Prediction of trace compounds in biogas from anaerobic digestion using the MATLAB Neural Network Toolbox

D.P.B.T.B. Strik, A.M. Domnanovich, L. Zani, R. Braun, P. Holubar

Research output: Contribution to journalArticleAcademicpeer-review

114 Citations (Scopus)

Abstract

The outlook to apply the highly energetic biogas from anaerobic digestion into fuel cells will result in a significantly higher electrical efficiency and can contribute to an increase of renewable energy production. The practical bottleneck is the fuel cell poisoning caused by several gaseous trace compounds like hydrogen sulfide and ammonia. Hence artificial neural networks were developed to predict these trace compounds. The experiments concluded that ammonia in biogas can indeed be present up to 93 ppm. Hydrogen sulfide and ammonia concentrations in biogas were modelled successfully using the MATLAB Neural Network Toolbox. A script was developed which made it easy to search for the best neural network models' input/output-parameters, settings and architectures. The models were predicting the trace compounds, even under dynamical conditions. The resulted determination coefficients (R2) were for hydrogen sulfide 0.91 and ammonia 0.83. Several model predictive control tool strategies were introduced which showed the potential to foresee, control, reduce or even avoid the presence of the trace compounds.
Original languageEnglish
Pages (from-to)803-810
JournalEnvironmental Modelling & Software
Volume20
Issue number6
DOIs
Publication statusPublished - 2005

Keywords

  • model no. 1
  • reactor

Fingerprint

Dive into the research topics of 'Prediction of trace compounds in biogas from anaerobic digestion using the MATLAB Neural Network Toolbox'. Together they form a unique fingerprint.

Cite this