Determination of Cloud Motion Applying the Lucas-Kanade Method to Sky Cam Imagery
The atmospheric conditions existing where concentrated solar power plants (CSP) are installed need to be carefully studied. A very important reason for this is because the presence of clouds causes drops in electricity generated from solar energy. Therefore, forecasting the cloud displacement trajec...
| Main Authors: | , , , |
|---|---|
| Format: | info:eu-repo/semantics/article |
| Language: | English |
| Published: |
MDPI
2020
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/10835/8410 |
| _version_ | 1789406670797930496 |
|---|---|
| author | Mondragón, Román Alonso Montesinos, Joaquín Blas Riveros Rosas, David Bonifaz, Roberto |
| author_facet | Mondragón, Román Alonso Montesinos, Joaquín Blas Riveros Rosas, David Bonifaz, Roberto |
| author_sort | Mondragón, Román |
| collection | DSpace |
| description | The atmospheric conditions existing where concentrated solar power plants (CSP) are installed need to be carefully studied. A very important reason for this is because the presence of clouds causes drops in electricity generated from solar energy. Therefore, forecasting the cloud displacement trajectory in real time is one of the functions and tools that CSP operators must develop for plant optimization, and to anticipate drops in solar irradiance. For short forecast of cloud movement (10 min) is enough with describe the cloud advection while for longer forecast (over 15 min), it is necessary to predict both advection and cloud changes. In this paper, we present a model that predict only the cloud advection displacement trajectory for different sky conditions and cloud types at the pixel level, using images obtained from a sky camera, as well as mathematical methods and the Lucas-Kanade method to measure optical flow. In the short term, up to 10 min the future position of the cloud front is predicted with 92% certainty while for 25–30 min, the best predicted precision was 82%. |
| format | info:eu-repo/semantics/article |
| id | oai:repositorio.ual.es:10835-8410 |
| institution | Universidad de Cuenca |
| language | English |
| publishDate | 2020 |
| publisher | MDPI |
| record_format | dspace |
| spelling | oai:repositorio.ual.es:10835-84102023-04-12T19:47:54Z Determination of Cloud Motion Applying the Lucas-Kanade Method to Sky Cam Imagery Mondragón, Román Alonso Montesinos, Joaquín Blas Riveros Rosas, David Bonifaz, Roberto cloud detection digitized image processing solar irradiance estimation solar irradiance forecasting solar energy sky camera tracking cloud motion Lucas Kanade optic flow remote sensing The atmospheric conditions existing where concentrated solar power plants (CSP) are installed need to be carefully studied. A very important reason for this is because the presence of clouds causes drops in electricity generated from solar energy. Therefore, forecasting the cloud displacement trajectory in real time is one of the functions and tools that CSP operators must develop for plant optimization, and to anticipate drops in solar irradiance. For short forecast of cloud movement (10 min) is enough with describe the cloud advection while for longer forecast (over 15 min), it is necessary to predict both advection and cloud changes. In this paper, we present a model that predict only the cloud advection displacement trajectory for different sky conditions and cloud types at the pixel level, using images obtained from a sky camera, as well as mathematical methods and the Lucas-Kanade method to measure optical flow. In the short term, up to 10 min the future position of the cloud front is predicted with 92% certainty while for 25–30 min, the best predicted precision was 82%. 2020-09-02T10:44:27Z 2020-09-02T10:44:27Z 2020-08-16 info:eu-repo/semantics/article 2072-4292 http://hdl.handle.net/10835/8410 en https://www.mdpi.com/2072-4292/12/16/2643 Attribution-NonCommercial-NoDerivatives 4.0 Internacional http://creativecommons.org/licenses/by-nc-nd/4.0/ info:eu-repo/semantics/openAccess MDPI |
| spellingShingle | cloud detection digitized image processing solar irradiance estimation solar irradiance forecasting solar energy sky camera tracking cloud motion Lucas Kanade optic flow remote sensing Mondragón, Román Alonso Montesinos, Joaquín Blas Riveros Rosas, David Bonifaz, Roberto Determination of Cloud Motion Applying the Lucas-Kanade Method to Sky Cam Imagery |
| title | Determination of Cloud Motion Applying the Lucas-Kanade Method to Sky Cam Imagery |
| title_full | Determination of Cloud Motion Applying the Lucas-Kanade Method to Sky Cam Imagery |
| title_fullStr | Determination of Cloud Motion Applying the Lucas-Kanade Method to Sky Cam Imagery |
| title_full_unstemmed | Determination of Cloud Motion Applying the Lucas-Kanade Method to Sky Cam Imagery |
| title_short | Determination of Cloud Motion Applying the Lucas-Kanade Method to Sky Cam Imagery |
| title_sort | determination of cloud motion applying the lucas-kanade method to sky cam imagery |
| topic | cloud detection digitized image processing solar irradiance estimation solar irradiance forecasting solar energy sky camera tracking cloud motion Lucas Kanade optic flow remote sensing |
| url | http://hdl.handle.net/10835/8410 |
| work_keys_str_mv | AT mondragonroman determinationofcloudmotionapplyingthelucaskanademethodtoskycamimagery AT alonsomontesinosjoaquinblas determinationofcloudmotionapplyingthelucaskanademethodtoskycamimagery AT riverosrosasdavid determinationofcloudmotionapplyingthelucaskanademethodtoskycamimagery AT bonifazroberto determinationofcloudmotionapplyingthelucaskanademethodtoskycamimagery |