Contribuye a Real-time demand response scheduling based on customized online learning of customers considering uncertainty. Tesis