LAUSR

This paper presents an adaptive multi-tracker optimization algorithm (AMTOA) for global optimization problems with an emphasis on applications in chemical engineering. To obtain the AMTOA, first, several modifications are performed on the conventional multi-tracker optimization algorithm (MTOA). Then a number of its parameters are considered to be adaptive. The modifications include a novel way of determining the search radius of each global tracker (GT\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$G_{\text{T}}$$\end{document}), and introducing a more efficacious technique of searching for a new solution by GT\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$G_{\text{T}}$$\end{document}s. GT\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$G_{\text{T}}$$\end{document}s are the main components of the MTOA which look for the global optimal point (GOP\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${\text{GOP}}$$\end{document}). Additionally, the adaptation rules are employed for GT\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$G_{\text{T}}$$\end{document}s search radii and their searching parameters. These modifications lead to increasing the precision of the solution and reliability of the algorithm, both of which are the most important properties of an optimizer. Reducing the number of parameters of MTOA is another advantage of AMTOA. The results of applying this algorithm to several unconstrained and constrained general benchmarks along with several chemical engineering optimization problems reveal that AMTOA outperforms other well-known methods such as genetic algorithm (GA), particle swarm optimization (PSO), gray wolf optimizer (GWO), whale optimization algorithm (WOA), and conventional MTOA. Additionally, comparing the results of AMTOA to other advanced optimization algorithms such as LSHADE44, MA-ES, and IUDE show its superiority for chemical engineering optimization problems. Thus, the development of AMTOA could be advantageous to the area of chemical engineering.