Comparisons between radial basis function and multilayer perceptron neural networks methods for nitrate and phosphate detections in water supply

• Main Authors: Md. Yunus, Mohd. Amri, Faramarzi, Mahdi, Ibrahim, Sallehuddin, Altowayti, Wahid Ali Hamood, Goh, Pei San, Mukhopadhyay, Subhas Chandra
• Format: Conference or Workshop Item
• Published: 2015
• Subjects: QA Mathematics
• Online Access: http://eprints.utm.my/id/eprint/60620/; http://ieeexplore.ieee.org/document/7244593/

This paper presents the comparisons between two models to classify nitrate and phosphate contamination in water supply based on artificial intelligence with multiple inputs parameters. The planar electromagnetic sensor array has been subjected to different water samples contaminated by nitrate and phosphate where output signals have been extracted. In the first method, the signals from the planar electromagnetic sensor array were derived to decompose by Wavelet Transform (WT). The energy and mean features of decomposed signals were extracted and used as inputs for an Artificial Neural Network (ANN) multilayer perceptron (MLP) and Radial Basis Function (RBF) neural networks models. The analysis models were targeted to classify the amount of nitrate and phosphate contamination in water supply. The result shows that the planar electromagnetic sensor array with the assistance of the MLP neural network method is the best alternative as compared to RBF neural network method.