Estimating missing data in historic series of global radiation through neural network algorithms Report as inadecuate




Estimating missing data in historic series of global radiation through neural network algorithms - Download this document for free, or read online. Document in PDF available to download.

Sistemas & Telemática 2016, 14 37

Author: El editor

Source: http://www.redalyc.org/articulo.oa?id=411546577006


Teaser



Sistemas & Telemática ISSN: 1692-5238 EditorSyT@icesi.edu.co Universidad ICESI Colombia editor, El Estimating missing data in historic series of global radiation through neural network algorithms Sistemas & Telemática, vol.
14, núm.
37, 2016, pp.
6-7 Universidad ICESI Cali, Colombia Available in: http:--www.redalyc.org-articulo.oa?id=411546577006 How to cite Complete issue More information about this article Journals homepage in redalyc.org Scientific Information System Network of Scientific Journals from Latin America, the Caribbean, Spain and Portugal Non-profit academic project, developed under the open access initiative Presentation Presentación Estimating missing data in historic series of global radiation through neural network algorithms opens this issue; this paper reports results obtained in a research project developed at the Universidad Francisco de Paula Santander (Cúcuta, Colombia).
This group of researchers deal with a frequent problem in data processing of meteorological data series: the absence of data in several time intervals.
They recognize the usefulness of two traditional methods – Autoregressive Integrated Moving Average (ARIMA) and Regression Analysis (interpolation) – but also their limitations under particular conditions, and address the problem from a neural networks perspective, using 10 years of data from the UFPS weather station: 125658 records of temperature, radiation and energy, with 9.98% of missing data. Estimación de datos faltantes en series históricas de radiación global mediante algoritmos de redes neuronales abre la edición, el artículo reporta los resultados del trabajo realizado por un grupo de investigadores de la Universidad Francisco de Paula Santander (Cúcuta, Colombia), quienes abordan un problema recurrente en los datos de las series de tiempo meteorológicas: la falta de información (datos) en algunos intervalos de tiempo.
Los investigadores, luego de revisar las limitaciones que tienen en alg...





Related documents