Estimation Of Higher Heating Value Research Articles

Biomass Higher Heating Value Estimation: A Comparative Analysis of Machine Learning Models

The research conducted focused on the capabilities of various non-linear and machine learning (ML) models in estimating the higher heating value (HHV) of biomass using proximate analysis data as inputs. The research was carried out to identify the most appropriate model for the estimation of HHV, which was determined by a statistical analysis of the modeling error. In this sense, artificial neural networks (ANNs), support vector machine (SVM), random forest regression (RFR), and higher-degree polynomial models were compared. After statistical analysis of the modeling error, the ANN model was found to be the most suitable for estimating the HHV biomass and showed the highest specific regression coefficient, with an R2 of 0.92. SVM (R2 = 0.81), RFR, and polynomial models (R2 = 0.84), on the other hand, also exhibit a high degree of estimation, albeit with somewhat larger modelling errors. The study conducted suggests that ANN models are best suited for the non-linear modeling of HHV of biomass, as they can generalize and search for links between input and output data that are more robust but also more complex in structure.

Estimation of the coal higher heating value for energy systems relied on ultimate analysis with machine learning techniques

The calorific value of solid fuels, also referred to as the gross calorific value (GCV) or the higher heating value (HHV), is a crucial property for its use as a fuel in energy systems. The HHV of coal as a resource can be predicted by more effective algorithms that use schedule information in engineering, like ultimate analysis, enabling fast decisions about its use as fuel in energy systems. The goal of this research was to acquire a global artificial prediction model relied on an interesting algorithm, a nonlinear model termed multivariate adaptive regression splines (MARS), in addition to the grid search (GS) optimizer, for characterization of coal HHV (output variable) using constituents of coal ultimate analysis: carbon (C), nitrogen (N), oxygen (O), hydrogen (H) and sulphur (S) (5 specific input variables). Moreover, a multivariate linear regression (MLR) and a multilayer perceptron-type (MLP) artificial neural network (ANN) were adjusted to the observed data as well as known empirical correlations for comparison purposes. The current investigation has produced two results. The MARS model is used to first demonstrate the significance (or strength) of each input variable on the coal HHV (output variable). Second, the most accurate predictor of the coal HHV was the MARS–relied approximation. In fact, using this method on coal testing samples resulted in a MARS regression with coefficients of determination and correlation coefficients for the coal HHV estimation of 0.9921 and 0.9960, respectively. The agreement between the data that were observed and those that were predicted using the GS/MARS–relied approximation proved that the latter had performed satisfactorily.

Assessment of lignin and extractive content based models used in estimation of higher heating value of lignocellulosics: Use of model mixtures

Cellulose, lignin and extractive material are mixed in certain proportions by having isolated from lignocellulosic materials, such as Zeyrek stem, hazelnut shell and Scotch pine, respectively. Their higher heating values (HHVs) are determined by using a bomb calorimeter system. Estimated HHVs are calculated by applying these mixture ratios to some Multiple (Non)-Linear Regression (M(N)LR) and Artificial Neural Network (ANN) models from the literature. MLR3 model is developed from the data of this study and this model reveals the highest R2 (0.974), lowest MAPE (0.012) and RMSE (0.278) values. The closest estimation accuracy to the MLR3 model is obtained from MLR2 (R2:0.972, MAPE:0.066 and RMSE:1.714) in the comparative analysis. MNLR and ANN equations containing quadratic terms are found to show deviations up to 132.6% (ANN3). It is attributed to the lower size and poor homogeneity of the individual group of samples from which model equations are developed.

Energy Potentials of Agricultural Biomass and the Possibility of Modelling Using RFR and SVM Models

Agricultural biomass is one of the most important renewable energy sources. As a byproduct of corn, soybean and sunflower production, large amounts of biomass are produced that can be used as an energy source through conversion. In order to assess the quality and the possibility of the use of biomass, its composition and calorific value must be determined. The use of nonlinear models allows for an easier estimation of the energy properties of biomass concerning certain input and output parameters. In this paper, RFR (Random Forest Regression) and SVM (Support Vector Machine) models were developed to determine their capabilities in estimating the HHV (higher heating value) of biomass based on input parameters of ultimate analysis. The developed models showed good performance in terms of HHV estimation, confirmed by the coefficient of determination for the RFR (R2 = 0.79) and SVM (R2 = 0.93) models. The developed models have shown promising results in accurately predicting the HHV of biomass from various sources. The use of these algorithms for biomass energy prediction has the potential for further development.

