The power output of two straight-bladed vertical-axis wind turbines is simulated using computational fluid dynamics (CFD) as well as analyzed and optimized using the Taguchi method. Five operating factors of incoming flow angle (13), tip speed ratio (lambda), turbine spacing (S/d), rotational direction (RD), and blade angle (4)) along with four levels are taken into consideration to account for their influences on the performance of the dual turbine system. An orthogonal array of L-16 (45) is designed. The profile of extent indicates that the factors 2,, and play crucial roles in determining power output, whereas the factor phi almost plays no part on the power output. The influence strength order of each factor is featured by lambda > beta > RD > S/d > phi. Furthermore, the analysis of the signal-to-noise (S/N) ratio suggests that the combination of the five factors for maximizing the power output of the system is located at lambda = 2, beta = 120, (clockwise, counterclockwise), S/d =3, and phi = 0 degrees. With this operation, flow velocity in three regions beyond, below, and between the two turbines is enhanced from their interaction, whereas it drops drastically in the wake regions. Compared to the single wind turbine operated at lambda = 2 along with the same wind speed (=8 m s(-1)) and counterclockwise rotation, the mean power coefficient (C-p, (average)) of the dual turbine system operated at the optimal combination is enlarged by 9.97%. (C) 2016 Elsevier Ltd. All rights reserved.