风力发电机转速控制系统的建模与仿真
|
摘要
针对风力发电机非线性、随机性、扰动大等特点,设计基于带动量项的粒子群优化的支持向量机的风力发电机转速控制建模的新方法。利用支持向量机对小样本、高维度、非线性特性的映射能力,将风电系统的采样数据映射到高维的特征空间进行建模。支持向量机惩罚因子C和核参数σ的选择对建模效果影响较大,经验试凑的方法难以获得较好的参数,引入粒子群算法进行参数寻优。为了克服传统粒子群算法易陷入局部最优且收敛速度慢的缺陷,提出带动量项的改进粒子群算法寻优。以采集的风速、风力发电机转矩、桨距角作为输入信号,发电机转速数据作为输出信号,在Matlab环境中进行建模。实验结果表明,与传统算法相比,采用该方法的模型在准确性和收敛速度方面得到较大改善。
A new wind turbine speed control modeling method based on support vector machine optimized by particle swarm algorithm with momentum item was designed for the characteristics of nonlinearity,randomness and large disturbance of wind turbine. Since the support vector machine has the mapping ability of small- sample,high - dimensional and nonlinear characteristics,the sampling data of the wind power system is mapped to the high-dimensional feature space for modeling. The selection of penalty factor C and kernel parameter σ of support vector machine has a great influence on the modeling effect,and the optimal parameter is difficult to be obtained by the trial and error method,so the particle swarm optimization algorithm is introduced for parameter optimization. In order to overcome that the traditional particle swarm optimization algorithm is easy to fall into local optimum and has slow convergence speed,an improved particle swarm optimization algorithm with momentum term is proposed. The collected wind speed,wind turbine torque and pitch angle of wind turbine are taken as input signals,and the generator speed data is taken as output signal. The system is modeled in Matlab. The experimental results show that the model adopting the proposed method has a great improvement in accuracy and convergence speed than the traditional algorithm.
A new wind turbine speed control modeling method based on support vector machine optimized by particle swarm algorithm with momentum item was designed for the characteristics of nonlinearity,randomness and large disturbance of wind turbine. Since the support vector machine has the mapping ability of small- sample,high - dimensional and nonlinear characteristics,the sampling data of the wind power system is mapped to the high-dimensional feature space for modeling. The selection of penalty factor C and kernel parameter σ of support vector machine has a great influence on the modeling effect,and the optimal parameter is difficult to be obtained by the trial and error method,so the particle swarm optimization algorithm is introduced for parameter optimization. In order to overcome that the traditional particle swarm optimization algorithm is easy to fall into local optimum and has slow convergence speed,an improved particle swarm optimization algorithm with momentum term is proposed. The collected wind speed,wind turbine torque and pitch angle of wind turbine are taken as input signals,and the generator speed data is taken as output signal. The system is modeled in Matlab. The experimental results show that the model adopting the proposed method has a great improvement in accuracy and convergence speed than the traditional algorithm.
引文
[1] BAYATI I,BELLOLI M,BERNINI L,et al. On the functional design of the DTU10 MW wind turbine scale model of LIFES50+project[J]. Journal of physics,2016,753(5):1-6.
[2] BIANCHI F D,BATTISTA D H,MANTZ R J. Wind turbine control systems:principles,modelling and gain scheduling de-sign[M]. Berlin:Springer,2011.
[3]张涛,胡立锦,张新燕,等.风力发电机通用化建模研究[J].电力系统保护与控制,2013(13):1-6.ZHANG Tao,HU Lijin,ZHANG Xinyan,et al. Modeling of general application of wind turbines[J]. Power system protec-tion and control,2013(13):1-6.
[4]贺文,王小立,王伟,等.基于RTDS的风力机自适应转速启动控制的仿真研究[J].电力系统保护与控制,2017,45(22):152-157.HE Wen,WANG Xiaoli,WANG Wei,et al. Simulation study on adaptive speed start control of wind turbine based on RTDS[J]. Power system protection and control,2017,45(22):152-157.
[5]杨震宇,王青,魏新刚,等.基于RBF神经网络的风力发电机组系统辨识研究[J].机电工程,2017(6):639-642.YANG Zhenyu,WANG Qing,WEI Xingang,et al. Study on system identification of wind turbine based on RBF neural net-work[J]. Mechanical and electrical engineering,2017(6):639-642.
[6]叶飞.基于数据驱动的风力发电机组系统辨识研究[D].杭州:浙江大学,2014.YE Fei. Study on system identification of wind turbine based on data drive[D]. Hangzhou:Zhejiang University,2014.
[7]丁志勇,杨苹,杨曦,等.基于连续时间段聚类的支持向量机风电功率预测方法[J].电力系统自动化,2012,36(14):131-135.DING Zhiyong,YANG Ping,YANG Xi,et al. SVM wind power forecasting method based on continuous time clustering[J].Power system automation,2012,36(14):131-135.
[8]田强.基于模糊PID的风力发电机组变桨距控制研究[J].现代电子技术,2013,36(16):146-148.TIAN Qiang. Study on variable pitch control of wind turbine based on fuzzy PID[J]. Modern electronics technique,2013,36(16):146-148.
[9] RAI S,YANG S Y,TSAO T C. Wind turbine system identifi-cation and individual pitch control[C]//2017 American Control Conference. New York:IEEE,2017:1371-1376.
