Syngas Compositions Prediction by Neural Estimator Based on Multi-Scale Analysis and Dynamic PCA

Prediction of syngas compositions, the most important parameter in determining the product's grade and quality control of raw syngas produced in coal gasification, was studied. A neural estimator model based on dynamic principal component analysis (DPCA), back-propagation (BP) networks, and multi-scale analysis (MSA) was proposed to infer the syngas compositions from real process variables. DPCA was carried out to select the most relevant process features and to eliminate the correlations of input variables; multi-scale analysis was introduced to acquire much more information and to reduce uncertainly in the system; and BP networks were used to characterize the nonlinearity of the process. A prediction of the syngas compositions in Texaco coal gasification process was taken as a case study. Research results show that the proposed method provides promising prediction reliability and accuracy, and supposed to have extensive application prospects in coal gasification processes.