1 Evaluation Optimal Friction Factor Correlation in Turbulent Pipe Flow by Genetic Algorithm Qais Abid Yousif 1 , Omar Rafae Alomar 1 , Ibrahim Atiya Mohamed 1 , Majid Kh. Najm 1 (kaisyusuf@ntu.edu.iq, omar.alomar@ntu.edu.iq, ibams_1962@ntu.edu.iq, dr.majid.najim@ntu.edu.iq) 1 Northern Technical University, Engineering Technical College of Mosul, Cultural Group Street, Mosul, Iraq Abstract. The prediction accuracies of friction factor correlations in turbulent pipe flow have remained unsatisfactory due to Colebrook equation is characterized as an implicit correlation. Thus, this works deals with numerical simulation for optimization the correlation of friction factor (fD) in turbulent pipe flow. Genetic Algorithms (GAs) method has been used to evaluate the accuracy of six most used explicit models as an alternative to the Colebrook equation. The fD has been estimated for higher ranges of Reynolds Number (Re) and the relative roughness of pipe (  ⁄ ). The evaluation process has been implemented through comparing the percentage of differences between the values of fD obtained using those correlations with that obtained using Colebrook equation. The optimized results clearly show that the Model-1 and Model-5 provide the lowest percentage of difference as compared to the other explicit models. Results indicated that GAs has succeeded in reducing the computational time by eliminating the iterative process. Keywords: Colebrook equation, friction factor, genetic algorithms. 1. Introduction Despite the great scientific progress that has been made in computer systems and the multiplicity of mathematical methods and programs developed, but still, the Colebrook equation is used in the evaluation of the coefficient of friction [1]. Although the accuracy of Colebrook equation is debatable, it is occasionally necessary to achieve an accurate solution of this equation extremely important for the scientific calculations and frequently for comparisons [2]. Various techniques have been utilized either to compute or to appraise the friction factor precisely. The greater part of recent efforts has been utilized Genetic Algorithms (GAs) and Artificial Neural Network method (ANN) for computing friction factor [3]. However, the use of GAs in the area of heat transfer is very newfangled. This is perhaps due to the computational time needed to reach the final solutions is too long. Recently, the GAs has been increasingly used in heat transfer problems. In 1994, Queipo et al. [4] have expected that the heat transfer community will witness considerable attention for involving new methods such as techniques (i.e. GAs) in many intricate thermo-science problems that confessing optimization. One that has helped to facilitate the use of these exciting technologies in the development of high-performance computing devices, which has increased interest in their use in the field of heat transfer. The IMDC-SDSP 2020, June 28-30, Cyberspace Copyright © 2020 EAI DOI 10.4108/eai.28-6-2020.2297927