A conceptual framework for biomass gasifier design using a semi-empirical model and heuristic algorithm

• Published in: Chemical engineering journal (Lausanne, Switzerland : 1996) Switzerland : 1996), 2022-01, Vol.427, p.130881, Article 130881
• Main Authors: Yan, Beibei, Zhao, Sheng, Li, Jian, Chen, Guanyi, Tao, Junyu
• Format: Article
• Language: English
• Subjects: Artificial neural network; Biomass gasification; Gasifier design; Product composition prediction; Thermodynamic equilibrium model
• ISSN: 1385-8947; 1873-3212
• DOI: 10.1016/j.cej.2021.130881

[Display omitted] •A novel gasifier design method was proposed by coupling T-ANN model and NSGA-II.•The T-ANN model was based on the mechanisms of the TE and ANN models.•The T-ANN model predicted biomass gasification products with superior accuracy.•The design method could effectively deal with multiple designing targets.•The design method could increase LHVsyn, Vyield and CGE by 6.95%, 1.9% and 8.3%. Biomass gasification is a cost-effective process that converts biomass into clean and combustible gaseous products. Gasifier design plays an important role in the performance of biomass gasification. This study proposed a novel gasifier design method which used the thermodynamic artificial neural network (T-ANN) model to accurately predict gasification products, and the non-dominated sorting genetic algorithm-II to search for optimal gasifier design parameters. 166 reported experimental data were used to establish the T-ANN model and evaluate its performance. The results showed that the semi-empirical T-ANN model performed better prediction accuracy and robustness than traditional theoretical and empirical models. The optimal model showed an accuracy within 8.81 ± 1.21% at 95% confidence interval. The satisfying performance could be attributed to accurate prediction of gasification temperature and a comprehensive consideration of influential parameters. Among various input parameters, gasification temperature contributed the most (>15%) to the prediction accuracy. Applicability of the proposed design method was demonstrated in designing a 200 kW biomass gasifier. It was also found that this design method was promisingly superior for tasks with multiple design targets. The design results show that the lower heating value of syngas, syngas yield and cold gas efficiency were increased by 6.95%, 1.9% and 8.3%, respectively. Hopefully, this study could shed a light on the application of state-of-the-art models and algorithms in biomass gasifier design and promisingly enhance the development of biomass gasification industry.