Average Absolute Relative Error Research Articles

Experimental investigation of the convective heat transfer coefficients at different angular orientations inside the void in a cold flow pebble bed reactor using a non-invasive heat transfer pebble probe

This study focuses on investigating the effect of the surface location of the pebble inside the bed, by varying the angular orientation of the heat transfer pebble probe and the velocity of the gas on the convective heat transfer coefficients in a cold-flow pebble bed reactor. An advanced non-invasive technique consisting of a cartridge heated copper pebble, a micro-foil sensor and a thermocouple probe was used to obtain accurate measurements of the local heat transfer coefficients from three surface locations inside the void. The effect of the wall on the local convective heat transfer coefficients was highlighted due to the low aspect ratio of the bed (bed diameter to pebble diameter (5cm)) used in this work. The local measurements showed, for the first time ever, that the surface location inside the void substantially impacts the heat transfer coefficients, with higher heat transfer coefficients at the center of the void and bottom of the void. Furthermore, the measured overall heat transfer coefficients showed good agreements with the predictions of the correlations of [52,21] and [14]. A second order polynomial model with an R2=0.9807 and an average absolute relative error (AARE) equal to 9.08%, was developed for the prediction of the local heat transfer coefficient within the design and operation conditions of this work. The accurate local heat transfer data collected in this work is valuable as benchmark data for the validation of computational fluid dynamics (CFD) simulations and heat transfer calculations.

Microstructure Evolution and Constitutive Modelling of Deformation Behavior for Al-Mg-Si-Cu-Sc-Zr Alloy Processed with Isothermal Multidirectional Forging

This research is devoted to the microstructure evolution and deformation behavior of the Al-1.2Mg-0.7Si-1.0Cu-0.1Sc-0.2Zr alloy during the isothermal multidirectional forging (MDF) in a large cumulative strain and temperature range. The structure investigation of the studied alloy revealed several phases precipitated during solidification, among which θ(Al2Cu), Q(Al5Cu2Mg8Si6), Mg2Si, Sc-bearing W(AlScCu) and V(AlSi2Sc2) phases were observed. The MDF at 150–350 °C and a maximum cumulative strain of 14.4 significantly refined grain structure providing a mean grain size of 1.2–2.1 µm. The L12 structured Al3(Sc,Zr) dispersoids with a mean size of 10 ± 1 nm were formed during two-step homogenization annealing. Due to Zener pinning of the nanoscale dispersoids and fine-grained structure, the alloy exhibited near-superplastic behavior in a temperature range of 460–500 °C and strain rate range of 2 × 10−3–1 × 10−2 s−1 with the maximum elongation to failure of ~300%. After a strengthening heat treatment, the forged alloy exhibited the yield strength of 326 ± 5 MPa, ultimate tensile strength of 366 ± 5 MPa, and elongation of 10 ± 3%. The hot deformation behavior was described using the Arrhenius type model. The developed model demonstrated high predictability accuracy with a maximum average absolute relative error of 6.6%.

Comparative study of the efficiency, effectiveness and robustness of the TGE S.A. Day-Ahead Market model and system

The article contains selected results of comparative research on the quality of the parametric model, corrected in selected situations with the use of ANN and the Day-Ahead Market system of TGE S.A. carried out in MATLAB and Simulink. The System Identification Toolbox library was used for identification tests and Simulink for simulation and comparative tests. The comparative studies used suchmeasures of model and system quality as: efficiency, effectiveness and robustness. Their waveforms as well as their average values and absolute errors and relative errors between the identification model or the identification-neural model and the system were obtained. The results of general tests were shown for the hours: 6:00, 12:00, 18:00 and 24:00 in 2019, and the detailed tests for 6:00. The sensitivity of the waveforms obtained in terms of model quality and the Day-Ahead Market system was also tested, depending on the assumed values of such parameters as e.g. electricity volume or volume-weighted average price of electricity.

