RETRACTED: Development of hybrid machine learning model for simulation of chemical reactors in water treatment applications: Absorption in amino acid

• DOI: 10.1016/j.eti.2022.102417
• Publication Date: 2022-02-23T00:00:30Z
• Authors: Yanjie Zhang; Lakshmi Thangavelu; Talib Zeedan Taban; Walid Kamal Abdel...; Wanich Suksatan; Mohd Sani Sarjadi; Md Lutfor Rahman; Shaheen M. Sarkar; May Alashwal; Mohammed Zwawi; Mohammed Algarni
• Keywords: Gradient boosting; Acid gas; Boosting

Separation and capture of CO2 from gas mixtures is great importance environmental point view which can be effectively achieved using amino acids as new class chemical absorbents. However, screening the proper absorbent with desired separation properties experimental measurements tedious costly. The predictive computational techniques employed to overcome this problem. In study, for estimating analyzing solubility in solvents based on acid salt solutions, we created two regression models different classes machine learning methods. main aim analyze effect physico-chemical parameters dissolution solvent carried out reactors separation/conversion applications. A number data are collected resources used training validation computations. Several inputs were considered developed models. Inputs task T (temperature), weight% (overall mass percentage solvent), PCO2 (partial pressure gas), MW-am (molecular weight salt), MPC (melting MWC cation). task, must predict alpha (CO2 loading solution) only output studied research Gaussian process decision tree boosted Gradient boosting. With R2 criterion, scores boosting obtained 0.985 0.993, respectively. As third efficiency metric models, RMSE criterion error rates 1.10E−01 1.44E−01. work indicated reliable robust enough a particular application save time cost measurements.