Causal simulation for residential electricity consumption

Abstract

The dynamic model use causal diagrams pretends to identify all factors involved behavior from specific system. This paper introduces a dynamic model based on causal diagrams that look at recognize it behavior and household energy consumption. Methodology applied includes random variables that looking for reply stochastic behavior inside house over day, the intention to apply a causal model implies identify interaction and dependency existing between variables. That means conditional and immediately response, such as criteria reflects in binary variables use, that represents ON / OFF states such as Boolean states TRUE / FALSE. For simulation, is considered typically values from consumption in electrical devices for household average in Quito, use time and turn on probability under specifically conditions. Simulation is running at VENSIM specific software for dynamics models. Results shows that proposed methodology presents is 24.95 % error respect to real measurements.