基于PCA和自适应神经模糊网络的风电机组执行器主动容错控制

Active Fault-tolerant Control of Wind Turbine Actuator Based on Pca and Adaptive Neural Fuzzy Detection

风电机组执行器的发电机转矩失效会导致系统无法工作在最大功率跟踪点,严重时甚至会引发机组停机,最终造成严重的经济损失和复杂的维修工作。为此,提出一种基于PCA和自适应神经模糊检测的风电机组执行器主动容错控制策略,以确保执行器故障时的高效运行。首先对风电发电机组进行建模,然后引入Luenberger观测器来估计执行器的转矩值,借助主动抗扰控制来抵消转矩执行器故障的影响。在诊断过程中使用PCA和自适应神经模糊算法来检测和分析执行器转矩测量值和观测值之间的差异,进而实现执行器故障的精确诊断。最后搭建实验样机平台对所提方案进行验证,实验结果表明所提PCA和自适应神经模糊检测算法可以实现整体95.2%的执行器故障诊断n准确率,n并在执行时故障发生时及时实现故障隔离。

The generator torque failure of the wind turbine actuator will cause the system to fail at the maximum power tracking point,and even cause the unit to shut down in serious cases,resulting in serious economic losses and complex maintenance work.Therefore,an active fault-tolerant control strategy for wind turbine actuators based on PCA and adaptive neural fuzzy detection is proposed to ensure the efficient operation of actuators when they fail.First,the wind power generator set is modeled,and then the Luenberger observer is introduced to estimate the torque value of the actuator,and the effect of the torque actuator failure is offset by active disturbance rejection control.In the process of diagnosis,PCA and adaptive neural fuzzy algorithm are used to detect and analyze the difference between the measured value and the observed value of the actuator torque,so as to realize the accurate diagnosis of the actuator fault.Finally,an experimental prototype platform is built to verify the proposed scheme.The experimental results show that the proposed PCA and adaptive neural fuzzy detection algorithm can achieve the overall fault diagnosis accuracy of 95.2%,and realize timely fault isolation in case of actuator fault.