1-01　Effect of Temperature and Dilution Gas Concentration on the Prediction of Reaction Rates of Water Gas Shift Reaction Using a Neural Network

Abstract

The reaction rates of water gas shift reaction based on detailed chemistry from 1373 K to 1573 K were analyzed. The obtained data sets were used to train a neural network. For validation, temporal changes of mole fraction were simulated using the trained neural network, and the results were compared with those using detailed chemistry. Since the results of neural network were in good agreement with these of detailed chemistry, the neural network with information of nitrogen and temperature in the input layer was able to predict reaction rates with high accuracy.