Design of optimal flow concentrator for vertical-axis wind turbines using computational fluid dynamics, artificial neural networks and genetic algorithm

Abstract

Wind energy extraction is one of the fastest developing engineering branches today. Number of installed wind turbines is constantly increasing. Appropriate solutions for urban environments are quiet, structurally simple and affordable small-scale vertical-axis wind turbines (VAWTs). Due to small efficiency, particularly in low and variable winds, main topic here is development of optimal flow concentrator that locally augments wind velocity, facilitates turbine start and increases generated power. Conceptual design was performed by combining finite volume method and artificial intelligence (AI). Smaller set of computational results (velocity profiles induced by existence of different concentrators in flow field) was used for creation, training and validation of several artificial neural networks. Multi-objective optimization of concentrator geometric parameters was realized through coupling of generated neural networks with genetic algorithm. Final solution from the acquired Pareto set is studied in more detail. Resulting computed velocity field is illustrated. Aerodynamic performances of small-scale VAWT with and without optimal flow concentrator are estimated and compared. The performed research demonstrates that, with use of flow concentrator, average increase in wind speed of 20%–25% can be expected. It also proves that contemporary AI techniques can significantly facilitate and accelerate design processes in the field of wind engineering.

Optimal Design of Novel Blade Profile for Savonius Wind Turbines

“…There are many different algorithms for optimizing VAWT efficiencies, such as artificial neural networks and genetic algorithms [27]. This study chooses the Taguchi Methods of Robust Design, an engineering methodology for improving productivity at the research and development stage.…”

Model-free adaptive sliding mode control for intelligent vehicle longitudinal dynamics

“…In this paper, genetic algorithm is used to optimize the parameters of MFASMC controller. 24,25 In order to ensure that the optimization effect meets the requirements of conventional scenarios, parameter optimization is carried out based on the following three typical working conditions: step input response, ramp input response and impulse input response. Individual fitness is the difference between simulation result of MFASMC controller and desired acceleration.…”

Development of an ANN model for prediction of tool wear in turning EN9 and EN24 steel alloy

“…The human brain’s inspiration in processing complex data leads to the evolution of Artificial Neural Network (ANN) to solve complex engineering problems. 40 ANN emulate the human brain and process data based on the input presented to them. 41 ANN process data fed to them with interconnected neurons, the transformation of information rest on the strength between two adjacent neurons termed as weights.…”