ISSN: 1999-8716

Keywords : Artificial Neural Network


NEURAL NETWORK ANALYSIS FOR SLIDING WEAR OF 13%CR STEEL COATINGS BY ELECTRIC ARC SPRAYING

Ali Khudhair

DIYALA JOURNAL OF ENGINEERING SCIENCES, 2010, Volume 3, Issue 0, Pages 157-169

ABSTRACT:- Artificial Neural Networks (ANNs) are a new type of information processing technique based on modeling the neural systems of human brain. The potential of using neural networks in prediction of wear rate quantities of 13%Cr steel coating produced by arc spraying, has been studied in the present work. The material is subjected to dry sliding wear test using pin-on-ring apparatus at room conditions. Effects of normal load, sliding speed and time on wear rate have been investigated by using artificial neural networks. The experimental results were used to train ANN model successfully with accepted mean square error (MSE) of 0.00077504. The ANN predictions shows very good agreement with experimental values with correlation coefficient of 0.99778, thus ANN can be considered excellent tool for modeling complex processes that have many variables.
Keywords:- Artificial Neural Network; Wear; Coating

ESTIMATING Of CO2 CONVERSION IN FALLING FILM REACTOR USING ARTIFICIAL NEURAL NETWORK

Ahmed D. Wiheeb

DIYALA JOURNAL OF ENGINEERING SCIENCES, 2008, Volume 1, Issue 1, Pages 86-100

This paper presents the development of Artificial Neural Network (ANN) model for absorption process of CO2 gas using monoethanolamine (MEA) as a solvent in a falling film reactor. Although studies on ANN applications in chemical engineering in the literature are more concentrated on utilizing dynamic models, there has been an increasing trend for diverse application of ANN to model steady state systems. The feed-forward artificial neural network was adopted and trained by back-propagation algorithm. In this paper 216 sets of data are used to train and test the network. This study shows that ANN model with one hidden layer and nine neurons in the hidden layer gives a very close estimation of the CO2 conversion and there is high potential for absorption application of ANN model.