A dynamic advanced radiation exchange module for use in simulation of spaces with radiant systems
The surface temperature of radiant systems can be significantly different than the temperature of other room surfaces. Therefore, radiation heat exchange often becomes the dominant mode of heat transfer. To accurately predict the surface temperature of radiant systems, it becomes necessary to model radiative heat fluxes in detail, in order to evaluate the thermal environment in terms of energy and comfort. In this study, a detailed thermal simulation module was developed for prediction of transient heat flux and surface temperatures in rooms with radiant floor heating, taking into account the impact of solar radiation on exposed thermal mass. The studied system consists of a room with a south facing window and a concrete floor with embedded tubing, located in Indianapolis, IN. Emphasis is placed on the radiation exchange module, which employs the full radiosity method, variable solar gains, and tracking of window projections on floors which are treated as separate surfaces in an 8-surface enclosure. For the thermal analysis, the heat balance method is used to build a comprehensive thermal network and solved explicitly using a finite difference approach to calculate surface temperatures and quantify thermal comfort. Comparison with simplified methods (linearized and constant radiation coefficients and absence of absorbed solar gains) and with other software is also presented in detail. The dynamic simulation module is used to investigate different modes of dynamic control of the radiant system in order to satisfy the space heating load and maintain thermal comfort. Specifically, a proportional-integral and a predictive control algorithm were utilized and compared, having operative temperature as the controlled variable. The impact of control constants and time step selection are discussed and analyzed in order to achieve the best possible results. The results show that (i) without considering the impact of absorbed solar radiation, the simplified models underestimate the higher limits of operative and floor surface temperatures by up to 3.5 degrees Celsius compared to the advanced developed module; (ii) taking into account the sunlit dynamic projections on the floor, simplified models with optimized proportional integral control of the system predict that the operative temperature remains below the lower comfort limit for 2.65% of the time and above the higher comfort limit for 6.3% of the time. Using the advanced module and same control parameters, these values changed to 2.8% and 26.0% respectively, while temperature differences in floor surface temperatures reach 2.8 degrees Celsius. Combining the developed model with hourly or sub-hourly weather data may lead to more accurate prediction of the annual dynamic thermal response spaces with radiant floor systems and exposed thermal mass, better understanding of control guidelines, and further insight to the impact different radiation heat transfer models have on the heating energy consumption of radiant type systems.
Horton, Purdue University.
Off-Campus Purdue Users:
To access this dissertation, please log in to our