Optimization of Maximum Lift to Drag Ratio on Airfoil Design Based on Artificial Neural Network Utilizing Genetic Algorithm

Abstract

.This paper deals with an alternative design method of airfoil for wind turbine blade for low wind speed based on combination of smart computing and numerical optimization. In this work, a simulation of Artificial Neural Network (ANN) for determining the relation between airfoil geometry and its aerodynamic characteristics was conducted. First, several airfoil geometries were generated through transformation of complex variables (Joukowski transformation), and then lift and drag coefficients of each airfoil were determined using CFD (Computational Fluid Dynamics). In present study, the ANN training was conducted using airfoil geometry and its aerodynamic characteristics as input and output, respectively. Therefore, lift and drag coefficients can be directly determined only by giving the airfoil geometry without having to perform wind tunnel experiment or numerical computation. Moreover, the optimization was conducted to obtain an airfoil geometry which gives maximum lift to drag ratio (CL/CD) for specific Reynolds number. For this purpose Genetic Algorithm (GA) was applied as optimizer. The results were validated using commercial CFD and it can be shown that the result are satisfactory with error approximately of 6%.

Aerodynamic surrogate model based on deep long short-term memory network: An application on high-lift device control

“…Intelligent optimization methods are always applied with surrogate models, which perform well in aerodynamic design. 40,41 Particle swarm optimization is a heuristic optimization algorithm simulating bird flocking/fish schooling and aggregation behaviors, which is applied widely in optimization. 32 Optimal design values are searched within a population called swarm.…”

A Transendothelial Leukocyte Transmigration Model Based on Computational Fluid Dynamics and BP Neural Network

“…For example, the BP model can effectively predict the displacement and dominant frequency of the vortex-induced vibration of flexible cylinders commonly found in engineering. But the neural network has a significant disadvantage, that is, the successful training of the network needs to rely on a considerable number of samples (Liu A sufficiently trained neural network model also can guide engineering design, such as the rational planning of the airfoil design process by predicting the leading edge pressure (Haryanto et al, 2014;Peng et al, 2020;Tang et al, 2020). Because of the wide application of BPNN model in fluid mechanics, we integrated the leukocyte transmigration CFD dataset to establish a time-dependent leukocyte transmigration prediction model.…”

A review of the artificial neural network surrogate modeling in aerodynamic design

“…This makes the use of ANN an efficient way to reduce the computational time, since ANN decouples the aerodynamic solver from the optimization process where the GA operates simultaneously with ANN. 36 ANN is employed as one of the reliable and fast methods of predicting aerodynamic coefficients to select optimized airfoils 10 as well as a lift-drag ratio optimization for airfoil 46 with GA. Approximated pre-evaluations based on ANN are used in a hybrid optimization procedure with GA to determine airfoil shape in three-dimensional wing design to benefit from the accumulated knowledge thus reducing the number of CFD evaluations required at each generation. 47 There are also other types of combination of optimization algorithms with ANN.…”