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)

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.

Estimation of Higher heating Value of Biomass from Proximate and Ultimate Analysis: A Novel Approach

Biomass is the organic matter formed by photosynthesis that occurs on the earth’s surface. They contain all forms of waste compost, including urban solid waste, municipal bio solids, animal wastes, forestry and agricultural wastes, and some industrial wastes. Efficient use of biomass oil would aid in the resolution of issues caused by fossil fuels. However, the biggest issue about using this energy is due to the gas composition of biomass material. As a result, properties of biomass are the critical parameter for assessing the fuel content of a special biomass substance in energetic applications. Gasification is the most mature thermo-chemical conversion technique available among the various methods of transforming biomass materials to bio resources. In this context, proximate and ultimate analysis has been used to classify two groups of biomass material that carry out in this experiment. The proximate analysis results have been obtained by the TGA technique while the ultimate analysis results will obtain by the GC mechanism. Then, based on the proximate analysis data various empirical equations containing linear and nonlinear terms were evaluated in order to predict the higher heating values (HHV) of the entire sample range. Since, the biomasses used in this analysis have different properties and fuel characteristics, the estimated HHV for the wood pellet sample are between 15.33 and 19.71 MJ/kg, while the rubber seed sample is between 15.18 and 18.64 MJ/kg. According to the experimental findings, the HHV of wood pellet is at around 2.95 MJ/Nm3 while the HHV of rubber seed of about 4.99MJ/Nm3. The comparison on the theoretical analysis have been show 0.19% compared to the results on wood pellet while the rubber seed have at around 2.07% difference compare each other. The experimental results on wood pellets, the findings reveal a 15.35% difference, while rubber seed indicates a 13.81% difference. Nonetheless, the finding and analysis on the properties, the results can be considered within reasonable limits.

A comparison of machine learning algorithms for estimation of higher heating values of biomass and fossil fuels from ultimate analysis

Higher heating value (HHV) is one of the most important parameters to consider while obtaining energy efficiently from fuels. It provides means to estimate the quality of the fuel. However, measuring the HHV of fuels requires advanced devices, is expensive and time consuming. Thus, sufficient estimation of the HHV is crucial for development of fuel technologies and numerous studies have been performed about this subject. In this study, several machine learning algorithms (DTR, SVR, GPR, RFR, Multi-Linear Regressions, and Polynomial Regression) were utilized to construct models for estimation of the HHV from the largest open dataset of ultimate analysis of fuels on a dry, ash-free basis. The mathematical analysis is conducted for numerous different solid, liquid, and gaseous fuels such as biomass, biochar, municipal solid waste, kerosene, gasoline, fuel oil, algae, natural gas etc. 10-fold cross validation method was used to assess the validity of the constructed models in an unbiased manner. The R2, adjusted R2, RMSE, N-RMSE, and AAE values of each model were computed for performance evaluations. Finally, the results were compared with the literature, and advantages and disadvantages of each method were discussed in terms of both computational complexity and prediction accuracy. The RFR and DTR models performed exceptionally well in estimation of HHV of all classes of fuels with R2 values of 0.9814 and 0.9664, respectively. In addition, the statistical values of the ultimate analysis for each constructed class of fuel are investigated for each class of fuel and an extensive Krevelen diagram was produced from the largest dataset to date, which illustrated the relationships between the atomic O/C ratio and the atomic H/C ratio of the fuels.

