Sami Abulnoun Ajeel; Ghalib A. Ali
Abstract
The intelligent techniques are used successfully in a broad band of applicationsone of these applications is the cathodic protection system. Examples of thesetechniques used in cathodic ...
Read More ...
The intelligent techniques are used successfully in a broad band of applicationsone of these applications is the cathodic protection system. Examples of thesetechniques used in cathodic protection are fuzzy logic and genetic algorithms. Thepresent work aims to use the neural network to predict the minimum currentdensity required in impressed current cathodic protection to protect low carbonsteel pipe which have been related previously.[1].This work deals with choosing the best network architecture for cathodicprotection system. This step used multilayer feed forward network fourenvironment variables (concentration C%, temperature T, distance D and pH) asinput to identify the minimum current density as output in a feed forward networkstructure with one hidden layer using the practical results data for the learningprocess. The best number of neurons in the hidden layer is chosen by trial and errorand it is found to be 25 neurons. the decision function used is the tan trainingalgorithm with one variable learning rate. Then, neural network training is doneusing 25 data samples from the experimental data for the current density in theabove four variables conditions. The stopping criterion for training was to obtain asum square error of 0.001 or read 10000 Epochs. An (SSE) than 0.001 wereobtained after 5226 Epochs.Generalization test used 5 data samples taken from the experimental resultsother than those data samples used in the learning process to check theperformance of the neural network on these data. The SSE for these samples was0.0053 and it shows a good generalization results for our application. Thecomparison between the actual experimental output and the neural network out putafter the learning process are almost identical which indicates that good learningprocess was achieved.The practical results indicate that neural network system can be usedsuccessfully to obtain minimum cathodic protection current density to protect lowcarbon steel pipes.