Online Warning Method of Wind Turbine Yaw to Wind Error Based on Interval Division
In view of the problem that additional equipment is needed to detect the yaw of wind turbines,the historical data stored in SCADA system is analyzed by using the fan mechanism and data-driven method,and an interval division based yaw warning method is proposed.Firstly,irrelevant data is deleted according to the state variables such as power limiting,yaw,failure and maintenance of the unit,and random noise in the data sample is eliminated by using the method combined with random forest and 3σ principle.Then,the power is divided into equal intervals,and the scatter plot within the two-dimensional interval of wind speed and yaw error is fitted into a quadratic curve within each power interval.The yaw error corresponding to the lowest point of the curve is the yaw to wind error.The test results verify the effectiveness of the method.Finally,an online early warning system for yaw to wind error is designed to realize the closed-loop landing of the unit yaw to wind error correction.
wind turbineyaw to wind errorinterval divisionleast square method
NETL
NSTL
万方数据