Forecast of the higher heating value based on proximate analysis by using support vector machines and multilayer perceptron in bioenergy resources

As biomass gets to be more relevant to energy feedstocks, the capacity to anticipate its Higher Heating Value (HHV) by more efficient algorithms from schedule information such as proximate analysis empowers quick choices around utilization in bioenergy. The present work studies a novel artificial smart model based on an interesting algorithm, relied on support vector machines (SVMs) jointly with the grid search (GS) optimizer, for characterization of HHV of raw biomass from parameters ascertained experimentally. Additionally, a multilayer perceptron (MLP) approach was built from the same experimental data for comparison objectives. The results of the current study are the relevance of each physico-chemical parameters on the raw biomass HHV through this novel model and forecasting the HHV of biomass. In this sense, when the novel model was applied to the observed dataset, a coefficient of determination and correlation coefficient equal to 0.8517 and 0.9229, were achieved for the HHV estimation, respectively. The importance of the use of learning machines is illustrated in the evaluation of the energy resources to energy systems to show an efficient algorithm for bioenergy purposes. The concordance between observed data and the GS/SVM–relied model indicated the good efficiency of the second.

A simple model for estimation of higher heating value of oily sludge

Oily sludge, as a waste containing a large amount of oil within the range of 10–90 wt%, has a potential to be utilized as a source of fuel. Heating value is of utmost importance in the assessment of thermal conversion of waste to energy and/or other valuable products. The aim of the current study is to develop correlations based on proximate and ultimate analysis to predict HHV of oily sludge accurately and easily. The analysis data were collected from previous studies conducted by global researchers. Four correlations were proposed by using the regression analysis method, one based on proximate analysis and three based on ultimate analysis. Among them, the correlations based on ultimate analysis show higher accuracy than the one based on proximate analysis and those established in the literature. Interestingly, the addition of variables has a negative contribution to the accuracy of correlations due to the collinearity between C and H. Thus, the simplest one (i.e. HHV = 0.431C + 1.234) only containing C presents the highest R2¯ and lowest bias error, especially for HHV higher than 20 MJ/kg (more preferable to be used as fuel source). The correlations were then validated with samples collected from local oilfield, showing a similar result with that obtained for the regression analysis.

Estimation of higher heating value (HHV) of bio-oils from thermochemical liquefaction by linear correlation

With increasing interest in renewable energy sources, particularly on biomass, and on the technologies to transform them into liquids fuels comes the need to assess its properties. One of the processes that can convert biomass into liquid fuel, bio-oil, is thermochemical liquefaction. However, the experimental determination of many properties can be time and budget consuming. Thus, the numerical models’ development to compute the higher heating value (HHV), with high accuracy, is needed. This work developed a model based on the linear correlation of a dataset composed of 54 samples. The carbon, hydrogen and oxygen content was correlated with HHV to develop a model. The model was compared with other from literature and finally validated with nine samples. Overall, the model exhibited excellent prediction accuracy. The absolute bias error was relatively low, 0.01%. Besides, the model’s good accuracy could be verified by the low absolute average error, 1.48%. The model’s good precision and performance was also demonstrated by the mean absolute error’s low value.

LHV and HHV prediction model using regression analysis with the help of bond energies for biodiesel

Lower heating value (LHV) is one of the most important fuel properties of an engine in terms of fuel economy and thermal efficiency. For this reason, various prediction models have been proposed in the literature to determine the LHV. In addition, higher heating value (HHV) prediction model, whose independent variables are generally preferred fuel characteristics such as density, viscosity, molecular weight, iodine value, cetane number, etc., has also been proposed. In this study, atomic bond structures in the fatty acid methyl and ethyl esters that make up biodiesel were investigated. Using the bond energy between atoms, LHV was predicted and HHV was calculated using the combustion reaction of esters. While determining the values of the bond energies, the experimental studies on esters were examined and the average bond energies of the ester were determined. LHV and HHV values of sixty-two esters were used as a training dataset for the second prediction model. The prediction equation for LHV was created by multiple linear regression analysis (R2 and Adj.R2 = 99.99%; p-value < 0.001) using the mass ratio of carbon (C), the mass ratio of oxygen (O), the mass ratio of hydrogen (H), and the unsaturation degree (UD). In this regression model, the standard error is 0.054 for sixty-two observations. Besides, the HHV estimation equation was obtained by developing the LHV equation. The validation of the model was carried out with LHV and HHV values measured in studies in the literature.

