Mean Absolute Relative Deviation Research Articles

McGFN-FF: an accurate force field for optimization and energetic screening of molecular crystals

Abstract This work presents a periodic extension of the GFN-FF force field for molecular crystals named mcGFN-FF. Non-covalent interactions in the force field are adjusted to reduce the systematic overbinding of the original, molecular version for molecular crystals. A diverse set of molecular crystal benchmarks for lattice energies and unit cell volumes is studied. The modified force field shows good results with a mean absolute relative deviation (MARD) of 19.9 % for lattice energies and 10.0 % for unit cell volumes. In many cases, mcGFN-FF approaches the accuracy of the GFN1-xTB quantum chemistry method which has an MARD of 18.7 % for lattice energies and 6.2 % for unit cell volumes. Further, the newly compiled mcVOL22 benchmark set is presented which features r2SCAN-D4/900 eV DFT reference volumes for molecular crystals with phosphorus-, sulfur-, and chlorine-containing compounds of various sizes. Overall, the mcGFN-FF poses an efficient tool for the optimization and energetic screening of molecular crystals containing elements up to radon.

Fundamental research on the condensation heat transfer of the hydrocarbon-mixture energy in a spiral tube described by a universal model using flow pattern based and general modes

The transportation of liquefied natural gas (LNG) is a crucial aspect of global energy application and security. However, there are still some challenges in liquefaction. This study establishes systematic experiments to investigate the condensation two-phase flow and heat transfer process of hydrocarbon-mixture energy (methane/ethane/propane/isobutane) in a spiral tube. The main influencing parameters, including mass flux (200–560 kg/(m2·s)), operating pressure (2–4 MPa), and vapor quality (0–1) on the heat transfer intensity and frictional pressure drop were analyzed. Different two-phase flow patterns were observed and categorized them as bubble flow, intermittent flow, stratified-wavy flow, and annular flow. For bubble flow and intermittent flow, vapor quality is found to remarkably affect the heat transfer; regarding stratified-wavy flow and annular flow, gravitational force and inertia force are found to prominently influence the heat transfer, respectively. Additionally, a newly universal model includes split flow patterns mode (SFPM) and the general mode (GM) has been proposed, which is with high-precision for describing the condensation heat transfer coefficients of different flow patterns. Furthermore, the analytical results show the SFPM's root mean square error and mean absolute relative deviation are 97.35 W/(m2·K) and 0.036.

A new heat transfer correlation based on flow patterns for hydrocarbon refrigerants condensation flow

To accurately predict heat transfer coefficients for hydrocarbon refrigerants condensation flow in heat exchangers is imperative for designing efficient systems. While various studies have proposed numerous correlations for predicting heat transfer coefficients, some of these correlations are based on fitted experimental data and are limited to predicting specific conditions. The prediction accuracy and capability for different working fluids remain to be evaluated. Currently, most researches mainly focus on pure working fluids, with relatively little attention given to mixed working fluids. In this study, six commonly used heat transfer correlations are evaluated and a new heat transfer correlation based on flow patterns is proposed. The research results indicate that the existing correlations have poor universality for predicting hydrocarbon refrigerants, and the prediction accuracy needs to be improved. The correlation developed in this study demonstrates good prediction performance for condensation heat transfer coefficients of 413 sets of hydrocarbon refrigerants, including methane, propane, ethane/propane, and methane/propane. The overall mean relative deviation (MRD) is −1.77%, and mean absolute relative deviation (MARD) is 8.12%. The correlation developed in this study can provide theoretical guidance for predicting condensation heat transfer coefficient for pure and mixed hydrocarbon refrigerants, aiding in the design of efficient heat exchangers.

