A cooperative multi-agent system to accurately estimate residential energy demand

摘要: In recent years much attention has been devoted to understanding and predicting when how occupants perform their daily routines use electric appliances within buildings. The purpose here is better estimate energy consumption. Despite efforts capture individual behavior with precision, models often neglect interact one another. Concretely, they do not accurately reproduce joint activities, such as having a meal or watching TV together, sole self-caring. Such inaccuracies, in turn, influence demand estimation, activities involve sharing non-sharing of electrical appliances. Therefore, this extended abstract, we propose cooperative multi-agent system, where interaction explicitly modeled by Interactive Markov chains. A preliminary study using data from five households Osaka, Japan suggests technique suited between than the traditional chain approach.