Grupo de Vibraciones Mecánicas

Permanent URI for this collectionhttp://48.217.138.120/handle/20.500.12272/1236

Browse

Search Results

Now showing 1 - 2 of 2
  • Thumbnail Image
    Item
    Four-layer spherical self-organized maps neural networks trained by recirculation to follow the phase evolution of a nearly four-year rainfall signal.
    (2019-01) Huggenberger, Dario Alberto
    This work is intended to organize a big set of time series of rainfall reanalysis built on the Fourier harmonic that corresponds to the 4.8year cycle of variability. To do that a self-organized map is implemented in four spherical layers trained by recirculation. The methodology is shortly described. It is used to organize time series on grid point around the Earth to follow the phase evolution of the signal. The phase and amplitude are the main criterion for organization. It is shown how the successive layers contain more general abstractions, their representativeness around the Globe and in regional scale. The main objective is to show how to use the neural network tool to follow the phase evolution of the signal around the Globe. It is described as an anomaly with highest amplitude in the central Pacific Ocean, this evolution and return after 4.8 years.
  • Thumbnail Image
    Item
    Four-layer spherical self-organized maps neural networks trained by recirculation to simulate perception and abstraction activity : application to patterns of rainfall global reanalysis
    (2018-07) Huggenberger, Darío Alberto
    Abstract This work is intended to organize a big set of time series. To do that a self-organized map is implemented in four spherical layers trainded by recirculation. This way tries to simulate aspects of perceotion and abstraction. The methodology and the fundamentals are describe. About the fundamentals, both from the problema point of view and the neural aspects as brain functioning, perception and abstraction concepts, psycho genetics and grouping ideas, and from the architecture of the network, scheme of training, spherical layers of the maps and algorithms involved in the iterative training, Then, it is used to organize a big set of time series of rainfall reanalysis on grid point around the Earth to show how it functions. After removing the average from the series, the annual cycle in shape and amplitude is the main criterion for oganization. It is shown how the successive layers contain more general abstractions, their representativeness around the Globe and in regional scale. It is compared with individual series in some points of grid. A posible change of behaviour is found in global scale around 1973 and with a variant in the methodogy a possible change in the annual cycle the same year.