Waste-to-energy as a tool of circular economy: Prediction of higher heating value of biomass by artificial neural network (ANN) and multivariate linear regression (MLR)
2022
Ezzahra Yatim, Fatima | Boumanchar, Imane | Srhir, Bousalham | Chhiti, Younes | Jama, Charafeddine | Ezzahrae M'hamdi Alaoui, Fatima
Circular economy is a global trend as a promising strategy for the sustainable use of natural resources. In this context, waste-to-energy presents an effective solution to respond to the ever-increasing waste generation and energy demand duality. However, waste diversity makes their management a serious challenge. Among their categories, biomass waste valorization is an attractive solution energy regarding its low cost and raw materials availability. Nevertheless, the knowledge of biomass waste characteristics, such as composition and energy content, is a necessity. In this research, new models are developed to estimate biomass wastes higher heating value (HHV) based on the ultimate analysis using linear regression and artificial neural network (ANN). The quality-measure of the two models for new dataset was evaluated with statistical metrics such as coefficient of correlation (R), root mean squared error (RMSE), mean absolute error (MAE) and mean absolute percentage error (MAPE). The methods developed in this work provided attractive accuracies comparing to other literature models. Additionally, it is found that the ANN, as machine learning method, is the best model for biomass HHV prediction (R = 0.75377, RMSE = 1.17527, MAE = 0.93315 and MAPE = 5.73%). Therefore, obtained results can be widely employed to design and optimize the reactors of combustion. In fact, the developed ANN software is a simple and accurate tool for HHV estimation based on ultimate analysis. Indeed, ANN is one of the most applicable and widely used software in the field of waste-to-energy.
اظهر المزيد [+] اقل [-]الكلمات المفتاحية الخاصة بالمكنز الزراعي (أجروفوك)
المعلومات البيبليوغرافية
تم تزويد هذا السجل من قبل National Agricultural Library