Estimation of Frost Layer Characteristics During Natural Convection on Vertical and Inverted Cold Surfaces Based on Smart Approaches

This study aimed to make a progress in the modeling of frost layer characteristics during natural convection on vertical and inverted cold surfaces, by directing focus towards the physical steps of frosting process. The modeling phase was relying on four enormous experimental datasets, embracing, (a) 727 data for frost thickness on vertical plates, (b) 735 data for frost thickness on inverted plates, (c) 218 data for frost surface roughness on vertical plates, and (d) 210 data for frost surface roughness on inverted plates. Three heuristic soft-computing paradigms, including multilayer Perceptron (MLP), radial basis function (RBF) and gaussian (GPR) were utilized to establish dimensionless predictive tools for each of the aforementioned four cases. It was found that the RBF paradigm presents the best performances in predicting the testing datasets pertinent to modeling cases (a)-(d), with average absolute relative errors (AAREs) of 1.28%, 2.30%, 4.31% and 3.57%, and R2 values of 99.97%, 99.83%, 98.53%, 98.74%, respectively. The recently developed models accurately estimated the values of frost thickness and surface roughness predictions at the critical moments of phase change, dendrite touch and reverse melting. Furthermore, they had excellent performances on the suit of describing the variations of frost characteristics with operational factors at all physical steps of frosting. The results implied the point that the trends of frost growth on vertical and inverted plates are quite different, indicating the significance of plate orientation in determining the characteristics of frost layer. Ultimately, a statistically-based investigation on the literature models revealed that all of them are fraught with deviations in describing the surface roughness and thickness of frost layer accumulated on vertical and inverted plates.

Toward empirical correlations for estimating the specific heat capacity of nanofluids utilizing GRG, GP, GEP, and GMDH

When nanoparticles are dispersed and stabilized in a base-fluid, the resulting nanofluid undergoes considerable changes in its thermophysical properties, which can have a substantial influence on the performance of nanofluid-flow systems. With such necessity and importance, developing a set of mathematical correlations to identify these properties in various conditions can greatly eliminate costly and time-consuming experimental tests. Hence, the current study aims to develop innovative correlations for estimating the specific heat capacity of mono-nanofluids. The accurate estimation of this crucial property can result in the development of more efficient and effective thermal systems, such as heat exchangers, solar collectors, microchannel cooling systems, etc. In this regard, four powerful soft-computing techniques were considered, including Generalized Reduced Gradient (GRG), Genetic Programming (GP), Gene Expression Programming (GEP), and Group Method of Data Handling (GMDH). These techniques were implemented on 2084 experimental data-points, corresponding to ten different kinds of nanoparticles and six different kinds of base-fluids, collected from previous research sources. Eventually, four distinct correlations with high accuracy were provided, and their outputs were compared to three correlations that had previously been published by other researchers. These novel correlations are applicable to various oxide-based mono-nanofluids for a broad range of independent variable values. The superiority of newly developed correlations was proven through various statistical and graphical error analyses. The GMDH-based correlation revealed the best performance with an Average Absolute Percent Relative Error (AAPRE) of 2.4163% and a Coefficient of Determination (R2) of 0.9743. At last, a leverage statistical approach was employed to identify the GMDH technique’s application domain and outlier data, and also, a sensitivity analysis was carried out to clarify the degree of dependence between input and output variables.

Predicting pKa of the carboxylic acid group in water solutions of amino acids based on molecular structures using machine learning QSPR methods

Amino acids as biomolecules are essential building blocks of proteins, which are fundamental to life and living organisms. The pKa values of amino acid residues in proteins can impact their structure and function. Furthermore, the pKa can be used to optimize experimental conditions for protein purification. In this research, a data set containing the pKa of the carboxylic acid group for 52 amino acids from various families in a water solution was used. The quantity structure property relationship (QSPR) approach was applied in our work for predicting the pKa using different machine learning methods. Initially, the amine molecules were drawn, and optimized and 3224 molecular descriptors were calculated for each molecule. Selecting among molecular descriptors using genetic algorithm-multi linear regression (GA-MLR) shows that the model with 8 descriptors is the best economic one. The comparison of results shows that the feed forward neural network (FFNN) is better than the particle swarm optimization-support vector machine (PSO-SVM) model. The coefficients of determination at different models follow the order of R2 FFNN= 0.9987 > R2 PSO-SVM= 0.9971 > R2 GA-MLR= 0.9951. The R2test for FFNN and PSO-SVM are near the R2train and this evidence proves the validation of models. Consequently, by using FFNN, the pKa of the carboxylic acid group in water solutions of amino acids can be predicted without any experimental costs with a low average absolute relative error (AARE%) of 2.82.

