Biomass is a carbon-neutral renewable energy resource.Biochar produced from biomass pyrolysis exhibits preferable characteristics and potential for fossil fuel substitution.For time-and cost-saving,it is vital to esta...Biomass is a carbon-neutral renewable energy resource.Biochar produced from biomass pyrolysis exhibits preferable characteristics and potential for fossil fuel substitution.For time-and cost-saving,it is vital to establish predictive models to predict biochar properties.However,limited studies focused on the accurate prediction of HHV of biochar by using proximate and ultimate analysis results of various biochar.Therefore,the multi-linear regression(MLR)and the machine learning(ML)models were developed to predict the measured HHV of biochar from the experiment data of this study.In detail,52 types of biochars were produced by pyrolysis from rice straw,pig manure,soybean straw,wood sawdust,sewage sludge,Chlorella Vulgaris,and their mixtures at the temperature ranging from 300 to 800℃.The results showed that the co-pyrolysis of the mixed biomass provided an alternative method to increase the yield of biochar production.The contents of ash,fixed carbon(FC),and C increased as the incremental pyrolysis temperature for most biochars.The Pearson correlation(r)and relative importance analysis between HHV values and the indicators derived from the proximate and ultimate analysis were carried out,and the measured HHV was used to train and test the MLR and the ML models.Besides,ML algorithms,including gradient boosted regression,random forest,and support vector machine,were also employed to develop more widely applicable models for predicting HHV of biochar from an expanded dataset(total 149 data points,including 97 data collected from the published literature).Results showed HHV had strong correlations(|r|>0.9,p<0.05)with ash,FC,and C.The MLR correlations based on either proximate or ultimate analysis showed acceptable prediction performance with test R2>0.90.The ML models showed better performance with test R^(2)around 0.95(random forest)and 0.97–0.98 before and after adding extra data for model construction,respectively.Feature importance analysis of the ML models showed that ash and C were the most important inputs to predict biochar HHV.展开更多
In recent decades,the generation of Municipal Solid Waste(MSW)is steadily increasing due to urbanization and technological advancement.The col-lection and disposal of municipal solid waste cause considerable environme...In recent decades,the generation of Municipal Solid Waste(MSW)is steadily increasing due to urbanization and technological advancement.The col-lection and disposal of municipal solid waste cause considerable environmental degradation,making MSW management a global priority.Waste-to-energy(WTE)using thermochemical process has been identified as the key solution in this area.After evaluating many automated Higher Heating Value(HHV)predic-tion approaches,an Optimal Deep Learning-based HHV Prediction(ODL-HHVP)model for MSW management has been developed.The objective of the ODL-HHVP model is to forecast the HHV of municipal solid waste,based on its oxy-gen,water,hydrogen,carbon,nitrogen,sulphur and ash constituents.In addition,the ODL-HHVP model contains a Deep Support Vector Machine(DSVM)regres-sion component that can accurately predict the HHV.In addition,the Beetle Swarm Optimization(BSO)method is utilised as a hyperparameter optimizer in conjunction with the DSVM model,resulting in the highest HHV prediction accu-racy.A comprehensive simulation study is conducted to validate the performance of the ODL-HHVP method.The Multiple Linear Regression(MLR),Genetic Pro-gramming(GP),Resilient backpropagation(RP),Levenberg Marquardt(LM)and DSVM approaches have attained an ineffective result with RMSEs of 4.360,2.870,3.590,3.100 and 3.050,respectively.The experimentalfindings demon-strate that the ODL-HHVP technique outperforms existing state-of-art technolo-gies in a variety of respects.展开更多
The higher heating value(HHV)of biomass is a crucial property for design calculations and numerical simulations in bioenergy utilization.However,existing models for HHV prediction faced challenges in terms of predicti...The higher heating value(HHV)of biomass is a crucial property for design calculations and numerical simulations in bioenergy utilization.However,existing models for HHV prediction faced challenges in terms of predictive accuracy and generalization capability across various solid waste types,especially for those with high ash content.This work proposed a novel HHV prediction model based on its reduction degree(D_(R))and ash content(Cash).First,ultimate analysis of biomass was applied to establish the calculation method of D_(R);then,the correlation between D_(R),Cash,and HHV was analyzed using the Pearson Correlation Coefficient;subsequently,the HHV=f(D_(R),Cash)model was developed using regression analysis.Furthermore,the accuracy was compared to previous literature in terms of correlation coefficient(R^(2)),root mean square error(RMSE),and mean absolute error(MAE).Results revealed that this model provided attractive accuracy with R^(2)=0.854,RMSE=0.900,and MAE=0.773 within a wide range of ash content from 0 to 83.32 wt%.Even higher accuracy was achieved with this model in predicting the HHV of coal,biochar,and bio-oil,with R^(2)of 0.961,0.989,and 0.939,respectively.Conclusively,this work proposed the use of D_(R)for HHV estimation,which was not only a simple and accurate approach but also widely applicable to various fuels.展开更多