Heat transfer characteristics in vertical tube condensers: Experimental investigation and numerical validation

The paper establishes a vertical steam condensation heat transfer experimental device to investigate the process of liquid film condensation heat transfer under the countercurrent heat transfer of a secondary coolant. An improved experimental method, integrating measurements and simulations, accurately calculates local heat transfer coefficients and heat flux. The validity of the data is confirmed through comparison with established condensation correlations, affirming the method's effectiveness. The correlations of previous research (i.e., Numrich, Carey, Oh, Lee, and Shah) demonstrate greater consistency with the experimental results, with mean relative deviations (MRD) of −4.90%, 7.18%, 11.42%, −8.63% −4.45%, and mean absolute relative deviations (MARD) of 11.61%, 21.57%, 23.60%, 18.97%, and 12.53%, respectively. Numrich's correlation demonstrated the closest congruence. Based on these results, the research proposes a new coupled heat transfer model, effectively simulating the interaction between external cooling water and steam condensation. This model simplifies the process by pre-defining coolant temperature changes per unit length, thereby reducing the need for extensive iterative calculations and eliminating assumptions about wall temperature. It accurately predicts the wall temperature, local condensing temperature, and coolant temperature in a vertical condenser. These experimental methods and models offer valuable insights for optimizing vertical condenser design and analysis. The study underscores the significance of combining simulation models with experimental measurements to enhance the understanding of heat transfer in complex systems. The findings promise to advance research in film condensation heat transfer and assist in condenser design and optimization.

Evaluating the generality of machine learning-based universal models used for prediction of condensation heat transfer coefficient in mini/macro channels

This research concerns the development of universal models founded on machine learning to predict the condensation heat transfer coefficient. In this regard, a consolidated database has been employed encompassing 8340 data points which belong to the flow condensation inside mini/macro channels. The database consists of 25 working fluids with hydraulic diameters ranging from 0.42 to 20.8 mm, mass velocities ranging from 13.1 to 1400 Kg m−2 s− 1 and reduced pressures ranging from 0.03 to 0.9. Four machine learning regression models namely artificial neural network (ANN), gradient-boosted regression (GBR), random forest regression (RFR) and support vector regression (SVR) are trained utilizing the database, and subsequently their accuracy and generality are compared. A multitude of dimensionless parameters are regarded as features, while three parameters specifically the heat transfer coefficient, Nusselt number and Nusselt number correction factor are each considered individually as targets. Within each model, the optimal values of the important hyper-parameters are tuned through appropriate search methods. In order to evaluate the generality of the models, the idea is to sequentially exclude each individual database from the consolidated database. The comparison of accuracy and generality of the models reveals that the RFR model using Nusselt number correction factor as target with a mean absolute relative deviation (MARD) of 3.01 % exhibits the highest level of accuracy, whereas the RFR model employing Nusselt number as target with a MARD for the predicted excluded values equal to 19.49 % demonstrates the superior generality.

Continuous time-resolved estimated synthetic 4D-CTs for dose reconstruction of lung tumor treatments at a 0.35 T MR-linac

Objective. To experimentally validate a method to create continuous time-resolved estimated synthetic 4D-computed tomography datasets (tresCTs) based on orthogonal cine MRI data for lung cancer treatments at a magnetic resonance imaging (MRI) guided linear accelerator (MR-linac). Approach. A breathing porcine lung phantom was scanned at a CT scanner and 0.35 T MR-linac. Orthogonal cine MRI series (sagittal/coronal orientation) at 7.3 Hz, intersecting tumor-mimicking gelatin nodules, were deformably registered to mid-exhale 3D-CT and 3D-MRI datasets. The time-resolved deformation vector fields were extrapolated to 3D and applied to a reference synthetic 3D-CT image (sCTref), while accounting for breathing phase-dependent lung density variations, to create 82 s long tresCTs at 3.65 Hz. Ten tresCTs were created for ten tracked nodules with different motion patterns in two lungs. For each dataset, a treatment plan was created on the mid-exhale phase of a measured ground truth (GT) respiratory-correlated 4D-CT dataset with the tracked nodule as gross tumor volume (GTV). Each plan was recalculated on the GT 4D-CT, randomly sampled tresCT, and static sCTref images. Dose distributions for corresponding breathing phases were compared in gamma (2%/2 mm) and dose–volume histogram (DVH) parameter analyses. Main results. The mean gamma pass rate between all tresCT and GT 4D-CT dose distributions was 98.6%. The mean absolute relative deviations of the tresCT with respect to GT DVH parameters were 1.9%, 1.0%, and 1.4% for the GTV D 98%, D 50%, and D 2%, respectively, 1.0% for the remaining nodules D 50%, and 1.5% for the lung V 20Gy. The gamma pass rate for the tresCTs was significantly larger (p < 0.01), and the GTV D 50% deviations with respect to the GT were significantly smaller (p < 0.01) than for the sCTref. Significance. The results suggest that tresCTs could be valuable for time-resolved reconstruction and intrafractional accumulation of the dose to the GTV for lung cancer patients treated at MR-linacs in the future.

