Artificial neural network based modelling approach for municipal solid waste gasification in a fluidized bed reactor
详细信息 查看全文
- 作者：Daya Shankar Pandeya ; Saptarshi Dasb ; Indranil Panc ; James J. Leahya ; Witold Kwapinskia ; witold.kwapinski@ul.ie" class="auth_mail" title="E-mail the corresponding author
- 关键词：LHV ; lower heating value ; LHVp ; lower heating value of product gas including tars and entrained char ; MSW ; municipal solid waste ; ANN ; artificial neural network ; LM ; Levenberg&ndash ; Marquardt ; WtE ; waste to energy ; FFNN ; feed forward neural network ; CFD ; computational fluid-dynamics ; ER ; equivalence ratio ; SCG ; scaled conjugate gradient ; BFGS ; Broyden-Fletcher-Goldfarb-Shanno quasi-Newton ; GDX ; gradient descent with momentum and adaptive learning rate ; MIMO ; multiple input and multiple output ; M
- 刊名：Waste Management
- 出版年：2016
- 出版时间：December 2016
- 年：2016
- 卷：58
- 期：Complete
- 页码：202-213
- 全文大小：3504 K
文摘
- •
ANN model is applied to predict the performance of MSW in a fluidized bed gasifier.
- •
Rigorous simulation is carried out to select the optimum ANN architecture.
- •
MISO vs MIMO ANN models are compared in both single and double layer architecture.
- •
Multiple Monte Carlo optimisation runs and data shuffling aid generalised training.
- •
Single layer MIMO model gives better predictive accuracy than other combinations.