Inferential Estimation of Texaco Coal Gasification Quality Using Stacked Neural Networks

The robust inferential estimation of syngas compositions using stacked neural network was presented. Data for building non-linear models is re-sampled using bootstrap techniques to form a number of sets of training and test data. For each data set, a neural network model was developed which were then aggregated through principal component regression. To improve the robustness and accuracy of the neural networks, the neural estimator model was obtained by stacking multiple neural networks which were developed based on the reorganization of the original data. Model robustness is shown to be significantly improved as a direct consequence of using multiple neural network representations. The implementation of the model was presented and the model was applied to Texaco coal gasification system to predict the syngas compositions. Research results show that the proposed method provides promising prediction reliability and accuracy.