A unique radiometric measurement station design incorporating spectral and image data collection

Date
2022-04
Journal Title
Journal ISSN
Volume Title
Publisher
Stellenbosch : Stellenbosch University
Abstract
ENGLISH ABSTRACT: The performance of photovoltaic (PV) panels is determined under standard test condi tions (STC). However, in practical PV installations, it is rare for PV panels to operate under the STC at which they are rated. The variation in spectral irradiance under dif ferent atmospheric conditions significantly i mpacts t he p erformance of P V p anels. The output of a PV power plant is also significantly a ffected a s a l arge c loud b ank travels across the sky above. The passing clouds attenuate the received solar spectral irradiance and reduce the output power. Sudden changes in output power of a grid connected plant, as a result of clouds, may lead to voltage and frequency oscillations and result in system instability. Accurate short-term irradiance prediction models are important tools used to manage these power fluctuations. However, many irradiance prediction models and software tools used in the PV industry do not take these spectral impacts on PV performance into account. There is a lack of measurement stations dedicated to collecting the data necessary to improve these prediction models and software tools. A unique radiometric measurement station is designed and built in this thesis for the collection of data required by researchers to improve upon irradiance forecasting models and PV software tools. The instruments on the measurement station include: a pyranometer, a pyrheliometer, a wide field-of-view (FOV) spectral irradiance meter, a narrow FOV spectral irradiance meter and a digital camera. The instruments are mounted on a pan/tilt unit that is able to accurately track the position of the sun in the sky throughout the day for solar elevation angles greater than 10◦ . The data collected by the station includes temperature, humidity, air pressure, broadband irradiance and spectral irradiance. Images of the sky correlating to the various irradiance measurements are captured by the onboard digital camera. The measurement station is designed to be solar powered, compact and weather-resistant, allowing it to be deployed in remote locations for data collection. The station is configured t o c apture a m easurement s et o f d ata o nce e very 1 0 minutes in 4 distinct measurement positions: facing directly towards the sun, facing north with optimal tilt, facing north with a fixed 30◦ tilt, and facing north with a 0◦ tilt. The mea surement frequency and measurement positions can be configured by the user as necessary. The broadband irradiance measurements are validated by comparison to data collected by an independent measurement station nearby. The correlation between the spectral irradiance measurements and the AM1.5 reference spectrum is shown, as well as the vari ation of the solar spectrum during different seasons, different times of day and different cloud conditions. The measurement station is found to fulfil all requirements and successfully collects the data required to improve irradiance prediction models and PV software tools. This station has the potential to help optimize PV power stations as well as domestic PV installations.
AFRIKAANSE OPSOMMING: Die werkverrigting van fotovoltaïese (FV) panele word by standaard toets toestande bepaal. In praktiese toepassings is dit selde dat daar aan standaard toets toestande voldoen word. Die variasie in spektrale bestraling onder verskillende atmosferiese toestande beïnvloed die werkverrigting van FV-panele aansienlik. Die uitset van ’n FV-kragstasie word aansienlik geaffekteer soos groot wolkbanke deur die lug beweeg. Die verbygaande wolke verswak die sonbestraling wat ontvang word en verminder die totale FV uittree drywing. Skielike veranderinge in uittree drywing as gevolg van wolke kan lei tot spanning en frekwensie ossillasies, wat lei tot stelsel onstabiliteit. Akkurate korttermyn bestraling voorspellingsmodelle is belangrike gereedskap wat gebruik word om hierdie krag fluktuasies te bestuur. Baie bestraling voorspellingsmodelle en sagteware wat in die FV-industrie gebruik word, neem egter nie hierdie spektrale impakte op FV-prestasie in ag nie. Daar is ’n tekort aan meetstasies wat toegewy is om die data in te samel wat nodig is om hierdie voorspellings modelle en sagteware te verbeter. ’n Unieke radiometriese meetstasie is in hierdie tesis ontwerp en gebou. Die meetstasie is verantwoordelik vir die insameling van data wat deur navorsers benodig word om die huidige bestraling voorspellingsmodelle en FV sagteware te verbeter. Die instrumente op die meetstasie sluit in ’n pyranometer, ’n pirheliometer, ’n wye gesigsveld spektrale bestralingsmeter, ’n nou gesigsveld spektrale bestralingsmeter en ’n digitale kamera. Die instrumente is op ’n sonvolgeenheid gemonteer wat die posisie van die son regdeur die dag akkuraat kan volg vir sonhoogte hoeke groter as 10 grade. Die data wat die meetstasie insamel sluit temperatuur, humiditeit, lugdruk, breëband bestraling en spektrale bestraling in. Beelde van die lug wat verband hou met die verskillende bestralingsmetings word deur die digitale kamera aan boord gestoor. Die meetstasie is ontwerp om sonkrag-aangedrewe, kompak en weerbestand te wees, sodat dit in afgeleë plekke vir data-insameling ontplooi kan word. Die meetstasie is opgestel om een keer elke 10 minute data op te neem in 4 verkillende metingsposisies. Dit sluit in ’n meting wat direk na die son kyk, ’n meting wat noord kyk met optimale kanteling, ’n meting wat noord kyk met ’n vaste 30 grade kantel, en ’n meeting wat noord kyk met ’n 0 grade kantel. Die metingsfrekwensie en metingsposisies kan deur die gebruiker verander word soos nodig. Die breëband-bestralingsmetings word geverifieer deur dit te vergelyk met data wat deur ’n nabye onafhanklike meetstasie ingesamel is. Die korrelasie tussen die spektrale bestra lingsmetings en die AM1.5-verwysingspektrum word getoon, sowel as die variasie van die sonspektrum gedurende verskillende seisoene, verskillende tye van die dag, en verskillende wolktoestande. Daar word gevind dat die meetstasie aan alle vereistes voldoen. Die nodige data om die bestraling voorspellings modelle en FV-sagteware te verbeter word suksesvol versamel deur die gebruik van die meetstasie. Hierdie stasie het die potensiaal om FV-kragstasies sowel as huishoudelike FV-installasies te help optimeer.
Description
Thesis (MEng)--Stellenbosch University, 2022.
Keywords
Spectral imaging, UCTD, Radiation -- Measurement, Panels, Photovoltaic, Photovoltaic power systems, Image data mining
Citation