A heating value estimation of refuse derived fuel using the genetic programming model

Refuse Derived Fuel (RDF) makes an increasingly important contribution to sustainable waste management as an energy source in cement kilns. The most important parameter of RDF in an evaluation of its performance as a fuel is Higher Heating Value (HHV). The two methods of HHV determination are the direct method and the indirect method. The direct method requires the use of a calorimetric bomb and the indirect method requires ultimate or proximate analysis. As in the direct method, the ultimate analysis based indirect method requires the use of specific equipment and a skilled analyst. Most cement plants do not have special equipment. From this point of view, this study aims to predict the HHVs of RDF samples using the results of proximate analysis. Two Genetic Programming (GP) Models, namely GP Model #1 and GP Model #2 are used for the prediction. GP Model #1 denotes a modest nonlinear mapping function used for the prediction of HHVs, whereas GP Model #2 is a more inclusive nonlinear correlation analysis model as an improved version of GP Model #1. To assess the developed models, the test data is simulated and statistical results to the estimation of HHVs are reported as R2 equal to 0.9951 and 0.9988, Root Mean Square Error (RMSE) equal to 1.4126 and 0.6971 and Average Absolute Error (AAE) equal to 0.0543 and 0.0251, for GP Model #1 and GP Model #2, respectively. It can be seen that GP Model #2 may be confidently used for HHV estimation.

Estimation of biomass higher heating value (HHV) based on the proximate analysis: Smart modeling and correlation

In order to evaluate the potential and make a technical assessment of biomass energy, it is crucial to determine the higher heating value (HHV) of biomass fuels. Thus, multilayer perceptron artificial neural network (MLP-ANN) genetic algorithm-adaptive neuro fuzzy inference system (GA-ANFIS) differential evolution-ANFIS (DE-ANFIS), GA-radial basis function (GA-RBF), least square support vector machine (LSSVM) methods and an empirical correlation (multivariate polynomial regression (MPR)) were employed for the estimation of the HHV of biomass fuels. The comparisons of results show that GA-RBF and MPR models have higher accuracy as coefficients of regression (R2) values equal to 0.9591 and 0.9597, respectively. The average Absolute Relative Errors (% AARD) were obtained as 3.9547 for GA-RBF and 3.9791 for MPR models. The results show that proposed techniques are working efficiently in the estimation of HHV of different sources of biomass.

On the estimation of higher heating value of municipal wastes using soft computing approaches

ABSTRACT High values of higher heating value (HHV) for municipal solid wastes (MSWs) turn them into promising sources of energy nowadays. Accurate prediction of HHV values is investigated using intelligent algorithms of multilayer perceptron artificial neural network (MLP-ANN) and least squares support vector machine (LSSVM). The proposed methods will give the HHV values as a function of different chemical species’ mass percentages. These models compared with previously reported correlations. Better performance achieved for the proposed MLP-ANN and LSSVM regarding mean squared error (MSE) values of 0.2325 and 0.000186. Models considered as reliable predictive tools to estimate the HHV values of solid wastes.

Multiple regression and genetic programming for coal higher heating value estimation

ABSTRACTThe higher heating value (HHV) is an important characteristic for the determination of fuels quality. Nevertheless, its experimental measurement requires intricate technologies. In this work, the HHV of coal was predicted from ultimate composition using two methods: multiple regression and genetic programming. A dataset of 100 samples from literature was exploited (75% for training and 25% for testing). A comparative study was elaborated between the developed models and published ones in terms of correlation coefficient, root mean square error, and mean absolute percent error. The adopted models gave a good statistical performance.Abbreviations: C: Carbon; CC: Correlation coefficient; H: Hydrogen; HHV: Higher heating valueI; GT: Institute of gas technology; GP: Genetic programming; LHV: Lower heating value; MAPE: Mean absolute percent error; N: Nitrogen; O: Oxygen; RMSE: Root mean square error; S: sulfur; Wt: Weight percentage

Application of MLP-ANN as a novel predictive method for prediction of the higher heating value of biomass in terms of ultimate analysis

