Available Thermal Comfort Models in THESEUS-FE

Thermal comfort models allow to assess the thermo-physiologic interaction of a human body with its environment in terms of feeling cold, neutral or warm, comfortable or un-comfortable. Global models consider the complete thermal state, and local models hold for certain body parts, e.g. for the seat contact zone at the human back. In simulations those comfort models are part of the post-processing, because temperatures or heat fluxes must be derived first. Mathematical models for thermal manikin simulations and comfort models should be compatible, e.g. the meaningful use of local comfort models requires manikins with well validated response functions for the local skin temperatures. On the other hand, simple global comfort models (like PMV) make an advanced simulation based on Fiala's manikin more or less dispensable.

Stationary comfort models typically assess the thermal sensation level (cold-hot) via given boundary conditions. Such suggestions hold for moderate boundary conditions that change in a quite slow manner. Dynamic models (like Zhang/Fiala) consider time-changing reactions on thermal boundary conditions, typically human skin temperatures and their derivations in time (dTsk/dt). That's how thermal comfort indices might depend on the previous history, too. A well known phenomenon is the following: A cold human entering a warm bath feels hot during the first period of time, later on (with rising skin temperatures) the thermal sensation gets moderate and the feeling changes from "to hot" to "warm". Dynamic effects remain completely unconsidered in stationary models.

Today, the most advanced model is based on Zhang. It considers time-dependent skin temperatures as an input for different body parts. That's why it is an ideal extension to Fiala's thermal manikin that guarantees high quality for local thermal response functions, like skin temperatures. This model developed at the University of California (Berkeley) is based on a huge number of climate chamber test cases that helped to derive correlations between human temperatures and local thermal comfort indices.

	Fanger (1970) PMV-Index	Fiala (1998) DTS-Index	Assessment of equiv. temperatures (EN ISO 14505-2)	Zhang (2003) Local comfort model (available in 2009)
Input	activity level global boundary cond.: air- and wall temperature, air-velocity, humidity clothing	mean skin temperature core temp.	local heat loss values	local skin temperatures mean skin temperature core temperature
Validity	stationary, global	dynamic, global	stationary, local + global, 6 assessment regions	dynamic, local + global, 13 body parts
Remarks	not coupled with thermal manikin response	DTS similar to PMV	differing assessment for summer and winter clothing	model also provides max. thermal comfort value ⇒ applicable for optimization
Handicap	not applicable for contact boundary conditions model requires global cloth. definition (clo-value)	less validated for dynamic load cases	compared with Zhang: local comfort predictions are quite undifferentiated	very complex model results sometime not transparent ("black box")
Output (Indices)	global therm. sensation on a 7-step-scale -3 .. cold -2 .. cool -1 .. slightly cool 0 .. neutral +1 .. slightly warm +2 .. warm +3 .. hot		local therm. sensation and comfort on a 5-step-scale 1 .. too cold (uncomfort.) 2 .. cold (but comfort.) 3 .. neutral (comfortable) 4 .. warm (but comfort.) 5 .. too warm (uncomfort.)	global and local therm. sensation on a 9-step-scale, from -4 (very cold) to +4 (very hot) thermal comfort on a 9-step-scale, from -4 (very uncomfortable) to +4 (very comfortable)