Abstract In this thesis, a simple dynamic model of the planetary gear train system is developed. Because of the dynamic equations deriving from designing a planetary gear train system are complex and nonlinear, and the controller design is difficult. If we take the planetary gear train system as a pure speed-down mechanism, and then the accuracy of the planetary gear train system will lose. So, we develop the dynamic equations of the planetary gear train system concerning with the conception of friction losses. Furthermore, the MGA method is used to identify the parameters of this system. The modified genetic algorithm (MGA) is proposed from the simple genetic algorithm (SGA) with some additional strategies, such as Elitist and Extinction strategies. From the computer simulations and the experimented results, it is concluded that the parameters of this system searched by using MGA will be more precise than the parameters searched by using LMS.

Two parallel planetary gear trains design are proposed to construct a dual input transmission mechanism system used in small power wind turbine systems. The time varied input wind powers are applied in the system with specified speed and torque. The Dynamic power flow variation in gear pairs are modeled and simulated in this work. Results indicate the proposed planetary gear train system is feasible in wind turbine system. The effect of gear train parameters on the operation safety and life will also be studied. The dynamic torque equilibrium equations between meshed gear pairs are employed to model the dynamic torque flow in this proposed dual input gear system. The nonlinear behavior of a synchronous generator has also included in the modeling. The dynamic responses of the dual input transmission mechanism system are simulated by using the 4th order Runge-Kutta method. The effect of system parameters used in this wind turbine system, i.e. the wind speed, the magnetic flux synchronous generator, the inertia flywheels, on the output electrical power variation have investigated in this study. The strength analyses of gear pairs with the bending fatigue and surface durability consideration have also studied in this work.