摘要
研究基于混合复数域CNN的风力发电机组塔架振动频率动态跟踪方法,动态跟踪风力发电机组塔架振动频率,抑制谐波干扰的影响。应用坎贝尔图法分析风力发电机组与塔筒共振点,确定自适应陷波器结构与陷波器自适应调整迭代搜索区间,应用黄金分割法在迭代搜索区间内,通过更新变步长、引入约束敏感因子,动态跟踪风力发电机组塔架振动中心频率,并应用融合估计方式确定振动基频;通过级联自适应陷波器与混合复数域CNN网络,抑制所跟踪到的风力发电机组塔架振动中心频率和基频中有用信号频率干扰,获取更精准的振动频率跟踪结果。实验结果表明:该方法可有效跟踪风力发电机塔架振动中心频率,降低风力发电机在运行时叶片拍打、叶片面外力矩振动幅度及塔架振动幅度,更好满足风力发电机组塔架振动控制需要。
A dynamic tracking method for wind turbine tower vibration frequency based on hybrid complex domain CNN was studied to dynamically track wind turbine tower vibration frequency and suppress the impact of harmonic interference.The campbell diagram method was applied to analyze the resonance point between the wind turbine generator and the tower,determine the adaptive notch structure and adaptive adjustment iterative search interval of the notch.The golden section method was applied to dynamically track the vibration center frequency of the wind turbine tower by updating the variable step size and introducing constraint sensitivity factors within the iterative search interval,and the fusion estimation method was applied to determine the vibration fundamental frequency;Cascaded adaptive notch filters and hybrid complex domain CNN networks were used to suppress useful signal frequency interference in the center frequency and fundamental frequency of wind turbine tower vibration tracked,and obtain more accurate vibration frequency tracking results.The experimental results show that this method can effectively track the center frequency of wind turbine tower vibration,reduce blade slapping,blade out-of-plane torque vibration amplitude,and tower vibration amplitude during operation of the wind turbine,and better meet the needs of wind turbine tower vibration control.
作者
李萍
LI Ping(College of Electrical and Information Engineering,Heilongjiang Institute of Technology,Harbin 150001,China)
出处
《微电机》
2023年第6期31-37,共7页
Micromotors
关键词
混合复数域
CNN网络
风力发电机组
塔架振动频率
动态跟踪
抗干扰
mixed complex field
CNN network
wind turbine
tower vibration frequency
dynamic track-ing
anti-interference
