Mahmood Alborzi; Seyed Amir Reza Abtahi
Volume 4, Issue 13 , June 2006, , Pages 41-66
Abstract
This paper uses neural network to predict corrosion rate. Corrosion can not modeled easily, because of wide range of causes either known or unknown. In mechanistic approach, physical, chemical and electrochemical reactions and processes are considered to model and predict. ...
Read More
This paper uses neural network to predict corrosion rate. Corrosion can not modeled easily, because of wide range of causes either known or unknown. In mechanistic approach, physical, chemical and electrochemical reactions and processes are considered to model and predict. But as stated before this models are not practically successful in prediction because of unknown parameters.
This paper uses genetic optimized neural network to predict corrosion rate. Among different neural networks, multi layer neural network with gradient descent learning algorithm has been selected. After developing the network, learning process has been done, using an oil refinery'S data. Then evaluation and test have been performed. After preparing the network Garson's algorithm and sensitivity analysis have been used for knowledge extraction.
According to results, neural network approach can predict corrosion rate with acceptable correlation coefficient (R) and mean squared error (MSE). Sensitivity analysis depicts the strength of each oil parameter influence on corrosion rate. Among these results, salt and sulphur are the most affecting parameters in corrosion rate.
Jamshid Salehi Sadaghiani; Seyed Amir Reza Abtahi
Volume 4, Issue 13 , June 2006, , Pages 89-122
Abstract
The purpose of this article is about soft computing and its different methods for modeling phenomena. Soft Computing refers to the evolving collection of methodologies used to build intelligent systems exhibiting human-like reasoning and capable of tackling uncertainty.
In this paper, we describe the ...
Read More
The purpose of this article is about soft computing and its different methods for modeling phenomena. Soft Computing refers to the evolving collection of methodologies used to build intelligent systems exhibiting human-like reasoning and capable of tackling uncertainty.
In this paper, we describe the neural networks approach in soft computing at first. Then, other approaches such as genetic algorithm and machine learning will be described. Since the main goal of building the model is knowledge extraction, finally, we will describe the various methods of knowledge and rule extraction from neural networks.
