摘要
随着风电机组单机容量的不断增大,风电机组的关键部件承受的载荷也越来越大,对风电机组可靠性的要求也越来越高,因此,要求风电机组控制策略与技术,既能实现功率优化控制,又能实现降载控制。研究基于模型预测控制(Model Predictive Control,MPC)理论,设计了一种基于风电机组多控制目标的运行区间划分方法的风电机组复合模型预测控制(Multi Model Predictive Control,Multi MPC)控制器。首先建立基于Matlab和TUV GL bladed的联合实时仿真平台,将MPC控制器与传统PI控制器进行对比分析,并以DUV GL Bladed软件中2 MW双馈式风电机组非线性模型作为研究对象,对Multi MPC控制器、MPC控制器和传统的比例积分微分(Proportional Integral Differential,PID)控制器进行了降载控制仿真分析。研究结果表明,MultiMPC控制器能够减小风电机组转速波动幅度,抑制转速超调量,降低传动链的载荷;能够抑制桨距角的波动幅度和变化速率,降低变桨距机构的运行载荷,提高机组运行可靠性。
The proportion of wind power in the global total power supply is also increasing year by year,and the single unit capacity of wind turbine is also increasing.At the same time,the load of the key components of wind turbine is also more and more large,and the reliability requirements of the structure and control technology are also more and more high,so the impact on the stability of the access system is more and more difficult to be ignored.Therefore,the control strategy and technology of wind turbine control system need not only the power optimization control ability,but also the load reduction control ability.The current research status of wind turbine load reduction is investigated,and the research status of model predictive control(MPC)technology is analyzed.Firstly,the wind turbine system,pitch system and drive chain system of wind turbine are modeled and analyzed,which lays the foundation for the further research of wind turbine control technology.Then,the model predictive control technology is studied deeply.Model predictive control is a kind of control algorithm to deal with nonlinear large time-delay system,which is essentially to solve an open-loop optimal control problem.In order to carry out the simulation comparative study,the variable pitch proportional integral controller of wind turbine is designed;based on the theory of model predictive control technology,the model predictive control controller of wind turbine and the multi model predictive control controller of wind turbine are designed.In the design of the model predictive control pitch controller of wind turbine,the cumulative error between the predicted output of the controlled actuator and the expected target is used as the objective optimization function;in the design of the multi model predictive control pitch controller of wind turbine,five continuous and non overlapping wind speed intervals are set,and five model predictions are designed for five different wind speed intervals At the same time,a division method of multi control objectives operation interval of wind turbine is designed to ensure that the multi model predictive control controller of wind turbine can continuously output control signals.Finally,in order to verify the load reduction ability of multi model predictive control controller for wind turbine,a real-time simulation platform based on MATLAB and TUV GL bladed is designed and written in C++.After verifying the advantages of model predictive control of wind turbine compared with traditional proportional integral controller,the nonlinear model of 2 MW doubly fed wind turbine in GL bladed software is taken as the research object,and the multi model predictive control controller,model predictive control controller and traditional proportional integral controller of wind turbine are simulated and analyzed.The results show that,compared with the traditional proportional integral controller and model predictive controller,model predictive controller can effectively reduce the fluctuation amplitude of wind turbine speed,restrain the overshoot of speed,reduce the load of the transmission chain of wind turbine,restrain the fluctuation amplitude and change rate of the pitch angle of wind turbine,and reduce the operation of the pitch mechanism of wind turbine Load.It can reduce the operation cost to a certain extent and improve the overall life of wind turbine.
作者
田德
陈忠雷
邓英
Tian De;Chen Zhonglei;Deng Ying(State Key Laboratory for Alternate Electrical Power System with Renewable Energy Sources,North China Electric Power University,Beijing 102206,China)
出处
《农业工程学报》
EI
CAS
CSCD
北大核心
2020年第21期65-70,共6页
Transactions of the Chinese Society of Agricultural Engineering
基金
国家863计划项目智能电网关键技术研发,风电场、光伏电站集群控制系统研究与开发子课题(2011AA05A104)。
关键词
模型
控制算法
风电机组
降载控制器
models
control algorithm
wind turbine
load shedding controller
作者简介
田德,博士,教授,主要从事风力发电系统理论与技术研究。Email:tdncepu@163.com。
