[Radiance-general] Greenhouse energy modeling

[Greg Ward]

Hello Valerie and welcome,

Radiance defaults to RGB-type calculations, but these are just values and you are free to choose different units or interpretations as fit your needs.  The math does not change.  Just think of these as three radiation channels that correspond to the reflectances and lighting values you have used them for.  There is nothing special in the simulation that says these have to correspond to visible wavelengths.

For your purposes, the photon map extension might indeed be useful.  However, the "classic" version will also be able to account for interreflections between surfaces, with certain limitations handling reflections from curved, specular surfaces.

It would help to frame your problem to define exactly what you mean by "light arriving at a point."  We usually specify an incident normal direction and compute a cosine-weighted average to arrive at irradiance or illuminance, but you may be after either a uniform spherical weighting or something similar.  I suggest looking into the "rsensor" program <http://radsite.lbl.gov/radiance/man_html/rsensor.1.html> as a possible calculation engine for your task.  However, I have not tried integrating this tool with the photon map, and I am not sure if this part is working.  (I would have to check.)

Cheers,
-Greg

> From: Valerie Tan <vlt26 at cornell.edu>
> Date: August 2, 2017 11:18:32 AM PDT
> 
> Hello,  
>  
> My name is Valerie Tan and I am a undergrad designing greenhouse energy modeling software for the Cornell Controlled Environment Agriculture group. I am a computer science major (Cornell Engineering, Class of 2019) working for the Cornell CEA group with Kale Harbick (researcher) and Kevin Ye (also an undergrad). I am new to using the Radiance libraries and have questions on the features and capabilities of Radiance. I have been, and will continue to look through tutorials, archived mailing lists and presentations.  
>   
> We are implementing a ray-tracing and energy modeling component of the NYSERDA funded GLASE project. See https://glase.cals.cornell.edu/ for a brief synopsis.  
>   
> I was given a big task of designing software that models a greenhouse and calculates the total amount of light (from all directions, not just one direction as in backwards ray-tracing) that reaches particular sample points, given the geometry, weather, time of day, artificial light configurations, etc. Imagine a typical greenhouse, with glass panels that reflect and refract light, and a rectangular plot where you grow the plants. We want to pick sample points on that rectangular plot and figure out how much energy reaches that point. Ultimately, we want to model and optimize everything such that all areas of the growing area receive optimal and equal amounts of light.  
>  
> My main question is, how do I make light calculations in micromoles instead of lumens or RGB values? From what I've seen, Radiance appears to have this type of focus, but I could be entirely wrong. (This is my first research position and that I am still in the learning process of understanding technical descriptions and tutorials.) The issue with lumens and RGB values is that they are calibrated for human vision and are very difficult to work with in terms of plant science, while micromoles measures the actual number of photons. It looks like the "Photon map extension user manual" from the website (https://www.radiance-online.org/learning/documentation/photonmap-user-guide) is a possible solution.  
>  
> I really appreciate any kind of advice on using the Radiance libraries. Thank you for your time and hope to hear from all of you. 
>  
> Best,  
>  
> Valerie 
> 
> -- 
> Valerie Tan 
> Computer Science
> College of Engineering
> Cornell University '19
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.radiance-online.org/pipermail/radiance-general/attachments/20170802/61b67258/attachment-0001.html>