<HTML dir=ltr><HEAD> <META http-equiv=Content-Type content="text/html; charset=unicode"> <META content="MSHTML 6.00.6000.16414" name=GENERATOR> <STYLE>  </STYLE> </HEAD> <BODY lang=EN-US vLink=purple link=blue> <DIV id=idOWAReplyText4759 dir=ltr> <DIV dir=ltr>Hi Achim,</DIV> <DIV dir=ltr> </DIV> <DIV dir=ltr>Thanks for your comments.</DIV> <DIV dir=ltr> </DIV> <DIV dir=ltr>"very interesting stuff, you describe! As the internal loads are surely one of the most important factors for the calculation of cooling load – and at the same time one of the most “guessed” usually – making more realistic patterns usable for simulation is a step forward, surely. However, how do you see this in the context of dimensioning (of a building – U-value, g-value etc. – and of HVAC equipment)? Too many degrees of freedom may be contra-productive, here?"</DIV> <DIV dir=ltr> </DIV> <DIV dir=ltr>Well, that applies to many advanced options found in any simulation program. You're likely not going to switch on every available option at your very first simluation, rather than gradually enabling relevant features as you progress. I guess it simply depends on the importance of human factors in a project. An anecdote: In a recent field study at BC Hydro offices (the British Columbia utility), colleagues at the NRC (ref. Anca Galasiu) have shown that occupancy sensors contributed to approx. 50% savings in lighting use. Quite a mobile population. One's ability to simulate such mobility patterns, with relative ease and speed, and to estimate the significant energy savings simply by defining occupancy-sensors in every workstation may certainly be worth the effort at some point during a project. The cost-effectiveness of adding daylighting controls would, in this case, be reduced given the initial savings. Quantifying such patterns could hardly possible with aggregated diversity factors. Acces to field-based behavioural attributes and models, which in most cases can only apply if modelling individuals, is another reason why such developments can become useful. I have many more useful anecdotes ...</DIV></DIV> <DIV> <DIV class=Section1>   "In regard to your second-to-last slide and the question on “what is the incidence angle behind the louvers”, my opinion is: -          the *incidence* angle is always in regard to the sun (sun azimuth and altitude in regard to the façade considered). For such louvers we could (a) model them individually in ESP-r as obstructions – if obstructions were enhanced with reflection and maybe even transmission (glass louvers are all ‘en vogue’, too) and in this way have direct-direct and direct-diffuse (and diffuse-diffuse, of course) transmission and reflection for the louver system / space behind or (b) model them as Venetian type blinds. Such a model could / will (?) then take direct-direct and direct-diffuse (and diffuse-diffuse, again) transmission / reflection into account for the space behind the louvers. -          if the interaction of ESP-r with Radiance could include “translation” of Radiance results to energy analysis for usage by ESP-r, the louvers would probably be no problem at all … maybe runtime, though."   My question was rather a rhetorical one, and was hinting that there wasn't a simple solution using existing ESP-r methods. If you're describing detailed specular components, then solar rays may be hitting the glass from multiple angles. You last suggestion was indeed what we're thinking; that in certain cases, you could define a pair of 2-dimensional arrays of DDS sensor points spread over the TMC area (in front and back). Based on the Radiance description of the louvers and glazing component, the simulation settings and the density of sensors, the DDS daylight coefficients (DCs) could take some time to calculate (maybe 24hours for an entire builing with such fenestration systems), but subsequently coupling the DCs with weather data within an ESP-r simulation wouldn't take much more time. The surface-averaged irradiances from sensors behind the TMC would give the transmitted insolation, while the difference between front and back values would give the total absorbed and reflected. Of course, one could add more DDS sensors within the room to track what surfaces were receiving what. I haven't worked out all the details, but the latest series of emails on this list suggests that solutions are possible.   All the best Denis Bourgeois  </DIV></DIV></BODY></HTML>