Neural network prediction of parameters of biomass ashes, reused within the circular economy frame

Piotr Sakiewicz, Krzysztof Piotrowski and Sylwester Kalisz

Renewable Energy, 2020, vol. 162, issue C, 743-753

Abstract: Artificial neural networks were used for the prediction of three biomass ash fusion temperatures: initial deformation temperature IDT, hemispherical temperature HT and flow temperature FT based on chemical composition of the ash. Applicability of 400 neural network configurations (of linear, MLP, RBF and GRNN types) was verified statistically. Multilayer perceptron with 12 inputs representing fractions of the ash compounds, 11 hidden neurons and three outputs (IDT, HT, FT) proved to be the optimal neural model configuration. Statistical analysis suggested also, that considering intrinsic dispersion within the raw experimental data (literature data supplemented with the authors’ own results describing the halloysite addition effect), quality of the resulting 3-output IDT-HT-FT model (IDT prediction with R2 0.615, HT with R2 0.756 and FT with R2 0.729) could be regarded satisfactory for the identification and generalization of the discussed relationships. Analysis of the neural model sensitivity in respect to the input variables demonstrated, that the most important factors affecting all ash transition temperatures in the 3-output IDT-HT-FT model were: K2O, SiO2, CaO and Al2O3 fractions. Moreover, individual sensitivity in respect to IDT, HT and FT temperatures slightly varied (characteristics provided by independently established 1-output networks – IDT model, HT model and FT model, respectively). Statistically verified neural network working as the 3-output IDT-HT-FT model can be applied in various computational tasks in biofuels energy sector required by Industry 4.0 principles, as well as in the selected Circular Economy problems.

Keywords: Biomass combustion energy; Artificial neural networks; Circular economy; Ash fusion temperature; Industry 4.0; Neural predictive model (search for similar items in EconPapers)
Date: 2020
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (9)

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S0960148120313331
Full text for ScienceDirect subscribers only

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:eee:renene:v:162:y:2020:i:c:p:743-753

DOI: 10.1016/j.renene.2020.08.088

Access Statistics for this article

Renewable Energy is currently edited by Soteris A. Kalogirou and Paul Christodoulides

More articles in Renewable Energy from Elsevier
Bibliographic data for series maintained by Catherine Liu ().