The higher heating value of five types of non- woody biomass and their torrefaction char was predicted and compared with the experimental data obtained in this paper. The correlation proposed in this paper and the one...The higher heating value of five types of non- woody biomass and their torrefaction char was predicted and compared with the experimental data obtained in this paper. The correlation proposed in this paper and the ones suggested by previous researches were used for prediction. For prediction using proximate analysis data, the mass fraction of fixed carbon and volatile matter had a strong effect on the higher heating value prediction oftorrefaction char of non-woody biomass. The high ash fraction found in torrefied char resulted in a decrease in prediction accuracy. However, the prediction could be improved by taking into account the effect of ash fraction. The correlation developed in this paper gave a better prediction than the ones suggested by previous researches, and had an absolute average error （AAE） of 2.74% and an absolute bias error （ABE） of 0.52%. For prediction using elemental analysis data, the mass fraction of carbon, hydrogen, and oxygen had a strong effect on the higher heating value, while no relationship between the higher heating value and mass fractions of nitrogen and sulfur was discovered. The best correlation gave an AAE of 2.28% and an ABE of 1.36%.展开更多
Higher heating value(HHV)is the key parameter for replacing Refuse-Derived Fuel(RDF)with fossil fuels in the cement industry.HHV can be measured with a bomb calorimeter or predicted from direct elemental data by using...Higher heating value(HHV)is the key parameter for replacing Refuse-Derived Fuel(RDF)with fossil fuels in the cement industry.HHV can be measured with a bomb calorimeter or predicted from direct elemental data by using regression models.Both methods require the continuous use of special laboratory equipment and are time consuming.To overcome these limitations,this study aims to predict the HHV value of RDF from predicted elemental data by using regression models.Therefore,once the predicted elemental data are generated,there will be no need to have continuous elemental data to predict HHV.Predicted elemental data were generated from direct elemental data and Near Infrared(NIR)camera-based spectrometric data by using a deep learning model.A convolutional neural networks(CNN)model was used for deep learning and was trained with 10,500 NIR image samples,each of which was 28×28×1.Different regression models(Linear,Tree,Support-Vector Machine,Ensemble and Gaussian process)were applied for HHV prediction.According to these results,higher R2 values(>0.85)were obtained with Gaussian process models(except for the Rational Quadratic model)for the predicted elemental data.Among the Gaussian models,the highest R2(0.95)but the lowest Root Mean Square Error(RMSE)(0.0563),Mean Squared Error(MSE)(0.0317)and Mean Absolute Error(MAE)(0.0431)were obtained with the Mattern 5/2 model.The results of predictions from predicted elemental data were compared to predictions from direct elemental data.The results show that the regression from predicted elemental data has an adequate prediction(R2=0.95)compared to the prediction from the direct elemental data(R^(2)=0.99).展开更多
The transportation industry still depends on fossil fuels,driving the need to transition to sustainable alternatives such as biofuels.Determining the viability of various biofuels as transportation fuels requires thor...The transportation industry still depends on fossil fuels,driving the need to transition to sustainable alternatives such as biofuels.Determining the viability of various biofuels as transportation fuels requires thorough knowledge of their fuel properties under different conditions.Conventional experimental methods of determining these properties are expensive and time-consuming,highlighting the need for reliable predictive techniques.This study is devoted to the precise prediction of the higher heating value of pure biodiesels through multiple linear and non-linear regressions,as well as machine learning techniques(artificial neural networks and support vector machines).The models are developed using density,kinematic viscosity,and cetane number data,and trained and validated over 241 biodiesel samples.The relative errors vary between 0.0039%-4.9195%,0.0041%-8.9644%,0.0062%-5.6850%,and 0.0018%-7.0050% for multiple non-linear,multiple linear,artificial neural networks,and support vector machines approaches,respectively.The average relative errors are computed as 0.9999%,1.2612%,0.9283%,and 0.9272% for multiple non-linear,multiple linear,artificial neural networks,and support vector machines approaches,respectively.The root mean square error values are computed as 0.5783 MJ/kg for multiple non-linear correlation,0.7212 MJ/kg for multiple linear correlation,0.5353 MJ/kg for artificial neural networks technique,and 0.6088 MJ/kg for support vector machines technique.The correlation coefficient value is computed as 0.9999 and 0.9998 for the multiple non-linear correlation and multiple linear correlation.The multiple non-linear regression model achieves the highest accuracy in predicting higher heating value of pure biodiesels.The study establishes also acceptable minimum ranges for higher heating value by solving a constrained optimization problem using the Newton method and Hessian matrix.The determined limited are approximately 35 MJ/kg and 33 MJ/kg,respectively,for EN 14214 and ASTM D6751 standards.Advanced machine learning methods will be evaluated to improve model accuracy in the future studies by incorporating a wider range of fuel properties.展开更多
基金The work was supported by the National Natural Science Foundation of China(No.51808278)the Science Foundation for Youths of Jiangxi Province,China(20192BAB213012)This research was also supported by the College Students’Innovative Entrepreneurial Training Plan Program,China(No.201910403049).
