A novel population-based occupancy modeling approach for district-scale simulations compared to standard-based methods

• Published in: Building and environment Vol. 181; p. 107084
• Main Authors: Mosteiro-Romero, Martín, Hischier, Illias, Fonseca, Jimeno A., Schlueter, Arno
• Format: Journal Article
• Language: English
• Published: Oxford Elsevier Ltd 15.08.2020; Elsevier BV
• Subjects: Building occupant modeling; Buildings; Cooling; Deviation; Electric appliances; Electric power demand; Electricity; Electricity consumption; Energy demand; Energy simulation; Energy sources; Modelling; Occupancy; Occupant presence models; Peak demand; Peak load; Population-based studies; Renewable energy sources; Schedules; Space heating; Stochasticity; Urban areas; Urban building energy modeling; Urban building energy modeling; Building occupant modeling; Energy simulation; Occupant presence models
• ISSN: 0360-1323; 1873-684X
• DOI: 10.1016/j.buildenv.2020.107084

District-scale building energy models can be a powerful tool for the integration of renewable energy sources and efficiency measures in urban areas. One key limitation of these models, however, has been their rather simplified treatment of building occupants. Since it is their activities which create the needs for energy in an area, an improved analysis of the effects of occupants on demand at the district scale is needed. This paper presents a novel population-based approach (PopAp) inspired by agent-based transportation models, in which a population of occupants was defined based on class and employee registers and each was given an individual daily schedule. This approach was then used to assess the effect of occupant presence modeling on district-scale energy demand simulations by comparing the data-centric PopAp method to standard-based deterministic and stochastic approaches. The maximum number of occupants in the area was found to be 33% higher for the deterministic model compared to the data-centric PopAp results, a deviation that was especially pronounced in education buildings. The results for space heating, space cooling and electricity demand for lighting and appliances show that while the mean deviation between models on a yearly basis is within 10% for all demands, on an hourly scale the deviation for space cooling and electricity exceeded 15%. Given the importance of the hourly scale for peak demand prediction for technology sizing, more detailed occupant modeling approaches should be considered when planning energy systems. •A methodology to model occupant presence at the district scale is presented.•The methodology follows an activity-based approach inspired by transportation models.•Standard-based approaches were found to predict occupancies up to 33% higher.•The deviation between models was analyzed for different use types and time scales.•On an hourly basis the deviation exceeded 15% for cooling and electricity.