A study of 5 kW wind turbine performance in a wind flow obstruction building using the CFD technique

Abstract

This dissertation presents a study on techniques for a small and low wind speed turbine installation based on the small and low wind speed machine performance prediction in the wind flow obstruction area using the Computational Fluid Dynamics (CFD) technique. A Horizontal Axis Wind Turbine (HAWT) is commonly used in the wind energy sector compared to the other types of the machines. Small Wind Turbines (SWTs) in the scale of wind machines being not over 10 kW in power generation are commonly used in urban areas where are mostly surrounded by buildings that cause obstructed wind flows to the machines. The power performance of SWTs is significantly affected by the turbulence and wind shear. As a consequence, before installing SWTs in such areas, it is necessary to consider the effects of some parameters in order to reduce power loss and installation time. In this study, three units of 5 kW HAWTs installed in the Defense Energy Training Center (DETC) in Rayong Province, Thailand, were studied. The three turbulent models of CFD, comprising k-ε, k-ω, and RNG, were applied in the boundary conditions to investigate the performances and characteristics of wind flows to the machines. The wind flows being studied were from two directions, the North-East (NE) and the South-West (SW), both of which are the common wind flow directions in Thailand. Additionally, the wind velocity profile and the site information were collected and compared with the CFD data. From the results of the CFD technique, the wind machines No.1 (WT1), No.2. (WT2), and No.3 (WT3) were compared with the wind machine located without any obstruction to the wind flow (WT0). It was found that: (1) Using the k-ε model with the wind flow from NE, the power performances of WT1, WT2, and WT3 were higher than that of WT0 by approximately 31%. For the wind flow from SW, the power performances of WT1 and WT2 were higher than that of WT0 by approximately 7%, while the power performance of WT3 was lower than that of WT0 by approximately 7%; (2) Using the k-ω model with wind flow from SW, the power performances of WT1 and WT2 were higher than that of WT0 by approximately 12%, while the power of performance of WT3 was lower than that of WT0 by approximately 2%; (3) Using the RNG model with the wind flow from SW, the power performances of WT1, WT2 were higher than that of WT0 by approximately 9%,while the performance of WT3 was lower than that of WT0 by approximately 6%. The results of the CFD technique based on the three turbulent models showed similarities and trends in terms of power performances. Moreover, the study also measured the wind velocity behaviors in the site compared with the CFD technique results by using two anemometers which were installed at the same height of the SWTs in Building A and Building C in direction of wind flow from SW to Building A through to WT3-WT2-WT1-Building C. The measurement results from anemometers showed that the wind velocity coming into Building A was lower than that coming out of Building C, which were similar to the CFD investigation results. Therefore, the study showed that the behavior of the wind flow from SW and NE passing the buildings through to the SWTs affected an increase or decrease of power generation of the SWTs when compared with the SWT located in the site that did not obstruct the wind flow. Thus, it is suggested that the height of the turbine and the distance between the wind machine and the building were the key parameters to consider before installing this kind of the wind machines. Consequently, the CFD technique together with the three turbulence models can be used to predict the power performance and the installation of an SWT in any building-obstructed wind-flow area.