A methodology for the simulation of CPV systems is presented in four distinct sections: input, optics, uncertainty and electrical output.
In the input section, existing methods of describing the solar irradiation that is incident at the primary optical element of a CPV system are discussed, the inadequacies of the existing methods are explored and conditions of validity for their use drawn. An improved and spectrally extended model for a variable, spatially resolved solar image is arrived at. The model is used to analyse losses at the primary concentration device stage under varying solar profiles and air masses. A contextual analysis of an example Seattle based CPV system operating with constant solar tracking errors of 0.3-0.4° show a corresponding loss in isolation available to the optical system of 5-20%, respectively.
In the optics section, an optical ray trace model is developed specifically for this work. The optical ray trace model is capable of the spectrally resolved ray tracing of all insolation input models discussed above. Plano-convex and Fresnel lenses are designed, investigated and compared using each of the insolation models described in the input section. Common CPV component material samples for the plano-convex and Fresnel lenses are analysed for their spectrally resolved optical properties. The computational expense of high resolution spatial and spectral modelling is addressed by means of a spectrally weighted banding method. The optical properties parameter spectral weighting method can be applied to any arbitrary spectral band. The bands used herein correspond to the active ranges of a typical triple-junction solar cell. Each band shows a different spectral dependency. Banded beam irradiation proportions are shown to change by as much as 10% in absolute terms within the air mass range of 1 to 3. Significant variations in spectrally banded illumination profiles are found with the extended light source insolation model. These banded variations are mostly unaccounted for with the use of approximated insolation models, further compounding the argument for extended light source Sun models in CPV system simulations.
In the uncertainty section, the limitations of the manufacturing process are explored. Manufacturing tolerance errors from manufacturer datasheets are presented. These production uncertainties are used in the design of an erroneous plano-convex lens which is then analysed with the optical modelled presented in the optics section and compared to the ideal design specification. A 15% variation in maximum intensity value is found alongside a linear shift in the focal crossover point of approximately 0.2mm, although the optical efficiency of the lens remains the same. Framing manufacture errors are investigated for a square Fresnel lens system resulting in a linear shift of the focal centre of approximately 0.85mm. A process for the calculation of wind loading force on a CPV array is also presented. The process uses real 2 second resolution wind data and highlights the chaotic nature of loading force. A maximum force of 1.4kN was found on an example day for a 3m by 3m by 0.1m cuboid (i.e. CPV array); corresponding to a wind speed of approximately 13m/s, which is well within the typical operating range of a CPV tracking system.
In the electrical output section, a spatially resolved solar cell model is identified and used for the investigation of solar cell performance under the inhomogeneous cell illumination profiles produced in the uncertainty section. Significant differences in the maximum power point of the cell IVs are found for the ideal and erroneous system illumination profiles. Approximately, a 15% variation is found in the plano-convex lens example, with a relative difference of 4% attributable to illumination profile distortion, and a 6% variation in the module framing component example. These results further highlight the need for the consideration of production uncertainties in CPV system simulation.
Funding
none
History
School
Mechanical, Electrical and Manufacturing Engineering
This work is made available according to the conditions of the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0) licence. Full details of this licence are available at: https://creativecommons.org/licenses/by-nc-nd/4.0/
Publication date
2015
Notes
A Doctoral Thesis. Submitted in partial fulfilment of the requirements for the award of Doctor of Philosophy of Loughborough University.