Studies on flow stress behaviour prediction of AZ31B alloy: Microstructural evolution and fracture mechanism

The hot tensile flow behaviour of the AZ31B alloy is investigated at various varying deformation temperatures (200–350 °C) and strain rates (0.1s−1, 0.01s−1, and 0.001s−1). The deformation condition significantly influenced the mechanical properties and microstructural evolution. Observation from the tensile tests indicated a strong dependence of flow stress on deformation temperature and strain rates. At a constant strain rate, flow stress decreased as the temperature increased, while at a constant deformation temperature, flow stress decreased with decreasing strain rates. The strain rate sensitivity varies from 0.01 to 0.25, suggesting a diffusion-controlled dislocation climb mechanism. Dynamic recrystallization (DRX) initiation was observed at 250 °C and a strain rate of 0.001s−1, characterized by the formation of necklace-type grains with low pole intensity. Predominantly, the DRX softening mechanism, including continuous dynamic recrystallization (CDRX) and discontinuous dynamic recrystallization (DDRX), was observed at 350 °C and 300 °C for a strain rate of 0.001s−1. Fracture morphology analysis of the tested samples indicated a micro-void coalescence mechanism. Equiaxed dimples were found at 350 °C and a strain rate of 0.001s−1, while oval-shaped dimples were observed at 300 °C and 0.1s−1. A strain-compensated Arrhenius model was incorporated to estimate the flow stress prediction for hardening and softening regions. Statistical parameters such as the average absolute relative error (AARE = 13.50) and coefficient of determination (R = 0.97) were calculated. Good agreement between experimental and prediction stresses was achieved at a 0.001s−1 strain rate for all deformation temperatures.

Theoretical analysis of various permeation models for gas transport through a covalent organic framework (COF) mixed matrix membrane

AbstractModeling and theoretical description of gas transport through mixed matrix membranes synthesized by including a covalent organic framework are essential for the efficient design of these membranes. This paper reports a theoretical analysis of various permeation models for gas transport through covalent organic framework‐based mixed matrix membranes. The existing models, such as Bruggeman, Maxwell, and Lewis Nielson, were analyzed. The Maxwell model calculated the permeation through voids and filler by considering interfacial defects in the covalent organic framework mixed matrix membrane's structure. Hence, dual Maxwell model was used to estimate gas transport through mixed matrix membranes. The percentage absolute average relative error of the dual Maxwell model was found to be less than 4 %. The results suggest that the dual Maxwell model effectively describes gas transport through a covalent organic framework mixed matrix membrane.

Investigating the effects of deformation-induced densification on the constitutive behavior and corrosion resistance of Al–Zn–Mg powder metallurgy alloy during hot deformation

The electrochemical behavior is most strongly influenced by deformation, which may either accelerate or decelerate failure during service. Therefore, in the present study, the Al–5.6Zn–2Mg alloy fabricated by powder metallurgy was subjected to hot deformation at various temperatures (300–500 °C) and strain rates (0.5–0.005 s−1) on a hydraulic press, and then, the effect of deformation-induced densification on corrosion behavior was studied using electrochemical corrosion tests. Flow stress is mathematically modeled using the strain-compensated Arrhenius model. The model developed was reliable to predict, with an average absolute relative error (AARE) of 3.16% and an R2 value of 0.98. The potentiodynamic polarization results showed that as the degree of porosity in a sintered sample is reduced, there was a decrease in the corrosion current density and an increase in the pitting potential. Al–5.6Zn–2Mg corrosion resistance significantly improved with deformation temperature, whereas strain rates decrease it, due to a decrease in pore content. Icorr of 2.4982 µA/cm2 and Epit of 0.876 V have been found for the highly deformed preforms under 500 °C and 0.005 s−1 strain rate condition. EIS results showed a high contact resistance of 2659 Ohm·cm2 and high hardness value of 113.12 Hv for the preform deformed under 500 °C and 0.005 s−1 strain rate. The reduction in porosity, decrease in active surface area after the deformation process, densification behavior, and hardness are all related to the electrochemical behavior. Pitting corrosion was observed in deformed preforms as a corrosion mechanism; the dynamic recovery feature is observed from the flow behavior.

A novel Hybrid XGBoost Methodology in Predicting Penetration Rate of Rotary Based on Rock-Mass and Material Properties