Heat transfer model for downward two-component two-phase flows in inclined pipes

Downward two-component two-phase flows are commonly encountered in many chemical engineering applications. Accurate prediction of the heat transfer coefficient is of great importance to the design and operation of heat transfer systems. The mechanism of two-phase heat transfer is complicated because of the complex interactions between gas and liquid phases. Therefore, this study was aimed at developing a robust and theoretically supported heat transfer coefficient correlation for downward two-component two-phase flow in inclined pipes with the extended Chilton-Colburn analogy. First, an extensive literature review was performed to collect more than 1700 experimental data for downward two-phase heat transfer coefficients in inclined pipes and 10 heat transfer coefficient correlations. Poor predictive performance of the existing correlations was identified using the whole collected database. Then, the dependence of the heat transfer coefficient on void fraction and two-phase multiplier was analyzed using the extended Chilton-Colburn analogy. The dependence of the heat transfer coefficient on the pipe inclination angle was discussed in depth. The exponents in the newly-developed heat transfer coefficient correlation were correlated in terms of the pipe inclination angle. A robust and semi-theoretically supported correlation was developed for predicting the heat transfer coefficient of downward two-component two-phase flows in inclined pipes. The mean relative deviation (bias), mrel, and the mean absolute relative deviation (random uncertainty), mrel, ab, were −4.54 % and 13.8 %, respectively. The new heat transfer coefficient correlation will be of practical importance in understanding the two-component two-phase heat transfer characteristics.

Continuous Glucose Monitoring (CGM) in Sports-A Comparison between a CGM Device and Lab-Based Glucose Analyser under Resting and Exercising Conditions in Athletes.

The objective of this pilot study was to compare glucose concentrations in capillary blood (CB) samples analysed in a laboratory by a validated method and glucose concentrations measured in the interstitial fluid (ISF) by continuous glucose monitoring (CGM) under different physical activity levels in a postprandial state in healthy athletes without diabetes. As a physiological shift occurs between glucose concentration from the CB into the ISF, the applicability of CGM in sports, especially during exercise, as well as the comparability of CB and ISF data necessitate an in-depth assessment. Ten subjects (26 ± 4 years, 67 ± 11 kg bodyweight (BW), 11 ± 3 h) were included in the study. Within 14 days, they underwent six tests consisting of (a) two tests resting fasted (HC_Rest/Fast and LC_Rest/Fast), (b) two tests resting with intake of 1 g glucose/kg BW (HC_Rest/Glc and LC_Rest/Glc), (c) running for 60 min at moderate (ModExerc/Glc), and (d) high intensity after intake of 1 g glucose/kg BW (IntExerc/Glc). Data were collected in the morning, following a standardised dinner before test day. Sensor-based glucose concentrations were compared to those determined from capillary blood samples collected at the time of sensor-based analyses and subjected to laboratory glucose measurements. Pearson's r correlation coefficient was highest for Rest/Glc (0.92, p < 0.001) compared to Rest/Fast (0.45, p < 0.001), ModExerc/Glc (0.60, p < 0.001) and IntExerc/Glc (0.70, p < 0.001). Mean absolute relative deviation (MARD) and standard deviation (SD) was smallest for resting fasted and similar between all other conditions (Rest/Fast: 8 ± 6%, Rest/Glc: 17 ± 12%, ModExerc/Glc: 22 ± 24%, IntExerc/Glc: 18 ± 17%). However, Bland-Altman plot analysis showed a higher range between lower and upper limits of agreement (95% confidence interval) of paired data under exercising compared to resting conditions. Under resting fasted conditions, both methods produce similar outcomes. Under resting postprandial and exercising conditions, respectively, there are differences between both methods. Based on the results of this study, the application of CGM in healthy athletes is not recommended without concomitant nutritional or medical advice.

