Development of wind turbines to operate in modified axial flows which contain swirl velocities and non-uniform distributions

In pan A of this thesis, a procedure based on lifting line theory for the design of wind turbines operating in non-uniform, non-axial but axisymmetric flows is presented. This procedure was used for the design of conventional turbines which were compared with turbine designs produced by momentum theory. The overall trends of both theories were found to be similar, although the lifting line procedure was found to produce a more conservative estimate of the turbine performance. The above mentioned design procedure was also used for the turbine blade design of the wind power systems presented in parts B and C of the thesis. Part B of the thesis deals with the development of the delta wing-turbine system: The system was scaled-up using the results of a previously developed design model, and its dimensions were compared with those of equivalent conventional turbines. It was found that the system compares well with conventional turbines up to rated power values equal to 100 kW. Its advantages were found to be the lower turbine diameter required for a given power output and the opportunity it provides for direct connection of the turbines to generators. The cost of this advantage is the relatively large delta wing required. A system prototype with power output in the order of 1 kW was designed for testing. The prototype turbine blades were designed taking into account Reynolds number effects. In order to overcome the detrimental Re effects, use of a low Re aerofoil (GOE 795), reduction of the turbine number of blades to 10, increase of the blade chord, linear blade chord distribution and variable optimum angle of attack were found to be necessary, leading to a reduction of the turbine power coefficient drop to 4.7% below that of the original high Re design. The prototype off design performance was predicted and it was found that increase of the blade chord at the hub region (for strength) and linearisation of the blade optimum twist angle (for ease of manufacture) did not affect the turbine performance significantly. The generator to be used with the prototype turbine was bench-tested. Its model parameters and power losses were identified. For matching the generator with the turbine, an appropriate load for the generator was found. The prototype long-term performance was also estimated using the turbine performance characteristics, the generator test results and the Weibull distribution of wind occurrence probability. It was found that the generator is not ideally suited for the prototype turbine and that a generator of larger ratings would be more suitable. Finally, the effects of yaw on the delta wing vortices were investigated experimentally. This was done in order to determine the feasibility of using the delta wing yaw to regulate the system power output. It was found that the above mentioned regulation technique can be used, provided that undue blade vibrations due to turbine-vortex misalignment and vortex bursting will not occur. In part C, a procedure for the design of the counter-rotating turbine blades was developed. The above mentioned lifting line procedure as well as the existing knowledge of wind turbine wakes and counter-rotating rotor aerodynamics were used for the design of the counter-rotating turbine blades and the semi-empirical modelling of the two rotors' interaction. The optimum axial distance between the two rotors was found to be equal to 1.4 times the rotors' radius. It was demonstrated that proper design of the turbine blades and appropriate axial positioning of the two rotors could increase the turbine performance by 27.4% above that of the original counter-rotating turbine design, called Trimble Mill. It was also found that a considerable increase of the generator effective rotational speed (equal to 58%) can be achieved by the counter-rotating turbine, compared to that of a conventional turbine with the same number of blades, while the two turbines' power output was found to be at the same levels