Minimize pressure drop and maximize heat transfer coefficient by the new proposed multi-objective optimization/statistical model composed of “ANN + Genetic Algorithm” based on empirical data of CuO/paraffin nanofluid in a pipe

• Published in: Physica A 2019-08, Vol.527, p.121056, Article 121056
• Main Authors: Bagherzadeh, Seyed Amin, Sulgani, Mohsen Tahmasebi, Nikkhah, Vahid, Bahrami, Mehrdad, Karimipour, Arash, Jiang, Yu
• Format: Article
• Language: English
• Subjects: Artificial neural network; CuO/liquid paraffin nanofluid; Empirical results; Genetic algorithm; Multi-objective optimization
• ISSN: 0378-4371; 1873-2119
• DOI: 10.1016/j.physa.2019.121056

A new multi-objective optimization model composed of the artificial neural network (ANN) and the genetic algorithm (GA) methods based on the empirical thermo-physical characteristics of CuO/liquid paraffin nanofluid flow in a pipe is presented for the first time. It means a new optimization /statistical approach is achieved based on ANN together with GA; so that at first ANN is employed to predict the nanofluid thermo-physical properties and then the heat transfer coefficient and the pressure drop ratios of the nanofluid to the basefluid, are optimized as well as to minimize the pressure drop ratio and maximize the heat transfer coefficient ratio by using the multi-objective optimization approach of GA. The results of the multi-objective optimization via the GA show that the Pareto optimal front quantifies the trade-offs in satisfying the two fitness function of heat transfer coefficient and the pressure drop ratios. •Minimize pressure drop and maximize heat transfer coefficient.•New proposed multi-objective optimization/statistical model.•ANN plus Genetic Algorithm based on empirical data.