Condensation heat transfer and correlation analysis in helical tube of mixed carbon-hydro refrigerant

The heat transfer coefficient of the mixed carbon-hydrogen refrigerant (hmix) in a helically tube has been confirmed as the core parameter related to natural gas liquefaction. Moreover, an experimental system of the two-phase flow in the helical pipe was designed, and a simulation model was built to analyze the changing law of hmix. The effects of operating parameters (operating pressure (2–4 MP), mass flux (200–600 kg/(m2·s), and vapor quality (0–1)) and structural parameters (spiral diameter (1–3 m), helix angle (4–14°), and inner diameter (8–12 mm)) on hmix of methane/ethane/propane in the spiral tube were examined. Subsequently, 11 heat transfer correlations were analyzed and then compared. As indicated by the comparison results, the correlations of Boyko were the most accurate among the above-mentioned correlations, laying a basis for this type of hmix. Furthermore, the mean absolute relative deviation and determination coefficient of Boyko's correlation were obtained as 20.30% and 0.48, respectively.

FRICTIONAL PRESSURE DROP DURING CONDENSING FLOW INSIDE MINI/MICRO-CHANNELS: PREDICTIVE TOOLS ASSESSMENT AND DATA TREATMENT

A total of 1210 new condensation friction pressure gradient points (database) is collected from 12 open articles on mini/micro-channels literature. The database covered a wide range of thermophysical properties associated with low-GWP refrigerants, physical dimension, mass velocity and vapor quality. A comparison of the experimental database against eleven two-phase flow frictional pressure drop prediction tools is carried out. Based on that, the mini/micro-channel developed by Xu and Fang exhibited the best ability to capture the tendency of the database with a mean absolute relative deviation of (36.59%) for all data. Keywords: Two-phase flow frictional pressure gradient; Mini/micro-channels; Condensing flow; New refrigerant fluid; Heat transfer; Collect database.

Lower limbs inter-joint coordination and variability during typical Tai Chi movement in older female adults.

Purpose: This study aimed to investigate the lower limb inter-joint coordination and variability during Tai Chi movements compared with normal walking in older adults. Methods: A total of 30 female Tai Chi practitioners (70.9 ± 5.2years) were recruited in this study. Herein, each participant performed three trials of the normal walking and Tai Chi movements. The lower limb kinematics data were collected with Vicon 3D motion capture system. The continuous relative phase (CRP) includes both spatial and temporal information of two adjacent joints, which was calculated to assess the inter-joint coordination of lower limbs. Coordination amplitude and coordination variability were assessed with mean absolute relative phase (MARP) and deviation phase (DP). MANOVOA was used to analyze inter-joint coordination parameters between different movements. Results: The CRP values of hip-knee and knee-ankle segments in the sagittal plane of the Tai Chi movements changed frequently. The MARP values of the hip-knee (p < 0.001) and knee-ankle segments (p = 0.032) as well as the DP values of the hip-knee segment (p < 0.001) were significantly lower in Tai Chi than in normal walking. Conclusion: More consistent and stable inter-joint coordination patterns of Tai Chi movements found in this study may be one of the critical factors that Tai Chi could be a suitable coordinated exercise for older adults.

Fusion of various optimisation based feature smoothing methods for wearable and non-invasive blood glucose estimation.

The traditional blood glucose estimation method requires to take the invasive measurements several times a day. Therefore, it has a high infection risk and the users are suffered from the pain. Moreover, the long term consumable cost is high. Recently, the wearable and non-invasive blood glucose estimation approach has been proposed. However, due to the unreliability of the acquisition device, the presence of the noise and the variations of the acquisition environments, the obtained features and the reference blood glucose values are highly unreliable. Moreover, different subjects have different responses of the infrared light to the blood glucose. To address this issue, a polynomial fitting approach to smooth the obtained features or the reference blood glucose values has been proposed. In particular, the design of the coefficients in the polynomial is formulated as the various optimisation problems. First, the blood glucose values are estimated based on the individual optimisation approaches. Second, the absolute difference values between the estimated blood glucose values and the actual blood glucose values based on each optimisation approach are computed. Third, these absolute difference values for each optimisation approach are sorted in the ascending order. Fourth, for each sorted blood glucose value, the optimisation method corresponding to the minimum absolute difference value is selected. Fifth, the accumulate probability of each selected optimisation method is computed. If the accumulate probability of any selected optimisation method at a point is greater than a threshold value, then the accumulate probabilities of these three selected optimisation methods at that point are reset to zero. A range of the sorted blood glucose values are defined as that with the corresponding boundaries points being the previous reset point and this reset point. Hence, after performing the above procedures for all the sorted reference blood glucose values in the validation set, the regions of the sorted reference blood glucose values and the corresponding optimisation methods in these regions are determined. It is worth noting that the conventional lowpass denoising method was performed in the signal domain (either in the time domain or in the frequency domain), while the authors' proposed method is performed in the feature space or the reference blood glucose space. Hence, the authors' proposed method can further improve the reliability of the obtained feature values or the reference blood glucose values so as to improve the accuracy of the blood glucose estimation. Moreover, the individual modelling regression method has been employed here to suppress the effects of different users having different responses of the infrared light to the blood glucose values. The computer numerical simulation results show that the authors' proposed method yields the mean absolute relative deviation (MARD) at 0.0930 and the percentage of the test data falling in the zone A of the Clarke error grid at 94.1176%.

