PhotoClass contributors Syira Al Husna, Elena Koubli and George Koutsourakis were each awarded oral presentations at this year’s UK national PV conference, The 13th PV Science, Applications and Technology Conference held at Bangor University, Wales.
The trio presented a wide range of work from their PhD studies – all coming to a close in the next few months. This included elements of Syira’s work on the polychromatic spectral response method developed during the PhotoClass project, Elena’s energy yield modelling for fault detection and diagnoses, branching from work conducted on her Researcher Mobility Grant visit to PhotoClass partner JRC, while George presented colleagues’ work on REG(LU)’s PhotoClass contribution to linearity characterisation of PV devices in workpackage 3 (detector characterisation). PVSAT is the largest UK conference for photovoltaic research and applications, and an important meeting of researchers and commercial PV operators. We wish all three students the best as they finalise their research theses.
Linearity paper abstract:
M. Bliss, B.V. Mihaylov, G. Koutsourakis, T.R. Betts, R. Gottschalg
Proceedings of the 13th Photovoltaic Science, Applications and Technology Conference
5th-7th April 2017, Bangor University, Wales
Especially for reference devices, the linearity of photocurrent over irradiance is an important characteristic that requires a low measurement uncertainty. This work investigates the uncertainty contributions when using a typical flash solar simulator with attenuation masks to determine the linearity characteristics of a device. Due to the complexity in measurement correlations, a Monte-Carlo simulation model was developed to estimate the final uncertainty. Results show that attenuation masks are not necessarily spectrally neutral and, if left uncorrected, this can significantly impact the measurement results. Furthermore, uncertainty in linearity is also dependent on the linearity of the sample under test itself. A shunted, non-linear device can have double the linearity uncertainty in low light conditions than a similar, linear sample