摘要
针对风电机组的传动系统、发电机和桨距系统故障的有限时间重构和实际应用中气动转矩无法准确获取等问题。提出一种新的自适应非奇异终端滑模观测器。在观测器中引入自适应律,确保滑模观测器不受未知扰动的影响。设计的非奇异滑模面能有效解决常规滑模观测器的抖振问题,避免了抖振现象造成的故障误判和漏判等问题,提高了故障诊断效率。针对变桨系统故障,通过引入故障指示参数,将桨距执行机构的液压压降模型转化为加性故障。然后,利用两个级联滑模观测器对桨距系统进行观测,给出了有限时间状态的估计和故障重构。最后,仿真结果验证了风电机组状态的有限时间估计和执行器故障的重构,达到了风电机组故障诊断快速诊断的目的。证明所提方法的正确性、可行性。
The finite time reconstruction of the fault of the transmission system, generator and pitch system of the wind turbine and the accurate acquisition of the pneumatic torque in practical application are discussed. A new adaptive nonsingular terminal sliding mode observer is proposed. An adaptive law is introduced into the observer to ensure that the sliding mode observer is not affected by unknown disturbance. The proposed non-singular sliding mode surface can effectively solve the chattering problem of the conventional sliding mode observer, avoid the problems of fault misjudgment and missing judgment caused by chattering phenomenon, and improve the efficiency of fault diagnosis. Aiming at the fault of variable propeller system, the hydraulic pressure drop model of pitch actuator was transformed into additive fault by introducing fault indication parameters. Then, two cascade sliding mode observers are used to observe the pitch system, and the finite time state estimation and fault reconstruction are given. Finally, the simulation results verify the finite time estimation of the wind turbine state and the fault reconstruction of the actuator, and achieve the goal of fast fault diagnosis of the wind turbine. The correctness and feasibility of the proposed method are proved.
作者
王昊
郝万君
李泽
崔国增
吴宇
张正夫
孙志辉
WANG Hao;HAO Wanjun;LI Ze;CUI Guozeng;WU Yu;ZHANG Zhengfu;SUN Zhihui(Institute of Electronics and Information Engineering,Suzhou University of Science and Technology,Suzhou 215009)
出处
《机械工程学报》
EI
CAS
CSCD
北大核心
2021年第14期261-269,281,共10页
Journal of Mechanical Engineering
基金
国家自然科学基金(61703059)
中国博士后科学基金(2018M632621)资助项目。
关键词
风电机组
非奇异终端滑模观测器
自适应
级联滑模观测器
稳定性分析
故障模式
故障重构
wind turbine
nonsingular terminal sliding mode observer
adaptive
cascade sliding mode observer
stability analysis
fault mode
fault reconstruction