Not only about wind energy, natural energies of ocean, sun, etc. are clean and safe, but fluctuation of properties are large and has poor reproducibility, so that most important subject is to find out methods to take out those energies effectively.
To simulate and control windmill power system by control sequence, we need to obtain the accurate model of the wind power generating system previously. In chapter 2, we discuss a new model of a windmill and a method for determination of the windmill torque loss coefficients of a windmill power system. Also, we apply the neural network method to model a small-scale windmill power system. We manifested the validity of the proposed method by simulation results from both the simulated results for sinusoidal change of wind speed and natural wind speed by random sequence.
In chapter 3, we describe the development of a windmill simulator, to simulate accurately the movement of a windmill. The simulator has a model to facilitate the change of windmill
parameters, such as power coefficients or moment of ineertia, by overcoming the weak points of research and development environment for wind utilization.
In chapter 4, for the purpose of cooperative group control of genetators, a control method which is possible to take out wind energy effectively even system characteristics is unknown, is suggested. This control method is possible to follow quasi-maximum output only by measurements of load terminal voltage and current which are relatively simple to measure high accuracy. Hill-climbing method among adaptive control method was adopted. Secondly, fuzzy infrence is applied to the output control. The control method hereby proposed is the one to allow the measured wind speed and windmill speed to be the input of the fuzzy control rule by regarding these factors as ambiguous data and to permit the values of the load resistance to be the output. Finally, we proposed output control method by estimating system characteristics measuring wind speed and windmill speed. Less