Machine learning-based models for frictional pressure drop prediction of condensing and adiabatic flow in micro, mini and macro channels utilizing universal data

This study proposes a universal machine learning-based model to predict the adiabatic and condensing frictional pressure drop. For developing the proposed model, 11,411 data points of adiabatic and condensing flow inside micro, mini and macro channels are collected from 80 sources. The database consists of 24 working fluids, hydraulic diameters from 0.07 to 18 mm, mass velocities from 6.3 to 2000 Kg/m2s, and reduced pressures from 0.001 to 0.95. Using this database, four machine learning regression models, including “artificial neural network”, “support vector regression”, “gradient boosted regression”, and “random forest regression”, are developed and compared with each other. A wide range of dimensionless parameters as features, “two-phase friction factor” and “Chisholm parameter” are each considered separately as targets. Using search methods, the optimal values of important hyperparameters in each model are determined. The results showed that the “gradient boosted regression” model performs better than other models and predicts the frictional pressure drop with a mean absolute relative deviation of 3.24%. Examining the effectiveness of the new model showed that it predicts data with uniform accuracy over a vast range of variations of each flow parameter.

Void Fraction Measurement Method in Gas/Liquid Two-Phase Stratified Flow

Summary The void fraction is a basic parameter to characterize the two-phase flow. Accurate measurement of its value is of great significance for industrial production. In this study, an optical correction system combined with software and hardware is developed. Stereoscopic particle image velocimetry (SPIV) technology is used to measure the void fraction in gas/liquid two-phase stratified flow. The measurement results are compared with those of 2D particle image velocimetry under the same conditions, and the correction effect appears to be reasonably good. The void fraction value is obtained by the image processing of the corrected image. The accuracy of the measurement results is verified by comparing the experimental liquid-phase volumetric flow rate with that measured through a standard flowmeter. Twenty types of void fraction prediction models are evaluated using the mean absolute relative deviation (MARD), root mean square error (RMSE), and gray correlation analysis methods. The results suggest that the Armand (1946) model has the smallest MARD, RMSE value, the largest correlation degree (Ri) value, and the best prediction performance. On this basis, an improved model is proposed by fitting the relationship between void fraction and gas volume holdup.

A systematic approach based on artificial intelligence techniques for simulating the ammonia removal by eighteen deep eutectic solvents

The current research attempts to model the ammonia removal capacity of deep eutectic solvents (DES) employing artificial intelligence (AI) techniques. Four well-known AI classes apply to estimate the ammonia dissolution amount in eighteen DESs. The DES chemistry (type, molar concentration of components, and water dose), equilibrium temperature/pressure, and DES-ammonia absorption enthalpy help AI models estimate the ammonia solubility in DESs. The combination of trial-and-error and statistical analysis determines the best structure of AI models first. Then, several well-established model selection strategies apply to find the highest-accurate AI class. It was found that the adaptive neuro-fuzzy inference system (ANFIS) is more reliable than the other tested AI models. The designed ANFIS model precisely estimates 1356 experimental samples of ammonia removal by DESs with the mean absolute relative deviation percent of 3.51, mean absolute deviation percent of 0.121, root mean squared error of 0.216, and correlation of determination of 0.9980. The selected AI model monitors the influence of DES components, molar dose, equilibrium pressure, and equilibrium pressure on ammonia elimination by these green solvents. Both trend and relevancy investigations approved that the ammonia absorption tendency of DESs intensifies by increasing the equilibrium pressure, absorption enthalpy, the number of moles of hydrogen bond acceptor (HBA), and molecular weight of hydrogen bound donor (HBD). In addition, increasing the equilibrium temperature, water dosage in DES structure, HBA molecular weight, and the number of HBD moles decreases ammonia removal by DESs.

