摘要
随着风力发电在新能源发电中比例的逐渐提高,以及风电机组单机容量的增加,风电经串联补偿线路输出引起的次同步振荡问题也变得突出。为此,本文通过对轴系扭转振动特性进行分析,以期为解决次同步振荡问题提供依据。基于转轴、行星齿轮系统和平行齿轮副的扭转振动模型和运动微分方程,推导风机传动系统各质量单元间状态量的传递关系,采用Riccati传递矩阵法计算轴系的扭振固有特性,最后以某1.5 MW双馈风电机组为例,将采用此方法得到的计算结果与传统计算结果进行对比,并分析齿轮啮合刚度变化对主要固有频率的影响。结果表明:采用考虑齿轮啮合作用的风电机组扭转转动模型计算得到的扭转振动特性与采用简单集中质量模型的计算结果相比更加精确,这将为揭示次同步振荡发生的机理提供理论基础。
With the gradual increase in the proportion of wind power generation in new energy generation,and the increase in the capacity of single wind turbine unit,the problem of subsynchronous oscillation caused by wind power output through series compensation lines has also become prominent.The torsional vibration characteristics of the shafting is analyzed in this paper to provide basis for solving the subsynchronous oscillation problem.On the basis of the torsional vibration model and motion differential equation of the shaft,planetary gear system and parallel gears,the transitive relation of state variables among the mass units of wind turbine transmission system is deduced.Moreover,the Riccati transfer matrix method is adopted to calculate the natural torsional vibration characteristics.Finally,a 1.5 MW doubly-fed wind turbine is taken as an example for calculating the natural frequencies of torsional vibration by the method proposed,and the result is compared with that calculated by the conventional method.The influence of gear meshing stiffness on major natural frequency is also analyzed.It shows that,the torsional vibration characteristics calculated by the model considering gear meshing stiffness are more accurate than that by the simple lumped mass model.This provides theoretical basis for revealing the mechanism of subsynchronous oscillation.
作者
赵鹏程
王剑钊
ZHAO Pengcheng;WANG Jianzhao(Clean Energy Research Institute Co.,Ltd.,China Huaneng Group Co.,Ltd.,Beijing 102209,China)
出处
《热力发电》
CAS
北大核心
2020年第7期35-40,共6页
Thermal Power Generation
基金
中国华能集团有限公司总部科技项目(HNKJ16-H25)。
关键词
风电机组
轴系
扭振
行星轮
状态量
传递矩阵法
固有频率
wind turbine
shafting system
torsional vibration
planetary gear
state variable
transfer matrix method
natural frequency