Published: Aug. 8, 2024
Software programs for parameter estimation, phase visualization and predictive modeling of supercritical extraction process data using algorithms is presented in this work. A contextually appropriate, iterative, ordinary least squares estimation selection method developed estimating model coefficients density based semi empirical equations associated with data. Visualization the behaviors projected by specific semiempirical equation(s) also performed iteratively plotting three-dimensional surfaces involving state variables solute solubility mole fraction. Predictive input has been implemented three supervised machine learning (Multilayer perceptron, K-nearest neighbors Support vector regression). Hyperparameter optimization prior to prediction. Detailed analysis prediction conducted standard scoring metrics descriptive charts. Theoretical inference discrepancies regarding predicted window maximum/optimal solubility, efficiency, vapor liquid equilibrium have elucidated from program outputs. In summary, these are unique, accurate, reliable simple computational tools evaluating/designing
Language: Английский