大型离岸风电机组的故障监测策略研究  

Fault monitoring scheme for large-scale offshore wind turbines

作　　者：李鹏[1] 宋永端[1] 刘卫 秦明 

机构地区：[1]北京交通大学智能系统及再生能源研究中心,北京100044 [2]国电联合动力技术有限公司风电技术研究所,北京100044

出　　处：《电力系统保护与控制》2012年第1期43-49,共7页Power System Protection and Control

基　　金：北京交通大学人才基金(KWRC11004536);国家973计划项目(2012CB125202)~~

摘　　要：针对大型水平轴离岸风力发电机,建立考虑塔架振动的动力学模型,研究大于额定风速区域,变桨控制模式下的机组状态监测方案,实现变桨执行器的故障估计。为满足高效实用的监控方案要求,将大于额定风速时的非线性机组动力学在额定工作点附近线性化,并设计一种基于Setpoint Overshoot分析法的PI控制器进行变桨,保留了传统PI控制器简单实用的优点,且优化了参数调节过程,从而实现更高效便捷的额定功率输出控制;将离散Kalman滤波与仿记忆数据预处理模式相结合,进行机组状态以及变桨执行器故障监测,提高了监测系统的处理性能。以面向海上应用的10MW风电机组为对象,进行仿真验证,表明所提方案对MW级大型离岸风电系统具有良好的故障监测性能。The dynamic model for large-scale offshore horizontal-axis wind turbines is built up considering the effects of tower vibrations. A monitoring scheme is developed for pitch actuator faults estimation, in the full load area where the wind speed is above the rated value. To meet the requirements of effective and applicable monitoring scheme, the nonlinear wind turbine dynamics are linearized near the operating point. Then, an improved PI pitch controller based on Setpoint Overshoot analysis method is designed. This controller not only reserves the merits of being simple and practical of traditional PI controllers, but also optimizes the parameter adjustment process, so that a more efficient and convenient power output control scheme is realized. Moreover, the discrete Kalman filter and memory-like data preprocessing method are blended for states monitoring and pitch actuator faults estimation, to improve the performance of monitoring system. The offshore 10 MW wind turbine is used as a benchmark for simulation validation, which demonstrates that the proposed scheme has satisfactory monitoring performances for MW level offshore wind turbines.

关 键 词：风力发电 离岸风机 SO-PI控制 仿记忆 故障监测 

分 类 号：TM315[电气工程—电机] TM76