A new heat transfer model based on non-equilibrium film theory for annular flow condensation of binary zeotropic mixtures

In recent years, zeotropic mixtures have been widely used as substitutes for pure fluids. Accordingly, their condensation properties have attracted increasing attention. However, it is a challenge to accurately describe the heat transfer performance of zeotropic mixtures in condensation, which is limited by the estimation of mixture effects, especially the mass transfer resistance therein. Focusing on this issue, the mass transfer due to the axial and radial concentration gradients in the vapor phase was considered, and a new heat transfer model for the annular flow of binary zeotropic mixtures in horizontal tubes was developed based on non-equilibrium film theory. In this model, the boiling point difference was introduced to approximate the effect of concentration gradients by combining with the vapor-liquid phase equilibrium properties of binary zeotropic mixtures. Secondly, equivalent specific heat capacities were defined for an analogy with single-phase heat transfer, which further simplifies the form of the model. The simplified model was evaluated with a consolidated database of 11 mixtures from 13 sources. The results show that the predictions of the model are in good agreement with the experimental results, with a mean absolute relative deviation (MARD) of 15.88 and 91.14% of points falling in the ±30% deviation band. Predictions were also performed for excluded data from other sources and satisfactory results were obtained with a MARD of 13.47%. Moreover, the mechanism of mixture effects is elaborated. It indicates that besides the commonly mentioned heat transfer degradation, there is also heat transfer enhancement for binary zeotropic mixtures condensation due to the opposite effects of the axial and radial concentration gradients.

Modeling of turbulent diffusion terms for one-dimensional interfacial area transport equation in vertical round channels

The two-fluid model is comprised of the mass, momentum, and energy equations for liquid and gas phases separately. The terms of interfacial transfer govern mass, momentum, and energy transfer across the gas-liquid interface. The interfacial area concentration is critical to formulating the interfacial transfer terms. The one-dimensional Interfacial Area Transport Equation (IATE) has excellent potential in dynamically predicting the interfacial area concentration. The sink and source terms modeling for the interfacial area concentration is vital in the successful IATE development. The current IATE in bubbly flows considers the sink terms caused by bubble coalescence owing to random bubble collision and wake entrainment. The source term is caused by bubble breakup owing to turbulent impact. The turbulent diffusion term has not been considered in the current practice of one-dimensional IATE. This study developed the equation to predict the turbulent diffusion terms in the one-dimensional IATE for bubbly flow in vertical round channels. The predictive capability of the developed equation was validated with the collected database. The non-dimensionalized turbulent diffusion terms predicted by the developed equation agreed with those directly calculated by the experimental data. The developed turbulent diffusion term was also implemented in the one-dimensional IATE. The interfacial area concentrations predicted by the one-dimensional IATE with the turbulent diffusion term had a satisfactory agreement with the experimental results by the statistical parameters of the mean relative deviation of 3.76% and the mean absolute relative deviation of 10.1%.

EXPERIMENTAL INVESTIGATION ON FLOW CONDENSATION HEAT TRANSFER OF METHANE/ETHYLENE MIXTURE INSIDE A HORIZONTAL SMOOTH TUBE

