How to apply the DElight method in Honeybee energyplus simulation?

Dear friends,

Since I saw a paper that uses EnergyPlus to calculate the annual indoor lighting index, I thought of a problem, that is, how to use EnergyPlus to obtain the illuminance values of all reference points (more than 2) in indoor space.
After read the engineering reference, I find that DElight Daylighting Calculations can do this.

My questions are:

  1. Is there any way to apply this calculation into honeybee energyplus simulation process?
  2. How to apply the HB sensor grid from Rooms into DElight Daylighting Calculations?
  3. Can I use these rough and inaccurate illuminance values to calculate annual daylighting metrics?

Thanks for LBT teams!@chris @mostapha

I guess you could technically do it by editing the OSM generated by Honeybee using the OpenStudio SDK as described here. But I don’t know why you would make that much work for yourself when you can just use Radiance for a daylight calculation. An Annual Daylight simulation with Radiance is going to be way more accurate than anything EnergyPlus can give you and there’s no limit on the number of sensors. And you can use the illuminance results from the Radiance simulation to generate a schedule that dims the lights in an energy simulation using the HB Daylight Control Schedule component.

Thank you for your reply.
Because I plan to optimize a dynamic shading design. Different shading states are described based on HB shade transmittance schedules which cannot be recognized by Radiance. Maybe it can also be simulated with Radiance in other ways, but I think it will be more complex. So I try to use EnergyPlus to calculate daylighting as accurate as possible. If a large number of simulations are needed, the simulation will be much faster. By the way, can I call DElight by input additional string?

Thanks for clarifying, @Nevermore . I understand what you’re after but trying to use EnergyPlus for something that it’s not really designed isn’t really the best approach here. I can also assure you that, while the methods of modeling dynamic transmittance with Radiance are still complex, they are much, much simpler than what you are proposing here with EnergyPlus.

Ladybug Tools has ways of setting up Radiance models for simulation of dynamic transmittance (see this example file here) but we don’t currently have a recipe exposed that will model them correctly. However, we are in the process of exposing a 3-Phase Radiance recipe that will be capable of simulating this and it should be available before the next stable release.

In the meantime, the workflow for modeling dynamic transmittance with Radiance is conceptually very simple. You basically run two simulations: one with the shade and one without it. The difference between those two at each hour becomes the “shading contribution” for that hour. And you basically multiply that by the hourly transmittance and add it to the simulation with the shade to get the hourly illuminance values. The whole process becomes even simpler if there are only two transmittances that you need to model. In that case, you can just interpolate between two simulations at the two transmittances. Doing something like this manually with some native Grasshopper components should be a lot simpler than trying to edit your OSM using the OpenStudio SDK. And it will be a lot more accurate because you’re using the Radiance engine, which is actually designed for this type of thing.

1 Like

Yes, I compared the difference between Radiance and EnergyPlus in daylighting simulation. Splitflux algorithm obviously overestimates the illuminance value. I’ll try to use Radiance to model it.

Thanks for your advices!