摘要
为使风电机组控制技术持续进步,不断提升机组风能利用率,为碳达峰、碳中和服务。在风电机组偏航控制中引入了数据驱动优化方案。利用二维度数据归类方法实现了偏航静态偏差大数据辨识策略的构建,完成了准确辨识偏航静态偏差并合理补偿纠偏的控制实现。在偏航动态控制过程中,构建了考虑误差值和误差停留时间联合判断偏航动作需求的新型控制方案;并利用KNN数据聚类的方法,寻找最优的误差停留时间参数的数据驱动优化策略。新型数据驱动偏航优化策略在选定风场做了长时间实验验证,数据表明,偏航静态偏差辨识结果准确,偏航动态控制方法能有效自适应风况变化,偏航次数得以优化,偏航对风准确度概率得以提升,整体发电效率提升明显。
In order to make continuous progress in wind turbine control technology,continuously improve the utilization rate of wind energy,and serve carbon peak and carbon neutralization. The data-driven optimization scheme is introduced into the yaw control of wind turbine. Using the two-dimensional data classification method,the construction of large data identification strategy of yaw static deviation is realized,and the control realization of accurately identifying yaw static deviation and reasonable compensation and correction is completed. In the process of yaw dynamic control,a new control scheme considering the joint judgment of error value and error residence time is constructed. KNN data clustering method is used to find the optimal data-driven optimization strategy of error residence time parameters. The new data-driven yaw optimization strategy has been tested for a long time in the selected wind field. The data show that the identification result of yaw static deviation is accurate,the yaw dynamic control method can effectively adapt to the change of wind conditions,the yaw times can be optimized,the yaw accuracy probability to the wind can be improved,and the overall power generation efficiency can be significantly improved.
作者
刘栋
王景丹
卢晓光
LIU Dong;WANG Jing-dan;LU Xiao-guang(Xuchang XJ Wind Power Technology Co.,Ltd.,Xuchang 461000,China)
出处
《自动化与仪表》
2022年第10期30-34,共5页
Automation & Instrumentation
关键词
数据聚类
风电机组
偏航控制
风能利用率
data clustering
wind turbine
yaw control
wind energy utilization
