Artificial intelligence for energetic potential analysis in Colombia
Climate change and pollution have driven the consumption of cleaner energy sources for the past years. Solar energy, has been one of the main demanded and efficient renewable energy sources. Therefore there has been great interest in innovation towards better allocation of the solar resource. In Col...
- Autores:
-
Acosta Cevallos, Paulina Michelle
- Tipo de recurso:
- Trabajo de grado de pregrado
- Fecha de publicación:
- 2020
- Institución:
- Universidad de los Andes
- Repositorio:
- Séneca: repositorio Uniandes
- Idioma:
- eng
- OAI Identifier:
- oai:repositorio.uniandes.edu.co:1992/51540
- Acceso en línea:
- http://hdl.handle.net/1992/51540
- Palabra clave:
- Inteligencia artificial
Recursos energúticos renovables
Radiación solar
Ingeniería
- Rights
- openAccess
- License
- https://repositorio.uniandes.edu.co/static/pdf/aceptacion_uso_es.pdf
| Summary: | Climate change and pollution have driven the consumption of cleaner energy sources for the past years. Solar energy, has been one of the main demanded and efficient renewable energy sources. Therefore there has been great interest in innovation towards better allocation of the solar resource. In Colombia, around 40% of the country is a ZNI (Not Interconnected Zone), meaning this locations dont have access to the national electrical wiring, generally due to the lack of electrical infrastructure as they are not easily accesible locations, with low capital and are situated in rural areas.Hence, they are highly dependent on solar energy in order to supply their energy consumption thus the importance of efficient allocation of the solar resource. Consequently, this paper proposes the creation of a solar irradiation open source atlas that provides information of the solar resource and other metereological variables on 64,000 points located in Colombia, separated by a 4km2 radius for each year from 1998 to 2019 in 30 minute intervals supplied by the NSRDB. Furthermore, this atlas would provide solar irradiation forecasting, predicting to different time horizons. Even though, solar irradiation forecasting has been widely studied, this paper explores a method in which information from different geographical points can be fusioned in order to improve the prediction of solar irradiation at a given point and time horizon.Then proposing a framework that considers the combination of spatial and temporal information by using attention modules in a LSTM network. |
|---|