Rigorous modeling of gypsum solubility in Na–Ca–Mg–Fe–Al–H–Cl–H2O system at elevated temperatures

摘要: Precipitation and scaling of calcium sulfate have been known as major problems facing process industries oilfield operations. Most scale prediction models are based on aqueous thermodynamics solubility behavior salts in electrolyte solutions. There is yet a huge interest developing reliable, simple, accurate models. In this study, comprehensive model least-squares support vector machine (LS-SVM) presented, which mainly devoted to dihydrate (or gypsum) solutions mixed electrolytes covering wide temperature ranges. respect, an aggregate 880 experimental data were gathered from the open literature order construct evaluate reliability presented model. Solubility values predicted by LS-SVM well accordance with observed yielding squared correlation coefficient (R 2) 0.994. Sensitivity for some important parameters also checked ascertain whether learning has succeeded. At end, outlier diagnosis was performed using method leverage value statistics find eliminate falsely recorded measurements assembled dataset. Results obtained study indicate that can successfully be applied predicting Na---Ca---Mg---Fe---Al---H---Cl---H2O system over temperatures ranging 283.15 371.15 K.