Comparing artificial neural network algorithms for prediction of higher heating value for different types of biomass

Abstract

The paper presents the comparative qualitative and quantitative analysis of twelve algorithms for training artificial neural networks (ANN) which predict the higher heating value (HHV) of biomass based on the proximate analysis (fixed carbon, volatile matter, and ash percentage). The twelve networks, with the same structure but different training algorithm, were fed with 318 experimental data triplets from literature for different biomass species and trained with 318 corresponding HHV values used for the supervised learning. Our comparative analysis showed that several algorithms resulted in ANNs generating outputs well correlated with the true measured values of the biomass HHV. Of those, Levenberg-Marquardt algorithm gives the best results in terms of mean squared error calculated on the training set data, while Bayesian regularization gives the best results in terms of regression. When applied to new datasets, unknown to the ANNs trained here, the highest accuracy of the HHV prediction was obtained by Conjugate Gradient, Powell/Beale Restarts training function. It ensured prediction based on the unknown datasets better than Levenberg-Marquardt algorithm. The described approach can be used for predicting the calorific values of different biomass species, including newly proposed ones, as well as for optimizing the HHV for both pure biomass and biomass mixtures.

Accurate Prediction of Drill Bit Penetration Rate in Rock Using Supervised Machine Learning Techniques Base on Laboratory Test Data

“…The MPDE was in fact a special form of Tweedie's deviance estimator, wherein the exponential value was 1. Equation 2 was thus utilized for computation purposes (Jakšić et al, 2023).…”

A Dynamic Recurrent Neural Network for Predicting Higher Heating Value of Biomass

“…Therefore, the applicability of biomass feedstocks in the context of energy conversion processes requires a variety of characterization and investigations [ 12 ]. Fuel heating values are generally reported in two ways: the lower (net) and the higher (gross) heating values [ 13 ].…”

“…The literature includes several empirical correlations to predict the biomass HHV from ash (A), proximate analysis (PA), ultimate analysis (UA), volatile matter (VM), and fixed carbon (FC). For a variety of biomass, carbon (C), oxygen (O), hydrogen (H), sulfur (S), and nitrogen (N) are the four major components [ 13 , 16 ].…”

“…When combined with domain expertise and experimental data, the computational methods based on fuzzy inference systems can provide accurate models, as shown by Akkaya [ 34 ]. The calorific values of both individual and blends of biomass feedstocks may be predicted by ANN, as shown by Jakšić et al [ 13 ]. Pattanayak et al derived three ANN models to determine the HHV of different bamboo biomasses [ 36 ].…”

Fluid–Solid Coupling Characteristics of Methane-Containing Coal during Borehole Extraction of Coalbed: Numerical Modeling and Experimental Research

“…Coalbed methane (CBM) is a high-heat, clean energy source, with a heat generation of about (3.35–3.77) × 10 5 J/m 3 of CBM, − equivalent to the heat of 1 kg of standard coal, and the pollution it produces is only 1/40 of that of oil and 1/800 of that of coal. − It can be seen that methane has a very large potential as a common fuel and chemical raw material. − For the mine types of CBM outburst mines and high CBM mines, the simultaneous extraction of coal and CBM is the best measure for the utilization of coal mine CBM resources and the control of CBM disasters. − The seepage law of CBM in coalbeds is one of the basic problems in the research field of coal mine CBM disaster prevention and control, which has important guiding significance for the basic theoretical research on CBM outburst, CBM extraction, and coal and CBM outburst prevention and control. − …”