A new intelligent prediction model using machine learning linked to grey wolf optimizer algorithm for <scp>O₂</scp>/<scp>N₂</scp> adsorption

Abstract To address the deficiency and predict adsorption performance in different adsorbents, this study proposes a new optimizer linked to machine learning (ML) model considering of process. The main goal is under process conditions with adsorbents provide unified framework, leading prediction phenomena instead traditional isotherm models. This research focuses on predicting adsorbed amount O 2 N several carbon‐based using ML approach grey wolf algorithm (GWO). Experimental data (dataset 1344) adsorbent type, temperature, pressure, gas capacity were used as input output datasets. best was Broyden–Fletcher–Goldfarb–Shanno (BFGS), two‐layer network from multi‐layer perceptron (MLP) method applying 28 neurons. MLP‐GWO would have mean square error (MSE) efficiencies 0.00037, while R ( r ‐squared) 0.9934. ML‐generated can accurately behaviour various conditions. results potential assist wide range separation industries.