A systems dynamics approach to the bottom-up simulation of residential appliance load

Abstract

Residential demand from real residences can be resource intensive to collect. There is need to generate synthetic residential load in energy research, and new approaches are welcome. Most of the simulation models of synthetic residential load that output realistic loads are tightly coupled to historic correlations. This paper presents a high-resolution simulation model that generates a residential appliance load using the tools of System Dynamics via a bottom-up approach. In addition to being realistic, the model aims to minimise historic coupling. Whilst the intermediary outputs of the modelling process are subjected to systematic scrutiny, the final output is validated by comparing statistical characteristics of the model’s output to a validated model and data from real residences. The aims of the model were sufficiently met, and the modelling approach shows potential to simplify; by driving the model on average frequency of appliance use instead of probability distributions of human activities. Other outputs from the model, specifically distribution of appliances’ activation and operation, as well as complexity are discussed. Some benefits of the model are also discussed especially with regard to cost of modelling, interpretability of model and potential for transdisciplinary research. This study represents the first attempt to develop a bottom-up simulation model of residential load based on a System Dynamics approach.