Empirical modelling of the solar spectral influence on photovoltaic devices for improved performance forecasting
Photovoltaic performance modelling is essential for the successful development of PV systems. Accurate modelling can inform system design and financing prior to construction, help with fault detection during operation, and improve the grid penetration of PV energy. Whereas the models to account f...
| Main Author: | |
|---|---|
| Format: | Thesis (University of Nottingham only) |
| Language: | English English |
| Published: |
2024
|
| Subjects: | |
| Online Access: | https://eprints.nottingham.ac.uk/77212/ |
| _version_ | 1848800975769305088 |
|---|---|
| author | Daxini, Rajiv |
| author_facet | Daxini, Rajiv |
| author_sort | Daxini, Rajiv |
| building | Nottingham Research Data Repository |
| collection | Online Access |
| description | Photovoltaic performance modelling is essential for the successful development of PV systems. Accurate modelling can inform system design and financing prior to construction, help with fault detection during operation, and improve the grid penetration of PV energy.
Whereas the models to account for the effects of broadband irradiance, temperature, and so forth on PV performance are well established, those for the influence of the solar spectrum, known as spectral correction functions (SCFs), suffer a range of limitations. Existing models are typically based on proxy variables used to represent the solar spectrum, which are restricted in the amount of information they contain on the prevailing spectral irradiance conditions. Furthermore, validation of these models is restricted to climates that are not representative of the UK, where a broader range of spectral irradiance conditions is experienced due to its high northern latitude and frequent overcast or partially overcast skies.
Some studies have explored the possibility of characterising measured spectra with parameters such as the average photon energy to develop SCFs. However, these studies are limited in terms of their validation scope, such as duration of field data and types of PV module, and extension to a predictive model. In this project, two new SCFs are developed and validated in two distinct climate regions for multiple PV technologies. The first is based on the average photon energy alone (f(APE)), while the second is based on both the average photon energy and the depth of the 650--670nm water absorption band (f(APE,e)). Using data from Go (Golden, Colorado, USA), the former is shown to cut the prediction error for aSi modules by around 40% relative to a single-variable air mass SCF and a double-variable air mass and clearness index SCF. The latter, f(APE,e), addresses issues raised in the literature regarding the reliability of APE as a spectral characterisation index. Using the same data, f(APE,e) is shown to cut the prediction error by up to 60% with respect to a comparable multivariable proxy SCF based on the air mass and atmospheric precipitable water content.
These results are also validated at a new test site built at the University of Nottingham as part of this project. Although the overall errors are greater due to site-specific system characteristics, the relative improvements achieved by the APE-based models with respect to the proxy-based models are maintained in both climate regions.
The proposed spectral correction approaches can be integrated into wider PV performance models to improve their performance forecasting accuracy. |
| first_indexed | 2025-11-14T21:00:06Z |
| format | Thesis (University of Nottingham only) |
| id | nottingham-77212 |
| institution | University of Nottingham Malaysia Campus |
| institution_category | Local University |
| language | English English |
| last_indexed | 2025-11-14T21:00:06Z |
| publishDate | 2024 |
| recordtype | eprints |
| repository_type | Digital Repository |
| spelling | nottingham-772122025-02-28T12:27:39Z https://eprints.nottingham.ac.uk/77212/ Empirical modelling of the solar spectral influence on photovoltaic devices for improved performance forecasting Daxini, Rajiv Photovoltaic performance modelling is essential for the successful development of PV systems. Accurate modelling can inform system design and financing prior to construction, help with fault detection during operation, and improve the grid penetration of PV energy. Whereas the models to account for the effects of broadband irradiance, temperature, and so forth on PV performance are well established, those for the influence of the solar spectrum, known as spectral correction functions (SCFs), suffer a range of limitations. Existing models are typically based on proxy variables used to represent the solar spectrum, which are restricted in the amount of information they contain on the prevailing spectral irradiance conditions. Furthermore, validation of these models is restricted to climates that are not representative of the UK, where a broader range of spectral irradiance conditions is experienced due to its high northern latitude and frequent overcast or partially overcast skies. Some studies have explored the possibility of characterising measured spectra with parameters such as the average photon energy to develop SCFs. However, these studies are limited in terms of their validation scope, such as duration of field data and types of PV module, and extension to a predictive model. In this project, two new SCFs are developed and validated in two distinct climate regions for multiple PV technologies. The first is based on the average photon energy alone (f(APE)), while the second is based on both the average photon energy and the depth of the 650--670nm water absorption band (f(APE,e)). Using data from Go (Golden, Colorado, USA), the former is shown to cut the prediction error for aSi modules by around 40% relative to a single-variable air mass SCF and a double-variable air mass and clearness index SCF. The latter, f(APE,e), addresses issues raised in the literature regarding the reliability of APE as a spectral characterisation index. Using the same data, f(APE,e) is shown to cut the prediction error by up to 60% with respect to a comparable multivariable proxy SCF based on the air mass and atmospheric precipitable water content. These results are also validated at a new test site built at the University of Nottingham as part of this project. Although the overall errors are greater due to site-specific system characteristics, the relative improvements achieved by the APE-based models with respect to the proxy-based models are maintained in both climate regions. The proposed spectral correction approaches can be integrated into wider PV performance models to improve their performance forecasting accuracy. 2024-03-15 Thesis (University of Nottingham only) NonPeerReviewed application/pdf en cc_by https://eprints.nottingham.ac.uk/77212/1/RD_thesis_review_6.pdf text/plain en arr https://eprints.nottingham.ac.uk/77212/7/Email%20-%20Daxini.pdf Daxini, Rajiv (2024) Empirical modelling of the solar spectral influence on photovoltaic devices for improved performance forecasting. PhD thesis, University of Nottingham. photovoltaic spectrum spectral correction mismatch factor modelling energy solar forecasting Isc |
| spellingShingle | photovoltaic spectrum spectral correction mismatch factor modelling energy solar forecasting Isc Daxini, Rajiv Empirical modelling of the solar spectral influence on photovoltaic devices for improved performance forecasting |
| title | Empirical modelling of the solar spectral influence on photovoltaic devices for improved performance forecasting |
| title_full | Empirical modelling of the solar spectral influence on photovoltaic devices for improved performance forecasting |
| title_fullStr | Empirical modelling of the solar spectral influence on photovoltaic devices for improved performance forecasting |
| title_full_unstemmed | Empirical modelling of the solar spectral influence on photovoltaic devices for improved performance forecasting |
| title_short | Empirical modelling of the solar spectral influence on photovoltaic devices for improved performance forecasting |
| title_sort | empirical modelling of the solar spectral influence on photovoltaic devices for improved performance forecasting |
| topic | photovoltaic spectrum spectral correction mismatch factor modelling energy solar forecasting Isc |
| url | https://eprints.nottingham.ac.uk/77212/ |