文摘Biomass is a carbon-neutral renewable energy resource.Biochar produced from biomass pyrolysis exhibits preferable characteristics and potential for fossil fuel substitution.For time-and cost-saving,it is vital to establish predictive models to predict biochar properties.However,limited studies focused on the accurate prediction of HHV of biochar by using proximate and ultimate analysis results of various biochar.Therefore,the multi-linear regression(MLR)and the machine learning(ML)models were developed to predict the measured HHV of biochar from the experiment data of this study.In detail,52 types of biochars were produced by pyrolysis from rice straw,pig manure,soybean straw,wood sawdust,sewage sludge,Chlorella Vulgaris,and their mixtures at the temperature ranging from 300 to 800℃.The results showed that the co-pyrolysis of the mixed biomass provided an alternative method to increase the yield of biochar production.The contents of ash,fixed carbon(FC),and C increased as the incremental pyrolysis temperature for most biochars.The Pearson correlation(r)and relative importance analysis between HHV values and the indicators derived from the proximate and ultimate analysis were carried out,and the measured HHV was used to train and test the MLR and the ML models.Besides,ML algorithms,including gradient boosted regression,random forest,and support vector machine,were also employed to develop more widely applicable models for predicting HHV of biochar from an expanded dataset(total 149 data points,including 97 data collected from the published literature).Results showed HHV had strong correlations(|r|>0.9,p<0.05)with ash,FC,and C.The MLR correlations based on either proximate or ultimate analysis showed acceptable prediction performance with test R2>0.90.The ML models showed better performance with test R^(2)around 0.95(random forest)and 0.97–0.98 before and after adding extra data for model construction,respectively.Feature importance analysis of the ML models showed that ash and C were the most important inputs to predict biochar HHV.
文摘In recent decades,the generation of Municipal Solid Waste(MSW)is steadily increasing due to urbanization and technological advancement.The col-lection and disposal of municipal solid waste cause considerable environmental degradation,making MSW management a global priority.Waste-to-energy(WTE)using thermochemical process has been identified as the key solution in this area.After evaluating many automated Higher Heating Value(HHV)predic-tion approaches,an Optimal Deep Learning-based HHV Prediction(ODL-HHVP)model for MSW management has been developed.The objective of the ODL-HHVP model is to forecast the HHV of municipal solid waste,based on its oxy-gen,water,hydrogen,carbon,nitrogen,sulphur and ash constituents.In addition,the ODL-HHVP model contains a Deep Support Vector Machine(DSVM)regres-sion component that can accurately predict the HHV.In addition,the Beetle Swarm Optimization(BSO)method is utilised as a hyperparameter optimizer in conjunction with the DSVM model,resulting in the highest HHV prediction accu-racy.A comprehensive simulation study is conducted to validate the performance of the ODL-HHVP method.The Multiple Linear Regression(MLR),Genetic Pro-gramming(GP),Resilient backpropagation(RP),Levenberg Marquardt(LM)and DSVM approaches have attained an ineffective result with RMSEs of 4.360,2.870,3.590,3.100 and 3.050,respectively.The experimentalfindings demon-strate that the ODL-HHVP technique outperforms existing state-of-art technolo-gies in a variety of respects.
基金the National Natural Science Foundation of China(22378183,22378184)Partial support was also provided through the Cultivation Program for the Excellent Doctoral Dissertation of Nanjing Tech University(3800124701)the Postgraduate Research&Practice Innovation Program of Jiangsu Province(KYCX24_1567).
文摘The higher heating value(HHV)of biomass is a crucial property for design calculations and numerical simulations in bioenergy utilization.However,existing models for HHV prediction faced challenges in terms of predictive accuracy and generalization capability across various solid waste types,especially for those with high ash content.This work proposed a novel HHV prediction model based on its reduction degree(D_(R))and ash content(Cash).First,ultimate analysis of biomass was applied to establish the calculation method of D_(R);then,the correlation between D_(R),Cash,and HHV was analyzed using the Pearson Correlation Coefficient;subsequently,the HHV=f(D_(R),Cash)model was developed using regression analysis.Furthermore,the accuracy was compared to previous literature in terms of correlation coefficient(R^(2)),root mean square error(RMSE),and mean absolute error(MAE).Results revealed that this model provided attractive accuracy with R^(2)=0.854,RMSE=0.900,and MAE=0.773 within a wide range of ash content from 0 to 83.32 wt%.Even higher accuracy was achieved with this model in predicting the HHV of coal,biochar,and bio-oil,with R^(2)of 0.961,0.989,and 0.939,respectively.Conclusively,this work proposed the use of D_(R)for HHV estimation,which was not only a simple and accurate approach but also widely applicable to various fuels.