Predicting the drill penetration rate is a fundamental requirement in mining operations, profoundly impacting both the cost-effectiveness of mining activities and strategic mine planning. Given the intricate web of factors influencing rotary drilling performance, the necessity for advanced modeling techniques becomes evident. To this end, the hybrid extreme gradient boosting (XGBoost) was utilized to gauge the penetration rate of rotary drilling machines, utilizing random search, grid search, Harris Hawk optimization (HHO), and the dragonfly algorithm (DA) as metaheuristic algorithms. Our research draws from extensive data collected in copper mine case studies, encompassing both field and investigational data. This dataset incorporates critical material properties, such as tensile strength (TS), uniaxial compressive strength (UCS), as well as vital rock-mass characteristics including joint direction (JD), joint spacing (JS), and bit diameter (D). Our investigation evaluates the reliability of these prediction methods using various performance indicators, including mean absolute error (MAE), root mean square error (RMSE), average absolute relative error (AARE), and coefficient of determination (R2). The multivariate analysis reveals that the HHO-XGB model stands out, demonstrating superior prediction accuracy (MAE: 0.457; RMSE: 2.19; AARE: 2.29; R2: 0.993) compared to alternative models. Furthermore, our sensitivity analysis emphasizes the substantial impact of uniaxial compressive strength and tensile strength on the drill penetration rate. This underlines the importance of considering these material properties in mining operations. In conclusion, our research offers robust models for forecasting the penetration rate of similar rock formations, providing invaluable insights that can significantly enhance mining operations and planning processes.

Kinetic, artificial neural network, and statistical modeling to optimize the parameters of the air flotation process to remove latex suspended solids in ABS wastewater

As China's contemporary industrial production undergoes continuous enhancement and economic development, there is a remarkable surge in the consumption of acrylonitrile–butadiene–styrene (ABS) materials. The extensive application of dissolved air flotation (DAF) has been adopted to effectively eliminate a substantial quantity of suspended solids present in ABS wastewater. Attaining precise control over the DAF process stands as the pivotal factor in enhancing the removal efficiency. In this study, we had constructed an air flotation apparatus and developed an open-structure software within the master computer. We innovatively compared the response surface methodology (RSM), artificial neural network (ANN), and adsorption kinetic model (AKM) with ABS wastewater DAF treatment process by incorporating relevant parameters. The results demonstrated that the ANN exhibited superior fitting capability in comparison to both the RSM and AKM. Drawing from the comparative results, it was evident that, in the context of COD removal, the ANN surpassed the RSM by margins of 5.06 %, 80 %, and 4.44 % in terms of R-squared (R2), Average Absolute Relative Error (AARE), and Root Mean Square Error (RMSE), respectively. The ANN outperformed the AKM by percentages of 35.71 %, 95.65 %, and 74.66 % concerning R2, AARE, and RMSE, respectively. Similar conclusions were drawn for suspended solids (SS) removal. This study provided valuable support for the adaptive adjustment of successive reaction conditions and the prediction of outlet quality, thus possessing commendable applicative significance and practical utility.

Machine Learning-Based Models for Basic Sediment & Water and Sand-Cut Prediction in Matured Niger Delta Fields

Oil production from matured fields in the Niger Delta is characterised by basic sediment and water (BS&W) and sand or sand-cut (Scut) production. The predominant factor for this production is the unconsolidated nature of the formations in the Niger Delta. The available correlations for estimating BS&W and Scut are based more on the intrinsic reservoir properties than controllable wellhead variables during oil production. This study developed neural-based models to predict BS&W and Scut based on multiple-inputs single-output (MISO) and multiple-inputs multiple-outputs (MIMO) networks using 457 datasets from 43 oilfields in the Niger Delta. The performances of the neural-based models with new fields test datasets were determined using some statistical yardsticks: coefficient of determination (R2), correlation coefficient (R), mean square error (MSE), root mean square error (RMSE), average relative error (ARE), and average absolute relative error (AARE). The results indicate that the MISO neural-based models had overall R and MSE values of 0.9999 and 2.0698\(\times\)10-5, respectively, for BS&W and 0.9995 and 2.1529\(\times\)10-6 for Scut. In contrast, the MIMO neural-based model had overall R and MSE values of 0.9997 and 7.5865\(\times\)10-5. The generalisation performance of the MISO neural-based models with new field test datasets resulted in R2, R, MSE, RMSE, ARE and AAPRE of 0.97406, 0.98695, 2.08143, 1.44272, -0.00638 and 0.28755, respectively, for the BS&W model and R2 of 0.89558, R of 0.93544, MSE of 0.01736, RMSE of 0.13177, ARE of 0.01338 and AARE of 0.01759 for the Scut model. Furthermore, the MIMO-based model with new field test datasets resulted in R2, R, MSE, RMSE, ARE and AAPRE of 0.97317, 0.98650, 2.15293, 1.46729, -0.00713 and 0.25064, respectively, for BS&W, while the Scut model had R2 of 0.87505, R of 0.93544, MSE of 0.02118, RMSE of 0.14554, ARE of -0.02280 and AARE of 0.02996. Also, the relative importance of the input parameters of the MISO and MIMO neural-based models in predicting BS&W and Scut is \(q_0\) >Pr>Pwh>S> \(\gamma\)API . Based on the statistical indicators obtained, the predictions of the developed neural models were close to the actual fields’ datasets. Thus, the neural-based models should apply as tools for estimating BS&W and Scut in mature fields in the Niger Delta.

