For building energy modeling, I use a software platform called EnergyPlus developed by the U.S. Department of Energy. Here is the URL that you could use to download EnergyPlus at no cost.

https://energy.gov/eere/buildings/downloads/energyplus-0

Hope this helps answer your question.

Professor Yehia Khalil, Yale University, USA

Fellow of the University of Oxford, UK

Link:

https://energy-models.com/what-is-energy-modeling-building-simulation

Dear Chen, 

I'm interested in your question. Have you tried the two tools Pr. Sircar and Pr. Khalil talked about bellow ? 

What I understand from your question is that you're working at city scale and you are wondering how precise your model has to be at the building scale, is that it ? 

Dear Prof. Khalil and Prof. Sircar, thank you for your information. Actually, I am in the EnergyPlus development team at LBNL, and very familiar with energy modeling.

Cecile Forgue, you are right. For city scale energy modeling, we only have the 2D footprint (polygon) + building height + number of stories. I tried to know any existing researches that evaluated the difference in simulation results when different geometry simplification methods are applied.

I think this is pertinent, since performance simulation beyond the building scale is still contentious. So, thank you for the question. Considering its implications is quite helpful.

At UNSW we have been working on library based Precinct Energy Modelling (PIM) to improve energy analysis - both on the data scheme and procedural implications. LBNL is at the forefront of this so, I am sure you are aware of the tool SBT-1 (Space Boundary Tool), formerly Geometry Simplification Tool (GST). It works in conjunction with a Simergy tool element that is being simultaneously adjusted through the work of @Richard See, whom I had the fortune to correspond with to test some features. Just like ASHRAE and other data libraries, it embraces the space type data concept with important additions. Both the concept and the features of this tool gave good insights. (Your questions are set to bold in what follows.)

"Use any other methods?"

"1. Use prototype building (typically rectangular) with same floor area, orientation and aspect ratio?"

The key to this is that you don't conceive a building as an extrusion of just one polygon per building. That way you could be way off with your performance simulation. Consider that in an urban setting many buildings have a podium volume and then significant setbacks for the upper story extrusions. Typically, the taller the building the larger the setback. Importantly, those extrusions generally have different building programs often split between residential and commercial use. Beyond that, floor area, aspect ratio and orientation are also important, of course.

"2. Create one zone per floor?"

You might want to be able to assign multiple programs to the same spaces or zones. This means that multiple zones per floor is a necessity not just spatially but also temporally. Consider a client who want the option for a large part of the podium level, or several stories of the tower portion, to be used not for a commercial but a residential building program. The usage schedules for those zones will be vastly different.

"3. Separate the internal zones with the perimeter zones (e.g., 15 ft/ 5 m width for the perimeter zones)?"

I like the idea of assigning approximately twice the storey height in horizontal distance from the perimeter for a separate zone. This would come from the daylighting need and the average angle sunlight hits the windows. But consider doubling this for the perimeter zones that are double-storey high. Also, change either figures for significantly different climate zones where the angle of the average sunlight is steeper or shallower. This is pertinent both within the US, e.g. as within the ASHRAE types, but many provinces within China might also have significantly different climate zones.