风力发电系统容错控制策略研究

发布时间：2018-01-08 09:15

  本文关键词：风力发电系统容错控制策略研究　出处：《江南大学》2016年硕士论文　论文类型：学位论文

  更多相关文章： 风力发电系统 最大风能捕获 滑模控制器 传感器故障 T-S模糊滑模观测器 执行器故障 故障重构 传感器硬件冗余 自适应技术 容错控制

【摘要】：风能作为一种重要的自然能源,本身具有清洁环保、能量巨大、可持续供应的特点,相比自然界的煤炭、石油、天然气等其资源优势更加明显,目前在缓解世界能源危机中占有重要地位。风力发电设备通常建在高山或远离海岸的偏远地方,气候变化不可预测,在这样高度恶劣、复杂的工作环境中,传感器、执行器故障发生频繁,再加上风机本身具有非线性、多变量、强耦合等特点,对系统控制将会更加复杂。因此对风力发电系统进行实时故障诊断,实施有效的容错控制是确保风力发电系统可靠运行的重要手段。论文对控制系统现有的故障诊断技术、容错控制策略以及近年来风力发电系统的容错控制方法进行了详细调研,从风力机的空气动力学效应出发介绍了风力机的最大风能利用效率及各子系统的数学模型,讨论了系统在三种不同风速区域内的运行状态。通过对性能参数叶尖速比、风能利用系数最优值的计算及二者曲线图的分析阐述了最大风能捕获控制原理,并进行了仿真分析。针对风力发电系统的非线性、变量之间强耦合特点,利用结构简单、逼近能力强的T-S模糊算法,建立系统全局T-S模糊模型。对滑模控制理论分析给出等效控制量的求解方法并证明滑动模态对外界不确定因素的鲁棒性。选取合适滑模面,利用系统的输出信号作为控制器的输入,基于LMI方法设计模糊滑模控制器用于风力发电系统闭环反馈控制,确保系统性能参数叶尖速比和风能利用系数维持在最优值附近,实现部分负荷区的风能最大捕获。考虑滑模观测器在对非线性系统进行故障诊断时,能够保持对外界扰动不敏感,具有强的鲁棒性,论文将T-S模糊算法与滑模观测器理论相结合,设计模糊T-S系统滑模观测器对传感器故障进行重构。然后对传感器输出信号进行校正,以校正后的传感器输出信号代替控制器输入,实现风力发电系统主动容错控制的目的。针对风力发电系统中执行器故障与传感器故障并存情形,利用传感器硬件冗余技术结合状态观测器,建立残差逻辑判断表,实现多故障检测;其次通过引入一个简单的滤波器,将传感器故障转化为执行器故障,建立一个由原有的执行器故障和传感器故障组成的虚拟执行器故障,通过对虚拟执行器故障的重构来实现两种故障同时重构;将滑模算法与自适应技术结合用于风力发电系统的执行器故障容错控制中,保证故障系统能够准确跟踪所设计的期望状态和输出轨迹,达到容错控制目的。
[Abstract]:Wind energy is a kind of important natural energy, has tremendous energy, clean and environmental protection, sustainable supply characteristics, compared with the nature of coal, petroleum, natural gas and other resources more obvious advantages, currently occupies an important position in alleviating the energy crisis in the world. Wind power equipment are often built in remote areas or mountains away from the coast. Climate change is unpredictable, in such a highly complex work environment is bad, sensor, and the actuator faults occur frequently, and the wind machine itself has a nonlinear, multi variable, strong coupling characteristics, the control system will be more complicated. So the real-time fault diagnosis of wind power generation system, the implementation of fault tolerant control is effective an important means to ensure the reliable operation of wind power system. The fault diagnosis technology of the existing control system, fault-tolerant control strategy and system in recent years wind power capacity The wrong control method to make a detailed investigation of the mathematical model describes the maximum wind energy utilization efficiency and each subsystem from the aerodynamic effect of wind turbine, discusses the system in three different wind speed within the area of operation. Through the tip of the performance parameters calculation speed, wind energy utilization coefficient of the optimal value and the two analysis of the graph illustrates the maximum wind power capture control principle, and simulation analysis. Aiming at the nonlinear wind power system, variable strong coupling characteristics, with simple structure, strong approximation ability of T-S fuzzy algorithm, establish the system of global T-S fuzzy model. The theoretical analysis gives the equivalent control method of sliding mode control and that sliding mode robustness of uncertain external factors. Select the appropriate sliding surface, using the output signal as the input of the controller, the design model based on LMI method Fuzzy sliding mode controller for closed-loop feedback control of wind power generation system, ensure the system performance parameters of the tip speed ratio and wind power coefficient remained in the vicinity of the optimal value, realize the maximum wind energy capture part load. Considering the sliding mode observer in the fault diagnosis of nonlinear systems, can keep the external disturbance is not sensitive, has strong robustness, the T-S algorithm and fuzzy sliding mode observer theory are combined to reconstruct the sensor fault T-S system design of fuzzy sliding mode observer. Then the output of the sensor signal is corrected by the sensor output signal after correction instead of the input of the controller, realize the wind power generation system. The purpose of the active fault-tolerant control for wind power system actuator fault and sensor fault coexist situation according to the state observer, using the sensor hardware redundancy technology, the establishment of residual logic table implementation Multi fault detection; secondly, by introducing a simple filter, will be transformed into actuator fault sensor fault, the establishment of a composed original actuator and sensor faults of virtual actuator fault, through the reconstruction of virtual actuator failures to achieve two kinds of fault and reconstruction; sliding mode algorithm and adaptive technique used for actuator fault-tolerant control of wind power generation system, ensure the accuracy of tracking the design expectation state and output trajectory can achieve fault system, fault tolerant control.

【学位授予单位】：江南大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：TM614
，

本文编号：1396505