Arrhenius constitutive equation and artificial neural network model of flow stress in hot deformation of offshore steel with high strength and toughness

ABSTRACT In this study, the thermal deformation behaviour of a high strength offshore steel at different temperatures and rates was investigated through thermal compression experiments.An Arrhenius constitutive model and a back propagation artificial neural network (BP-ANN) were established to address more complex deformation characteristics. The performances of both models were was evaluated using standard statistical parameters such as the correlation coefficient (R) and average absolute relative error (AARE). The results showed that both models can accurately predict the rheological stresses generated during deformation.The BP-ANN outperforms the Arrhenius equation model with correlation coefficients of fit greater than 99.9% and less than 0.8% relative error. At a strain rate of 0.01 s−1 and 10 s−1, the accuracy of the ANN decreases slightly due to the fact that it exceeds the strain rate range of the training set, as compared to the Arrhenius constitutive equations as these are more accurately predicted.

Constitutive Modeling of Hot Deformation Behaviors and Processing Maps of Mg–4.83Gd–2.36Nd–0.21Zr Alloy

Accurately assessing the hot workability and flow behavior is crucial in the preparation of high‐performance wrought Mg–rare earth (RE) alloys. The hot deformation behaviors and workability of Mg–4.83Gd–2.36Nd–0.21Zr alloy are examined via Gleeble 3500 thermosimulation tests across a range of deformation temperatures from 350 °C to 450 °C and strain rates ranging from 0.001 to 1 s−1. The experimental results reveal that the flow stress of the experimental alloy decreases as the temperature increases and the strain rate decreases. The activation energies for deformation (Q) of the experiment alloy are calculated by the hyperbolic constitutive equation and range from 186.78 to 234.97 kJ mol−1. Strain compensation is incorporated in constitutive modeling, resulting in the correlation coefficient (R) of 0.9865 and the average absolute relative error (AARE) of 5.4639%. Further, the processing maps are constructed at different strains based on dynamic material models, from which the feasible processing window of the experimental alloy is determined in the areas of deformation temperatures of 425–450 °C and strain rates of 0.01–0.1 s−1.

Atomic connectivity group contribution method for predicting the boiling and melting points of organic compounds

Boiling point (Tb) and melting point (Tm) are the fundamental physical properties in the petroleum-related and chemical industries. In this work, the Tb and Tm models are established by the atomic connectivity group contribution (ACGC) method. In the proposed methodology, shape factors (SFs) and atomic connectivity factors (ACFs) are employed to distinguish among isomers and improve the prediction accuracy. In addition, external and internal validations and Y-randomization tests are performed to analyze and evaluate the predictability and robustness of the ACGC models. The results indicate that the ACGC method could predict the Tb with high accuracy, with a squared correlation coefficient (R2) of 0.9767 and an absolute average relative error (AARE) of 1.70%. Moreover, the ACGC model for Tm achieves an R2 of 0.8423 and an AARE of 8.73%, demonstrating that the ACGC method can effectively predict the Tm.

Hot workability behaviour of two P92 creep resistant steels: Constitutive analysis

This article reports the flow stress behaviour of two P92 steels at a temperature range of 850–1000°C and a strain rate of 0.1–10 s−1 using the Gleeble® 3500 thermomechanical simulator. A physically-based constitutive model was used to analyse the effects of deformation conditions on the flow stress behaviour during deformation. This model incorporates the influence in the variation of Young’s modulus and the self-diffusion coefficient as affected by temperature. The study developed constitutive equations that predict the flow stress behaviour of the two steels investigated. From the constitutive analysis of the results, the stress exponent n was: 9.8 (steel A) and 10.3 (steel B). The model used the self-diffusion activation energy of steel. The statistical parameters: correlation coefficient of 0.99 (for steel A and B), the absolute average relative error of 2.18% (steel A) and 2.20% (steel B) quantified the applicability of the model. The quantification results show that the constitutive equations developed have high accuracy in predicting the workability of the two P92 steels. The study has shown that this method is applicable in predicting the metal flow pattern of two P92 steels in the metalworking processes.

