A computational workflow for urban micro-simulation of buildings' energy performance

The micro-simulation of buildings’ energy performance at the urban scale en- tails the dynamic simulation of buildings’ energy demands in a spatially resolved way; and of the supply of energy to meet these demands. Two approaches have emerged to meet the increasingly important need for such simulations: repeated calls to detailed dynamic simulation software that was designed for simulating single buildings, or a single call to simplified dynamic simulation software that was designed for simulating arbitrary numbers of buildings simultaneously. In this paper we describe a workflow that is under development to support application of the latter approach at multiple scales, ranging from single buildings, through neighborhoods, to districts and whole cities. This approach is based on CitySim, which remains to our knowledge the most comprehensive dedicated urban energy micro-simulation software. This workflow focuses on the utilisation by CitySim of an emerging standard for the semantic attribution of 3D urban models called CityGML, the import and export of CityGML files from a 3D database called 3DCityDB, methods for the population of this database, and on the generalisation of CitySim to solve for scenes of different spatial extent and level of detail. This paper describes this new workflow and its potential application to the simulation of urban scenes of varying complexity, from a simple street canyon to an urban district. This workflow is to represents our first step in the development of an integrated urban modelling platform for the simulation of physical resource flows and the socioeconomic phenomena that influence them. We close this paper by discussing the challenges that await us in the development of such a comprehensive urban modelling platform; challenges that range from the preparation of data, interoperability through the exchange of data for scenes of contrasting scale, the orchestration of services utilising this data and the visualisation of resultant o