Methane (R50) and ethylene (R1150) are important components in mixed-refrigerant Joule-Thomson (MRJT) refrigeration systems. Among many efforts to improve MRJT refrigerators, the work focusing on the properties of condensing heat transfer in mixtures is essential. Therefore, this paper aims to investigate these characteristics of high-temperature glide mixture R50/R1150 (26.7/73.3&amp;#37; molar fraction) inside a horizontally placed smooth tube. The experiments were done for varying saturation pressures (1.7-2.3 MPa), mass fluxes (101-391 kg m&lt;sup&gt;-2&lt;/sup&gt; s&lt;sup&gt;-1&lt;/sup&gt;), and the entire range of vapor qualities. The results indicate that higher heat transfer coefficients (HTCs) occur at larger mass fluxes and smaller saturation pressures. In addition, the effect of temperature glide was discussed by comparing the HTCs of R50/R1150 with those of R50/R170 reported in the previous literature under the same test condition. It is indicated that higher temperature glide contributes to worse heat transfer performance. Moreover, six existing correlations combined with the equilibrium model were employed to predict the HTCs of R50/R1150. All predictions exhibit an underestimation of the HTCs, and only Dobson and Chato correlation and Shah correlation presented reasonable predictions with a mean absolute relative deviation (MARD) under 30&amp;#37;. Therefore, a new and improved heat transfer correlation was developed based on the pure fluid correlation proposed by Shah. The correlation takes into account the effects of temperature glide and mass transfer resistance and draws on experimental data for hydrocarbon mixtures reported in the existing literature. Comparison with the experimental data shows that the new correlation has good applicability to hydrocarbon mixtures, providing satisfactory predictive values of R50/R1150 with a MARD of 14&amp;#37; and 95.3&amp;#37; of the data within &amp;#177; 30&amp;#37; deviation.

Efficient Computation of the Interaction Energies of Very Large Non-covalently Bound Complexes

AbstractWe present a new benchmark set consisting of 16 large non-covalently bound systems (LNCI16) ranging from 380 up to 1988 atoms and featuring diverse interaction motives. Gas-phase interaction energies are calculated with various composite DFT, semi-empirical quantum mechanical (SQM), and force field (FF) methods and are evaluated using accurate DFT reference values. Of the employed QM methods, PBEh-3c proves to be the most robust for large systems with a relative mean absolute deviation (relMAD) of 8.5% with respect to the reference interaction energies. r2SCAN-3c yields an even smaller relMAD, at least for the subset of complexes for which the calculation could be converged, but is less robust for systems with smaller HOMO–LUMO gaps. The inclusion of Fock-exchange is therefore important for the description of very large non-covalent interaction (NCI) complexes in the gas phase. GFN2-xTB was found to be the best performer of the SQM methods with an excellent result of only 11.1% deviation. From the assessed force fields, GFN-FF and GAFF achieve the best accuracy. Considering their low computational costs, both can be recommended for routine calculations of very large NCI complexes, with GFN-FF being clearly superior in terms of general applicability. Hence, GFN-FF may be routinely applied in supramolecular synthesis planning.1 Introduction2 The LNCI16 Benchmark Set3 Computational Details4 Generation of Reference Values5 Results and Discussion6 Conclusions

Constructing an Intelligent Model Based on Support Vector Regression to Simulate the Solubility of Drugs in Polymeric Media

This study constructs a machine learning method to simultaneously analyze the thermodynamic behavior of many polymer-drug systems. The solubility temperature of Acetaminophen, Celecoxib, Chloramphenicol, D-Mannitol, Felodipine, Ibuprofen, Ibuprofen Sodium, Indomethacin, Itraconazole, Naproxen, Nifedipine, Paracetamol, Sulfadiazine, Sulfadimidine, Sulfamerazine, and Sulfathiazole in 1,3-bis[2-pyrrolidone-1-yl] butane, Polyvinyl Acetate, Polyvinylpyrrolidone (PVP), PVP K12, PVP K15, PVP K17, PVP K25, PVP/VA, PVP/VA 335, PVP/VA 535, PVP/VA 635, PVP/VA 735, Soluplus analyzes from a modeling perspective. The least-squares support vector regression (LS-SVR) designs to approximate the solubility temperature of drugs in polymers from polymer and drug types and drug loading in polymers. The structure of this machine learning model is well-tuned by conducting trial and error on the kernel type (i.e., Gaussian, polynomial, and linear) and methods used for adjusting the LS-SVR coefficients (i.e., leave-one-out and 10-fold cross-validation scenarios). Results of the sensitivity analysis showed that the Gaussian kernel and 10-fold cross-validation is the best candidate for developing an LS-SVR for the given task. The built model yielded results consistent with 278 experimental samples reported in the literature. Indeed, the mean absolute relative deviation percent of 8.35 and 7.25 is achieved in the training and testing stages, respectively. The performance on the largest available dataset confirms its applicability. Such a reliable tool is essential for monitoring polymer-drug systems' stability and deliverability, especially for poorly soluble drugs in polymers, which can be further validated by adopting it to an actual implementation in the future.