[10]魏毅立,韩素贤,时盛志.风力发电系统中组合风速的建模及仿真[J].可再生能源,2010,28(2):18-20.WEI Yili,HAN Suxian,SHI Shengzhi. Modeling and simula-tion of combined wind speed in wind power generation system[J]. Renewable energy,2010,28(2):18-20.
[11]田芳,周孝信,于之虹.基于支持向量机综合分类模型和关键样本集的电力系统暂态稳定评估[J].电力系统保护与控制,2017,45(22):1-8.TIAN Fang,ZHOU Xiaoxin,YU Zhihong. Evaluation of tran-sient stability of power systems based on the integrated classi-fication model of support vector machines and key sample sets[J]. Power system protection and control,2017,45(22):1-8.
[12] RANAEE V,EBRAHIMZADEH A,GHADERI R. Applica-tion of the PSO-SVM model for recognition of control chart patterns[J]. ISA transactions,2010,49(4):577.
[13]刘兴华,敬维,林威.GH Bladed和Matlab的交互软件设计及风力发电机的独立变桨控制器仿真研究[J].中国电机工程学报,2013,36(22):83-88.LIU Xinghua,JING Wei,LIN Wei. Interactive software de-sign of GH Bladed and Matlab and simulation study of an in-dependent variable oar controller for wind turbine[J]. Procee-dings of the CSEE,2013,36(22):83-88.
[14]李素,袁志高,王聪,等.群智能算法优化支持向量机参数综述[J].智能系统学报,2018,13(1):70-84.LI Su,YUAN Zhigao,WANG Cong,et al. Parameter review of swarm intelligence algorithm optimization support vector machines[J]. Journal of intelligent systems,2018,13(1):70-84.
[2] BIANCHI F D,BATTISTA D H,MANTZ R J. Wind turbine control systems:principles,modelling and gain scheduling de-sign[M]. Berlin:Springer,2011.
[3]张涛,胡立锦,张新燕,等.风力发电机通用化建模研究[J].电力系统保护与控制,2013(13):1-6.ZHANG Tao,HU Lijin,ZHANG Xinyan,et al. Modeling of general application of wind turbines[J]. Power system protec-tion and control,2013(13):1-6.
[4]贺文,王小立,王伟,等.基于RTDS的风力机自适应转速启动控制的仿真研究[J].电力系统保护与控制,2017,45(22):152-157.HE Wen,WANG Xiaoli,WANG Wei,et al. Simulation study on adaptive speed start control of wind turbine based on RTDS[J]. Power system protection and control,2017,45(22):152-157.
[5]杨震宇,王青,魏新刚,等.基于RBF神经网络的风力发电机组系统辨识研究[J].机电工程,2017(6):639-642.YANG Zhenyu,WANG Qing,WEI Xingang,et al. Study on system identification of wind turbine based on RBF neural net-work[J]. Mechanical and electrical engineering,2017(6):639-642.
[6]叶飞.基于数据驱动的风力发电机组系统辨识研究[D].杭州:浙江大学,2014.YE Fei. Study on system identification of wind turbine based on data drive[D]. Hangzhou:Zhejiang University,2014.
[7]丁志勇,杨苹,杨曦,等.基于连续时间段聚类的支持向量机风电功率预测方法[J].电力系统自动化,2012,36(14):131-135.DING Zhiyong,YANG Ping,YANG Xi,et al. SVM wind power forecasting method based on continuous time clustering[J].Power system automation,2012,36(14):131-135.
[8]田强.基于模糊PID的风力发电机组变桨距控制研究[J].现代电子技术,2013,36(16):146-148.TIAN Qiang. Study on variable pitch control of wind turbine based on fuzzy PID[J]. Modern electronics technique,2013,36(16):146-148.
[9] RAI S,YANG S Y,TSAO T C. Wind turbine system identifi-cation and individual pitch control[C]//2017 American Control Conference. New York:IEEE,2017:1371-1376.
[10]魏毅立,韩素贤,时盛志.风力发电系统中组合风速的建模及仿真[J].可再生能源,2010,28(2):18-20.WEI Yili,HAN Suxian,SHI Shengzhi. Modeling and simula-tion of combined wind speed in wind power generation system[J]. Renewable energy,2010,28(2):18-20.
[11]田芳,周孝信,于之虹.基于支持向量机综合分类模型和关键样本集的电力系统暂态稳定评估[J].电力系统保护与控制,2017,45(22):1-8.TIAN Fang,ZHOU Xiaoxin,YU Zhihong. Evaluation of tran-sient stability of power systems based on the integrated classi-fication model of support vector machines and key sample sets[J]. Power system protection and control,2017,45(22):1-8.
[12] RANAEE V,EBRAHIMZADEH A,GHADERI R. Applica-tion of the PSO-SVM model for recognition of control chart patterns[J]. ISA transactions,2010,49(4):577.
[13]刘兴华,敬维,林威.GH Bladed和Matlab的交互软件设计及风力发电机的独立变桨控制器仿真研究[J].中国电机工程学报,2013,36(22):83-88.LIU Xinghua,JING Wei,LIN Wei. Interactive software de-sign of GH Bladed and Matlab and simulation study of an in-dependent variable oar controller for wind turbine[J]. Procee-dings of the CSEE,2013,36(22):83-88.
[14]李素,袁志高,王聪,等.群智能算法优化支持向量机参数综述[J].智能系统学报,2018,13(1):70-84.LI Su,YUAN Zhigao,WANG Cong,et al. Parameter review of swarm intelligence algorithm optimization support vector machines[J]. Journal of intelligent systems,2018,13(1):70-84.