®®®® SIIA Público

Título del libro: Global Surface Temperature Model Using Coupled Sugeno Type Fuzzy Inference Systems And Neural Network Optimization

Autores UNAM:
BERNARDO ADOLFO BASTIEN OLVERA; CARLOS GAY Y GARCIA;
Autores externos:

Idioma:
Inglés
Año de publicación:
2015
Palabras clave:

Atmospheric composition; Atmospheric temperature; Carbon dioxide; Fuzzy neural networks; Fuzzy systems; Neural networks; Solar radiation; Surface properties; Atmospheric transparency; Carbon emissions; Cimate change; Fuzzy inference systems; Global surface temperature; Global temperatures; Neural network optimization; Temperature projection; Fuzzy inference


Resumen:

In this research, a model that projects the mean global temperature as a function of anthropogenic carbon emissions was generated with two fuzzy inference systems, sugeno type. We propose that the climatic system is energetically balanced, and the albedo, solar constant and atmospheric transparency are all constants. Nevertheless, we assume that the surface temperature varies when the CO2 concentration changes and depends on the system temperature itself. The second assertion states that any change in atmospheric CO2 concentration depends on anthropogenic carbon emissions and the system actual concentration. The fuzzy inference systems were optimized using artificial neural networks that adjust the parameters according to a different data base that the one that was used to create the initial system. So that, we assure to find the hidden patterns and avoid overfitting. The principal results of this work are the temperature projections under IPCC scenarios and the discovering of the historical data hidden patterns.


Entidades citadas de la UNAM: