Date Added: Dec 2010
This paper arises from an investigation into the use of an artificial neural network for modeling the retention process. The neural network model is used for those parts of the system where no physical model is readily available due to complexity. For the purpose of neural control structures, a direct and inverse neural model of retention process using three layer perceptron networks was created. The classical controller strategy has also been developed and analyzed. The performance tests for both the controllers (PID and ANN) were realized in the simulation environment Matlab /Simulink tool.