An artificial neural network-based peak demand and system loss forecasting system and its effect on demand response programs

Abstract

This paper describes a neural network application to reduce the computational complexities in a smart grid environment by predicting peak demand and losses for the next operating day. Mitigating peak demands and losses could significantly reduce outage risks and increase costs savings, hence improving the reliability and efficiency of Electrical Service Providers (ESPs).