engleski

Mathematical modelling of biomass gasification in fixed bed gasifiers

The aim of the research was to develop a mathematical model that is capable to predict process parameters with reasonable speed and accuracy in different and changeable operating conditions during biomass gasification. Process parameters such as fuel and air flow rate were considered as one of the model inputs which lead to prediction of other process parameters such as syngas temperature and syngas composition. Process dynamics were modelled and simulaiton results were analysed in order to enable further development of an on-line gasification process control concept. Model was designed to predict process parameters in different and changing operating conditions. For model development purposes different equlibrium models and artificial inteligence based models were utilised and their performance was analysed. Model prediction potential was validated on measurement data from a fixed-bed type gasification facility. Developed models were able to predict process parameters such as syngas temperature with average prediction error below 10% (R2 > 0.82) and syngas composition with average prediction error below 38% (R2 > 0.42).Dynamic modelling approach with active prediction error estimation was developed to predict process parameters in variable operating conditions. Models were further used to develop a control strategy that could improve process efficiency by 25%.

Biomass gasification, Numerical modelling, Artificial Neural Networks

nije evidentirano