A wide variety of alkanolamines like mono ethanolamine (MEA), diethanolamine (DEA), N-methyldiethanolamine (MDEA) and blended amine solutions are used in gas treating plants.The principle advantages of tertiary over primary and secondary alkanolamines are; their lower heat of reaction with acidic gases and consequent lower heat requirement for solution stripping and their lower vapor pressure permitting use of higher concentration solutions in the treating plants ,But the low reaction rate of CO2 with tertiary amines limits the use of MDEA solutions .Piperazine is one of the activators that are frequently used in combination of methylediethanolamine solution in acid gas removal plants. A thermodynamic model is also tuned with the experimental data to be used for the prediction of the solubility of CO2 in MDEA+PZ solutions at wide range of temperatures, and pressures. ANNs are appropriate tool for nonlinear fitting. They can learn through experimental data and implementation of this method is simple. In ANN model in this work for calculating CO2 loading a two layer feed forward neural model with 7 neurons in hidden layer has been used, this ANN model has 4 inputs and 1 output.
