Robust Determination of Performance Loss Rate for Photovoltaic Systems; IEEE Sensors Letters; Vol. 8, iss. 9
| Parent link: | IEEE Sensors Letters.— .— Piscataway: IEEE Vol. 8, iss. 9.— 2024.— Article number 7004504, 4 p. |
|---|---|
| Autor principal: | |
| Autor corporatiu: | |
| Altres autors: | , |
| Sumari: | Title screen The performance loss rate (PLR) of the photovoltaic (PV) system quantifies the change in the system's energy yield over time. To determine the PLR, readings from different sensors obtained for a certain time period are processed to get the linear regression that reflects the changes in system performance measured by relationship between incoming irradiation and energy produced by the PV system. Ordinary least squares (OLS) provide acceptable regression only under homoscedasticity, where analyzed sensory data are normally distributed and have the same variance. In the presence of heteroscedasticity and outliers, OLS needs additional efforts to improve the data. We propose a way for constructing a linear regression for PV system performance raw sensory data by means of the robust interval fusion with preference aggregation method. The proposed approach is insensitive to heteroscedasticity and outliers in data under analysis, which is demonstrated on small size set of synthetic data and on real-life data. The approach also does not require special preliminary sensory data preparation. Текстовый файл |
| Idioma: | anglès |
| Publicat: |
2024
|
| Matèries: | |
| Accés en línia: | https://doi.org/10.1109/LSENS.2024.3441854 |
| Format: | Electrònic Capítol de llibre |
| KOHA link: | https://koha.lib.tpu.ru/cgi-bin/koha/opac-detail.pl?biblionumber=675001 |
| Sumari: | Title screen The performance loss rate (PLR) of the photovoltaic (PV) system quantifies the change in the system's energy yield over time. To determine the PLR, readings from different sensors obtained for a certain time period are processed to get the linear regression that reflects the changes in system performance measured by relationship between incoming irradiation and energy produced by the PV system. Ordinary least squares (OLS) provide acceptable regression only under homoscedasticity, where analyzed sensory data are normally distributed and have the same variance. In the presence of heteroscedasticity and outliers, OLS needs additional efforts to improve the data. We propose a way for constructing a linear regression for PV system performance raw sensory data by means of the robust interval fusion with preference aggregation method. The proposed approach is insensitive to heteroscedasticity and outliers in data under analysis, which is demonstrated on small size set of synthetic data and on real-life data. The approach also does not require special preliminary sensory data preparation. Текстовый файл |
|---|---|
| DOI: | 10.1109/LSENS.2024.3441854 |