Use of Artificial Neural Networks Type MLP for the Prediction of Phosphorus Level from the Physicochemical Parameters of Sediments
Journal Title: IOSR Journals (IOSR Journal of Computer Engineering) - Year 2016, Vol 18, Issue 1
Abstract
Abstract: The present work is a contribution on the development of mathematical models for predicting the phosphorus contents based on the physicochemical properties of the sediments of the reservoir of the water damSidi Chahed (Meknes, Morocco). For that purpose, artificial neural networks (ANN) of type multilayer perceptron (MLP) was used. The data base used corresponds to 118 samples of superficial sediments taken fromseveral stations, and distributed in space and time at the level of the reservoir of the water power plant Sidi Chahed. This data base of the neural network, which was collected between 2010 and 2012, consists of the phosphorus content (variable to explain or to predict) and physical and chemical parameters as explicative or predictive variables. The performance of the mathematical models provided by artificial neural networks of typePMC was compared to the multiple linear regression model (MLP). This comparison showed that neural stochastic models are more efficient compared to the model on the MLR standard method, for the prediction ofthe phosphorus. This result can be explained by the existence of a non-linear relationship between the investigated physical and chemical parameters and the phosphorus contents of sediments from the dam's reservoir. The obtained results showed that the most efficient model is that of type PMC with the configuration [14-7-1], which uses, as transfer functions, the hyperbolic tangent function in the hidden layer and in the output layer, and learning algorithm of type quasi Newton BFGS.
Authors and Affiliations
Monyr Naoual, Abdallaoui Abdelaziz , El Hmaidi Abdellah
Keywords
Related Articles
- EP ID EP112139
- DOI -
- Views 103
- Downloads 0
Distribution of RSA Public Key with Security Device based Identity for Multi-Agent secured Distributed Computing System
Abstract: In Mobile Agent Technology, interoperability between agents is indispensably to secure the data from malicious agents under Multi-Agent System. To protect data and agents from malicious attacks, the multi-agent...
Security Enforcement with query routing Information Brokering in Distributed Information Sharing
Abstract: Information brokering system (IBS) shares information via on-demand access. IBS connect large-scale loosely federated data sources via a brokering overlay. It is a peer-to-peer overlay network that comprises di...
Analysis of blood samples for counting leukemia cells using Support vector machine and nearest neighbour
Abstract: analysis of blood samples for counting leukemia cells is discussed in this paper. support vector machine and nearest neighbour concept is presented in this paper. The identification of blood disorders,...
An Optimized Algorithm for Solving Combinatorial Problems using Reference Graph
Abstract: In this paper a novel optimized graph referencing method is proposed to tackle combinatorial problems with greater ease. The proposed algorithm is a modification of the existing backtracking algorithm....
Survey of different Web Application Attacks & Its Preventive Measures
Securing web is like securing our nation. Our whole world is Internet dependent In each sector internet is very much essential. So, internet security is very much promising task for us. More than 80% attacks are...