Modeling and classification of oil in a multilayer artificial neural network

Abstract

This research work presents modeling and classification of oil in a multilayered artificial neural network. Oil spill is one of the major sources of pollution to the sea which can be accidental or deliberate. In order to avoid this menace in our environment, early detection of oil spills and quick interventions are of paramount importance. In this research work, oil spills classification system based on laser fluorosensor spectra data was modeled and simulated. Artificial Neural Network (ANN) toolbox in Matlab/Simulink with MLP (multi-layer perceptron) based supervised architecture was used for the simulation. The network was trained to understand numerous spectra data of laser fluorosensor for different oil spill products (light oil, medium oil, and heavy oil) and other backgrounds (water, sand and stone). The trained network was tested using data set to the network. It was found that the ANN with MLP based supervised architecture performed well when the number of neurons in hidden layers is the same and an average of 100% classification result was achieved.  It was also observed that the network behaved badly and could not generalize well when the number of neurons in the two hidden layers differs. The performance, accuracy and precision were very poor in all the cases where the two hidden layers have different number of neurons.