Wind turbine wakes have significant effects on the production efficiency and fatigue loads, and these effects should be considered in the optimization of wind turbine structure and wind farm layouts. In this paper, a numerical model combining Lagrangian dynamic large-eddy models and actuator line methods (ALMs) is implemented to investigate the wind turbine near wakes at three representative tip speed ratios (TSRs). In the model, several model parameters that have been. justified based on the existing literature and experiments are utilized to enhance the numerical stability and accuracy. These parameters are related to a physically meaningful length scale in the Gaussian smoothing function, a Prandtl tip/hubloss factor and a 3D correction for airfoil data. The model is compared to the MEXICO measurements, in which a detailed stereo Ply measurement is carried out. According to the comparison of rotor power coefficients between the prediction and the measurements, there is a slight overestimate at TSRs of 6.67 and 10, while a slight underestimate at TSR of 4.17. Additionally, according to the comparison of streamwise traverses and spanwise distribution of axial velocities, good agreement is achieved at both TSRs of 4.17 and 6.67, while visible difference is found at TSR of 10. Moreover, the simulation result shows a helical behavior of wake tip vortices induced by the turbine rotor. This behavior gets more pronounced with a decreasing TSR. The tip vortices also give a reasonable explanation of why the maximum velocity deficit and turbulence intensity occur near the blade tip of wind turbines. (C) 2015 Elsevier Ltd. All rights reserved.