一种基于风速估计和NN/RISE的风电最大功率点跟踪方法  

A wind power MPPT method based on wind speed estimation and NN/RISE

作　　者：龚礼 罗浩 张道源 王国忠 焦绪国 GONG Li;LUO Hao;ZHANG Daoyuan;WANG Guozhong;JIAO Xuguo(School of Information and Control Engineering,Qingdao University of Technology,Qingdao 266525,China)

机构地区：[1]青岛理工大学信息与控制工程学院,山东青岛266525

出　　处：《广西大学学报(自然科学版)》2025年第1期105-118,共14页Journal of Guangxi University(Natural Science Edition)

基　　金：国家自然科学基金项目(62203249);山东省自然科学基金项目(ZR2021QF115)。

摘　　要：为了在有效风速未知、系统内存在未知动态以及有界噪声的情况下,实现风电机组最大功率点跟踪(MPPT),提出了一种基于风速估计和NN/RISE的MPPT控制方法。首先,利用卡尔曼牛顿算法对未知风速进行在线估计,滤除实际风速中的高频噪声,得到平滑、易于追踪的参考风轮转速,其次,基于NN/RISE原理设计转矩控制器,利用神经网络(NN)来学习风电机组中包含气动转矩的未知动态,利用积分器来抑制系统中符号函数对系统所造成的抖震现象。稳定性证明表明所提方法具有渐近稳定的转速跟踪性能。仿真结果表明,所提方法在提高转速跟踪精度与风能捕获效率的同时,也具有更平稳的转矩和功率输出。In order to achieve maximum power point tracking(MPPT)for wind turbines under conditions of unknown effective wind speed,unknown dynamics within the system,and bounded noise,a MPPT control method based on wind speed estimation and NN/RISE was proposed.Firstly,Kalman-Newton algorithm was used to estimate the unknown wind speed online,filter out high-frequency noise in the actual wind speed,and obtain a smooth and easily tracked reference wind turbine speed.Secondly,a torque controller was designed based on the NN/RISE principle,using neural networks(NN)to learn the unknown dynamics of aerodynamic torque in wind turbines,and using an integrator to suppress the chattering phenomenon caused by the sign function in the system.The stability proof shows that the proposed method has asymptotically stable speed tracking performance.Simulation results show that the proposed method can improve the speed tracking accuracy and wind energy capture efficiency,and provide a smoother torque and power output.

关 键 词：风电机组 最大功率点跟踪 风速估计 NN/RISE 转矩控制 

分 类 号：TM614[电气工程—电力系统及自动化] TP13[自动化与计算机技术—控制理论与控制工程]