Temperature rises and constitutive equation of homogenized 6063 aluminum alloy for extrusion

The effect of temperature rises on the flow stress and constitutive equation of homogenized 6063 aluminum alloy was studied by theoretical analysis, numerical simulation, and experiments. The results showed that the temperature rises increased with the decreases of deformation temperatures and the increases of strain rates, which caused the flow softening. The modification of the flow stresses of 6063 aluminum alloy was carried out. A strain compensation constitutive equation based on the Arrhenius equation and Zener-Hollomon parameter was introduced to predict the flow behavior of 6063 aluminum alloy. The calculated flow stresses were consistent with the experimental results, and its average absolute relative error was only 3.25%. Finally, the established constitutive equation was substituted into the Deform-3D software. The corresponding extrusion experiments were carried out. The maximum extrusion pressures and maximum exit temperatures in the numerical simulation were in good agreement with those in the experiments, which confirmed the accuracy and reliability of the established constitutive equation.

Applying a deep-learning surrogate model to simulate and compare achievable oil recovery by different waterflood scenarios

Reservoir pressure depletion is a critical aspect leading to declined oil production. Water flooding becomes mandatory to increase the recovery efficiency from certain reservoirs. Establishing well placements, injection rates, and injection schedules enables improved fluid displacement in such reservoirs. Standard reservoir simulators are widely used to provide near-optimal scenarios for maximizing oil recovery. Such reservoir simulations involved numerous iterations and required significant computational effort. Also, the conventional machine learning methods provide a relatively high uncertainty to predict the considered problem. In this context, a deep-learning surrogate model comprising long short-term memory networks and densely connected networks is proposed to predict cumulative oil production using various waterflood injection well patterns, injection rates, and injection period information. The relevance test clarifies that the oil recovery mainly controls by water injection duration and injection pattern. A total of 1485 data records are considered to validate the model’s performance, demonstrating less than 3.5% absolute average relative error and a prediction versus actual value regression coefficient of 0.99989. In addition, the prediction accuracy of the proposed deep-learning model is also shown to outperform two well-known traditional machine-learning models (i.e. artificial neural networks and support vector regressions).

Analysis of the High Temperature Plastic Deformation Characteristics of 18CrNi4A Steel and Establishment of a Modified Johnson–Cook Constitutive Model

To investigate the plastic deformation behavior of 18CrNi4A steel at high temperatures, an isothermal hot compression test was conducted on a Thermecmastor-Z series test machine at deformation temperatures from 1273 K to 1423 K and strain rates ranging from 0.01 s−1 to 10 s−1. The effects of these two factors on the flow stress were analyzed. Based on the true stress–strain experimental curves, the original Johnson–Cook constitutive model was applied to determine the flow stress data under different deformation conditions. The prediction results of the model were compared with the experimental data. The main reason for the large deviation observed between them was that the coupling relationship between the deformation temperature and the strain rate was not considered, so the original Johnson–Cook model was modified. The correlation coefficient and average absolute relative error of the original Johnson–Cook model were 0.962 and 16.36%. The prediction accuracy of the modified Johnson–Cook model was improved to 0.991 and 5.58%, respectively. The results show that the modified Johnson–Cook model exhibits higher prediction precision, which is beneficial for the broader application of 18CrNi4A steel in the industry.

Applying feature selection and machine learning techniques to estimate the biomass higher heating value

The biomass higher heating value (HHV) is an important thermal property that determines the amount of recoverable energy from agriculture byproducts. Precise laboratory measurement or accurate prediction of the HHV is essential for designing biomass conversion equipment. The current study combines feature selection scenarios and machine learning tools to establish a general model for estimating biomass HHV. Multiple linear regression and Pearson’s correlation coefficients justified that volatile matter, nitrogen, and oxygen content of biomass samples have a slight effect on the HHV and it is better to ignore them during the HHV modeling. Then, the prediction performance of random forest, multilayer and cascade feedforward neural networks, group method of data handling, and least-squares support vector regressor are compared to determine the intelligent estimator with the highest accuracy toward biomass HHV prediction. The ranking test shows that the multilayer perceptron neural network better predicts the HHV of 532 biomass samples than the other intelligent models. This model presents the outstanding absolute average relative error of 2.75% and 3.12% and regression coefficients of 0.9500 and 0.9418 in the learning and testing stages. The model performance is also superior to a recurrent neural network which was recently developed in the literature using the same databank.