Simulating heat load profiles in buildings using mixed effects models

Abstract The landscape of buildings is a diverse one and long-term energy system planning requires simulation tools that can capture such diversity. This work proposes model for simulating the space-heating consumption using linear mixed-effects model. modelling framework captures noise caused by differences are not being measured between individual buildings; e.g. preferences their occupants. proposed uses outdoor temperature at hourly resolution; thus, able to predict intra-day variations as well longer effects. Given stochastic nature simulation, prediction interval be estimated, which defines region where any unobserved building will fall in. A whole year has been simulated compared out-of-sample measurements from same period. results show data virtually always inside estimated 90% interval. Norwegian schools, although general built other categories. amount detail allows planners draw varied realistic map future needs given location.