)	Linear regression forecast	√		√
Bina and Ahmadi (2015a, b)	Gaussian Copulas		√	√		√	√
Mekhilef et al. (2012), Simmhan et al. (2013), Yang et al. (2018)	Tree-based	√	√
Goubko et al. (2016)	Bayesian learning				√
Cao et al. (2013)	K-means,			√
O’Neill et al. (2010), Wen et al. (2015)	Q-learning			√		√	√
Patyn et al. (2018), Ruelens et al. (2014), Xu et al. (2016a)	Use Fitted Q-iteration (FQI)					√	√