ABSTRACTIn the recent years, the energy issue is known as one of the main entries for economic and social development of human. So the biomass fuels as one of the approaches for supplying energy become the attractive topic for investigation. The higher heating value (HHV) is a key parameter for evaluation of energy of biomasses; so in the present study, a novel work was done to predict HHV as a function of ultimate analysis by utilization of multi-layer perceptron artificial neural network (MLP-ANN). To this end, a total number of 78 actual data were extracted from reliable references for training and validation of the model. The predicted HHVs were compared with the experimental data graphically and statistically, and the obtained results expressed that the MLP-ANN has a great potential for estimation of HHV of biomasses; so this approach can be used as a simple and accurate tool for forecasting HHV in terms of ultimate analysis. Based on the obtained results, this approach becomes one of the applicable softwares in industries.

Application of LSSVM algorithm for estimating higher heating value of biomass based on ultimate analysis

ABSTRACTThe higher heating value (HHV) is known as one of the energy evaluation parameters for biomass which has wide application in economic aspects investigation of energy sources. In this investigation the LSSVM algorithm as novel predicting model in the purpose of estimation of higher heating value in terms of ultimate analysis. A total number of 78 experimental data for training and testing of the algorithm were gathered from literature.in the purpose of evaluation of estimating algorithm the results are reported graphically and statistically. The calculated statistical indexes for overall data such as Root mean square error (RMSE), average absolute relative deviation (AARD) and the coefficient of determination (R2) are 9.2881, 0.038005 and 0.99996 respectively also the graphical results confirm the potential of LSSVM algorithm to be a predicting tool and a simple software for estimation of HHV as function of ultimate analysis.

Application of ANFIS-PSO algorithm as a novel method for estimation of higher heating value of biomass

ABSTRACTOne of the important parameters in economic study of energy sources and bioenergy is higher heating value (HHV). In this investigation, adaptive neuro fuzzy inference system (ANFIS) was applied as a novel method to predict HHV of biomass in terms of fixed carbon (FC), ash content (ASH), and volatile matters (VMs). Due to the fact that experimental investigations are time- and cost-consuming, this investigation was selected purely computational and a total number of 350 experimental data were extracted from literature for different steps of modeling. The proposed algorithm was evaluated by statistical indexes such as coefficient of determination (R2), root mean squared error (RMSE), and average absolute relative deviation (AARD), which are 0.90757, 1.1792, and 5.266, respectively. The reported indexes showed that ANFIS-particle swarm optimization can be used as a novel computational approach for prediction of HHV as function of proximate analysis.

Design of simplified models for the estimation of higher heating value of refused derived fuels

In this paper we disclose two new models for the estimation of the higher heating value of refused derived fuels as an answer to the question – “Are the available models to predict the HHV of biomasses suitable to be applied with an acceptable accuracy to RDFs?”. Based on the carbon and hydrogen content the newly developed equations were used to predict, accurately, the higher heating value of 25 samples available from open database sources and then compared with other 35 equations already introduced by other authors. The models herein presented have demonstrated, within the test group used to lead to very low average absolute and bias errors. The models were then validated with samples obtained locally, showing that the results were in line with that obtained for the test group.

Estimation of higher heating value of coal based on proximate analysis using support vector regression

To estimate the higher heating value (HHV) of coals based on proximate analysis, a nonlinear model termed support vector regression (SVR) is introduced in this work. A total of 167 Chinese coal samples and 4540 U.S. coal samples were employed to develop and verify the SVR-based correlations. The estimation results indicated that the average absolute errors from estimating the HHV of Chinese and U.S. coals were only 2.16% and 2.42%, respectively. Some published correlations were also employed and redeveloped with the Chinese and U.S. coals to obtain a comparison with the SVR-based correlations developed in the present work. The results indicate that the SVR-based correlations can be more accurate than the published correlations. Attempts were also made to develop a universal correlation for coals from different regions. The simulation results indicate that the correlation between the proximate analysis and HHV of coals from different geographical regions is varied. For coals from different regions, developing and using different correlations can obtain much higher accuracy in estimating the HHV from proximate analysis.