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.

This study proposes a simple correlation for approximating hydrogen solubility in biomaterials as a function of pressure and temperature. The pre-exponential term of the proposed model linearly relates to the pressure, whereas the exponential term is merely a function of temperature. The differential evolution (DE) optimization algorithm helps adjust three unknown coefficients of the correlation. The proposed model estimates 134 literature data points for the hydrogen solubility in biomaterials with an excellent absolute average relative deviation (AARD) of 3.02% and a coefficient of determination (R) of 0.99815. Comparing analysis justifies that the developed correlation has higher accuracy than the multilayer perceptron artificial neural network (MLP-ANN) with the same number of adjustable parameters. Comparing analysis justifies that the Arrhenius-type correlation not only needs lower computational effort, it also has higher accuracy than the PR (Peng-Robinson), PC-SAFT (perturbed-chain statistical associating fluid theory), and SRK (Soave-Redlich-Kwong) equations of state. Modeling results show that hydrogen solubility in the studied biomaterials increases with increasing temperature and pressure. Furthermore, furan and furfuryl alcohol show the maximum and minimum hydrogen absorption capacities, respectively. Such a correlation helps in understanding the biochemical–hydrogen phase equilibria which are necessary to design, optimize, and control biofuel production plants.

This study correlated biomass heat capacity (Cp) with the chemistry (sulfur and ash content), crystallinity index, and temperature of various samples. A five-parameter linear correlation predicted 576 biomass Cp samples from four different origins with the absolute average relative deviation (AARD%) of ~1.1%. The proportional reduction in error (REE) approved that ash and sulfur contents only enlarge the correlation and have little effect on the accuracy. Furthermore, the REE showed that the temperature effect on biomass heat capacity was stronger than on the crystallinity index. Consequently, a new three-parameter correlation utilizing crystallinity index and temperature was developed. This model was more straightforward than the five-parameter correlation and provided better predictions (AARD = 0.98%). The proposed three-parameter correlation predicted the heat capacity of four different biomass classes with residual errors between −0.02 to 0.02 J/g∙K. The literature related biomass Cp to temperature using quadratic and linear correlations, and ignored the effect of the chemistry of the samples. These quadratic and linear correlations predicted the biomass Cp of the available database with an AARD of 39.19% and 1.29%, respectively. Our proposed model was the first work incorporating sample chemistry in biomass Cp estimation.