文摘The higher heating value of five types of non- woody biomass and their torrefaction char was predicted and compared with the experimental data obtained in this paper. The correlation proposed in this paper and the ones suggested by previous researches were used for prediction. For prediction using proximate analysis data, the mass fraction of fixed carbon and volatile matter had a strong effect on the higher heating value prediction oftorrefaction char of non-woody biomass. The high ash fraction found in torrefied char resulted in a decrease in prediction accuracy. However, the prediction could be improved by taking into account the effect of ash fraction. The correlation developed in this paper gave a better prediction than the ones suggested by previous researches, and had an absolute average error （AAE） of 2.74% and an absolute bias error （ABE） of 0.52%. For prediction using elemental analysis data, the mass fraction of carbon, hydrogen, and oxygen had a strong effect on the higher heating value, while no relationship between the higher heating value and mass fractions of nitrogen and sulfur was discovered. The best correlation gave an AAE of 2.28% and an ABE of 1.36%.
基金supported by the Turkish Scientific and Technological Research Council(TUBITAK)(Project No.118Y135).
文摘Higher heating value(HHV)is the key parameter for replacing Refuse-Derived Fuel(RDF)with fossil fuels in the cement industry.HHV can be measured with a bomb calorimeter or predicted from direct elemental data by using regression models.Both methods require the continuous use of special laboratory equipment and are time consuming.To overcome these limitations,this study aims to predict the HHV value of RDF from predicted elemental data by using regression models.Therefore,once the predicted elemental data are generated,there will be no need to have continuous elemental data to predict HHV.Predicted elemental data were generated from direct elemental data and Near Infrared(NIR)camera-based spectrometric data by using a deep learning model.A convolutional neural networks(CNN)model was used for deep learning and was trained with 10,500 NIR image samples,each of which was 28×28×1.Different regression models(Linear,Tree,Support-Vector Machine,Ensemble and Gaussian process)were applied for HHV prediction.According to these results,higher R2 values(>0.85)were obtained with Gaussian process models(except for the Rational Quadratic model)for the predicted elemental data.Among the Gaussian models,the highest R2(0.95)but the lowest Root Mean Square Error(RMSE)(0.0563),Mean Squared Error(MSE)(0.0317)and Mean Absolute Error(MAE)(0.0431)were obtained with the Mattern 5/2 model.The results of predictions from predicted elemental data were compared to predictions from direct elemental data.The results show that the regression from predicted elemental data has an adequate prediction(R2=0.95)compared to the prediction from the direct elemental data(R^(2)=0.99).
文摘The transportation industry still depends on fossil fuels,driving the need to transition to sustainable alternatives such as biofuels.Determining the viability of various biofuels as transportation fuels requires thorough knowledge of their fuel properties under different conditions.Conventional experimental methods of determining these properties are expensive and time-consuming,highlighting the need for reliable predictive techniques.This study is devoted to the precise prediction of the higher heating value of pure biodiesels through multiple linear and non-linear regressions,as well as machine learning techniques(artificial neural networks and support vector machines).The models are developed using density,kinematic viscosity,and cetane number data,and trained and validated over 241 biodiesel samples.The relative errors vary between 0.0039%-4.9195%,0.0041%-8.9644%,0.0062%-5.6850%,and 0.0018%-7.0050% for multiple non-linear,multiple linear,artificial neural networks,and support vector machines approaches,respectively.The average relative errors are computed as 0.9999%,1.2612%,0.9283%,and 0.9272% for multiple non-linear,multiple linear,artificial neural networks,and support vector machines approaches,respectively.The root mean square error values are computed as 0.5783 MJ/kg for multiple non-linear correlation,0.7212 MJ/kg for multiple linear correlation,0.5353 MJ/kg for artificial neural networks technique,and 0.6088 MJ/kg for support vector machines technique.The correlation coefficient value is computed as 0.9999 and 0.9998 for the multiple non-linear correlation and multiple linear correlation.The multiple non-linear regression model achieves the highest accuracy in predicting higher heating value of pure biodiesels.The study establishes also acceptable minimum ranges for higher heating value by solving a constrained optimization problem using the Newton method and Hessian matrix.The determined limited are approximately 35 MJ/kg and 33 MJ/kg,respectively,for EN 14214 and ASTM D6751 standards.Advanced machine learning methods will be evaluated to improve model accuracy in the future studies by incorporating a wider range of fuel properties.
