The vacuum tube collector implemented and simulated in OptiCAD was also investigated experimentally at the outdoor test facility of the “Test Centre for Solar Thermal Systems (PZTS)” at the Fraunhofer ISE according to the guidelines of EN 12975-2. The tests include the measurement of the optical efficiency p0 and the IAM-values in the longitudinal and transversal plane.

The following problem has to be considered: The measured optical efficiencies are evaluated with respect to the measured global radiation. However, the simulations in OptiCAD are conducted by applying direct radiation. In order to compare simulated IAM-values with measured ones the specific fraction of diffuse radiation recorded during the measurement has to be taken into account. Moreover, during the IAM measurements — which are outdoor measurements — different levels of diffuse radiation may have occurred. This is considered in the comparison to the simulated IAM- values. In the measurements, also the diffuse radiation is not isotropic but may vary over the hemisphere and the surroundings within the viewing angle of the collector. However, in order to simplify the process the adjustment of the measured values is carried out by assuming a constant isotropic distribution. The tilt angle of the collector is also not taken into account.

Figure 6 shows the simulated transversal IAM of the vacuum tube collector over the incidence angle in transversal direction. Furthermore, the measured transversal IAM-values at their corresponding incidence angle of 20°, 40° and 60° are charted. Those values were adjusted as described above.

It can be seen that the measured values and the adjusted ones are almost exactly the same. It should be mentioned that in the measurements the fraction of diffuse irradiation is rather constant which contributes to the fact that the adjustments are small. Furthermore, the figure shows that the simulation result fits very well the measured data whereby the simulation output can be regarded as validated.

2. Conclusion

Ray-tracing simulations can be used as a powerful tool for the development of collectors [5]. Furthermore, the acceptance of diffuse radiation of a collector expressed as the diffuse IAM can be determined by utilisation of simulated IAM-values for direct radiation.

Once a certain collector type (single glazed flat-plate collector, vacuum tube collector with a given distance of vacuum tubes, with or without reflector, etc.) is modelled in OptiCAD, it is rather quick to carry out parameter variations and investigate different development possibilities. This may concern geometrical and material options. However, the result shown in Figure 6 requires detailed information about the collector parameters. The quality and exactness of the input data directly influence the quality of the results of ray-tracing simulations. With respect to the comparison of simulated against measured IAM-values, the model for the distribution of the diffuse radiation on the hemisphere has to be taken into consideration.

References

[1] Hefi, Stefan: Raytracing-Untersuchungen fur die Entwicklung von Prozesswarme-Kollektoren. In: 18. OTTI Symposium Thermische Solarenergie. Proceedings. Bad Staffelstein, 23.-25. april 2008, p. 416 — 421.

[2] Rommel, Matthias (Fraunhofer ISE); Weiss, Werner (AEE INTEC): Process Heat Collectors. State of the Art within Task 33/IV. Gleisdorf, Austria: AEE INTEC 2008.

Available: http://www. iea-shc. org/publications/downloads/task33-Process Heat Collectors. pdf

[3] Welford, W. T., Winston, R.: The Optics of Nonimaging Concentrators: Light and Solar Energy. San Diego, California; London: Academic Press, Inc. 1978

[4] TRNSYS 16: a TRaNsient SYstem Simulation program. Volume 5 — mathematical reference. Type 71, p. 5-343

[5] Geisshusler, Simon: IAM (Winkelfaktoren) Raytracing Simulationen von Solarthermischen Kollektoren. In: 17. OTTI Symposium Thermische Solarenergie. Proceedings. Bad Staffelstein, 09.-11. may 2007, p. 248 — 250.

Modelling complex systems within TRNSYS SIMULATION STUDIO

S. Kuethe*, C. Wilhelms, K. Zass, R. Heinzen, K. Vajen and U. Jordan

Kassel University, Institute for Thermal Energy Technology, 34125 Kassel, Germany
* Corresponding Author, solar@uni-kassel. de

Abstract

Solar thermal systems progress in complexity as the technology advances. Development and optimization of these systems is often carried out with the help of flexible numerical simulation tools like TRNSYS. With increasing complexity, maintaining flexibility and structure of the simulation model is an important issue. Therefore, Version 16 of TRNSYS comes along with the graphical user interface SIMULATION STUDIO. It assists the user in setting up the model. Furthermore, a graphical system representation is available automatically. However, for complex systems containing a lot of components and interconnections the system understanding and usability of the model gets difficult. Due to this fact this paper describes a new structure level introduced to SIMULATION STUDIO. Like TRNSYS itself, it also pursues a modular approach by organizing a model in several subsystems, where a subsystem is a collection of components. These subsystems are connected to each other with uniform interfaces only. This results in modularity on the component as well as on the subsystem level and allows the replacement of subsystems in SIMULATION STUDIO in a simple way. The solar thermal system developed within Task 32 of the Solar Heating and Cooling Programme of the International Energy Agency (IEA-SHC Task 32) has been implemented using the new approach.

Keywords: solar thermal, simulations, modelling complex systems, TRNSYS

1. Introduction

TRNSYS [1] is the most commonly used numerical simulation tool for development and investigation of solar thermal systems. Since Version 16, it comes along with the graphical user interface SIMULATION STUDIO. Besides the assistance of generating the TRNSYS ASCII input file, the graphical representation of the model is supposed to improve the maintenance of it. In addition, it should help to decrease the time for third party users to get familiar with new models. However, with increasing complexity of a model the graphical representation gets confusing due to the crossing of many interconnections between the several components (see Fig. 1). This fact leads to difficulties of the system understanding regarding the interaction between the components.

Portugal /

No_Parsons (Type9а) Internal Bams (Type 9a)

Fig. 1: Graphical representation of a complex model with a lot of overlapping interconnections between the

components in SIMULATION STUDIO.

Furthermore, the replacement of system parts like a collector loop of a solar thermal system is not possible in a simple way. SIMULATION STUDIO already has got two options in terms of structuring components. On the one hand components can be arranged on several layers, which can be hidden and unhidden by demand, and on the other hand components can be collected in macros, where the input and output dialog of a macro contains all inputs and outputs of the components within the macro. In the graphical representation the collected components of the macro are replaced by one symbol only. Unfortunately, both features only help to improve the design quality but do not provide simple to handle removals or transfers of complete system parts to other models. Therefore, a new structure level is introduced to SIMULATION STUDIO. It follows the modular concept of TRNSYS by defining subsystems, in which components are grouped. The subsystems are connected via uniform implemented interfaces containing all required inputs and outputs of the subsystems. As interfaces simple EQUATION blocks are used in SIMULATION STUDIO and connections of subsystems are realized via the defined interfaces only. The interfaces are implemented in a way that a replacement of subsystems is possible without deleting or reconfiguring graphical links between the subsystems.