Photovoltaic energy systems are of significant importance in the face of increasing energy demand and the push for a higher percentage of renewables in the global energy mix. Takeup of PV systems is limited in part by the unavoidable loss of power generation during the night time. Analytical studies of Storage Integrated Solar Thermophotovoltaic (SISTPV) systems however, have shown viability as a means of night time energy production. This thesis investigates the development of a CFD model of the SISTPV system’s thermal storage material. Motivation for this development is the validation of results from an existing analytical model, and to further incorporate non-linear effects such as natural convection, which cannot be analytically modelled. Heat losses can also be easily modelled in the CFD solver.
The basis for the CFD model is the OpenFOAM software package and the meltFoam solver. Dynamic heat fluxes were implemented on the thermal storage material via the third party library, swak4foam. This allowed for an accurate comparison to be made between the analytical model described in and CFD solutions. Results for night time operation showed a good match between the analytical model and CFD results and consequently a paper containing these findings is in preparation for submission to Solar Energy.
The meltFoam solver’s convection model has been validated against previous experiments which made it an ideal candidate to explore non-linear effects. Further investigation demonstrated the influence of natural convection on the running times of the model. This led to the conclusion that below a critical heat flux at the absorber, an inverted geometry reduces the melting time of the system while maintaining similar night time performance. The analytical model’s prediction of the thermal storage material temperature following the melting process was also shown to be accurate.
The existing analytical model assumed adiabatic side walls. By incorporating heat loss into the CFD model, this assumption was shown to be valid, as operational time losses were shown to be less than 1% for high heat loss cases.
Based on the positive results in this thesis, research on SISTPV technology should continue to a prototyping phase in the near future. In order to reach this stage, the CFD model should be further refined by adding a heat flux boundary condition calculated on a cell-by-cell basis and an detailed analysis of convection driven melting should be carried out.