최대출력제어를 위한 풍속예측에 관한 연구

Abstract

The power control of the wind turbine system can be divided with two types depending on the wind speed. The two types are the maximum power control(MPC) and the limited power control. MPC is used for obtaining the maximum power below the rated wind speed. The limited power control is utilized to regulate the rated power above the rated wind speed. Perturbation and Observation(P&O) method and Optimal Tip Speed Ratio(TSR) control are most widely known methods for MPC. Perturbation and Observation(P&O) method is controlled by comparing the present and past values ​​of the rotation speed. However, due to a number of processes, it has slow response time to track a maximum power. Optimal TSR control method requires a wind speed. However, measuring wind speed is not feasible since it is installed some distance away from wind turbines. For this reason, the estimated wind speed has been used for optimal TSR control. But the problem is that estimating the wind speed is related on inverse problems, thus, it is not so simple to solve. Because of these, neural networks or various numerical methods have been used to solve these problems. In this paper, two new methods are introduced for MPC. The first method is to use a Regression method for estimating the wind speed through the power and rotational speed. And the second method uses the power curve which is a function of the rotation speed and wind speed for estimating the wind speed. Matlab/Simulink is used for simulations. To confirm the performance of proposed methods, 9.4kW PMSG model is used in the two difference wind case(at constant wind speed and at variable wind speed).