基于GRU-IPSO算法的双馈风机控制参数辨识

English Version Reader LoginJournal Login

您当前的位置：

首页 >

文章列表页 >

基于GRU-IPSO算法的双馈风机控制参数辨识

更新时间：2025-10-29

- 基于GRU-IPSO算法的双馈风机控制参数辨识
- Issue 4, Pages: 70-80(2025)
- 作者机构：
  
  1. 三峡大学电气与新能源学院
  2. 华北电力大学电气与电子工程学院
  3. 国网宁夏电力有限公司电力科学研究院
  4. 国网甘肃省电力公司电力科学研究院
- 作者简介：
- 基金信息：
- DOI：
  CLC： TP18;TM315
- Published：2025
- 稿件说明：
移动端阅览
[J]. 2025, (4): 70-80.
DOI：

[J]. 2025, (4): 70-80. DOI：

摘要

针对特定风速下双馈风机(double fed induction generator

DFIG)控制参数辨识结果难以精确刻画随机风速下输出特性的问题，提出一种基于门控循环单元(gated recurrent unit

GRU)-改进粒子群优化(improved particle swarm optimization

IPSO)算法的双馈风机控制参数辨识方法。首先，将双馈风机运行状态分为MPPT、恒速和恒功率三种，并利用斯皮尔曼秩相关系数法分析控制参数与待选观测量的相关性，筛选出交/直轴电流和有功/无功功率作为观测量；然后，在三种运行状态下训练GRU神经网络，在不设寻优范围情况下初步获得控制参数值，以简化辨识流程，并利用IPSO算法对控制参数进一步寻优，以提高辨识精度；最后，根据各参数的相对误差提取三种状态辨识结果中的最优值作为最终控制参数，结合测试案例对GRU-IPSO控制参数辨识方法的有效性和可行性进行验证。测试结果表明：GRU-IPSO辨识算法的加权平均总偏差约为3%

相比于GRU辨识算法，GRU-IPSO辨识算法可将误差缩小5倍。

Abstract

Aiming at the problem that it is difficult to accurately describe the output characteristics of doubly fed induction generator(DFIG) control parameter identification results at specific wind speed under random wind speed

we propose a control parameter identification method of DFIG based on gated current unit(GRU)-improved particle swarm optimization(IPSO) algorithm. Firstly

the operation status of DFIG is divided into three kinds

namely

MPPT

constant speed

and constant power status. We use the spearman rank correlation coefficient method to analyze the correlation between control parameters and optional observations

and select dq-axis current and active/reactive power as the observations. Secondly

we use the GRU neural network in three operating states to obtain initial control parameter without setting an optimization range to simplify identification process

and use the IPSO algorithm to further optimize control parameters to improve identification accuracy. Finally

according to the relative error of each parameter

we extract the optimal values of the three operating states as the final control parameter

and verify the effectiveness and feasibility of the GRU-IPSO control parameter identification method with test case. The test results show that the weighted average total deviation of GRU-IPSO identification algorithm is approximately 3%. Compared to GRU identification algorithm

the GRU-IPSO identification algorithm can reduce the error by five times.

关键词

Keywords

references

Du Wenjuan,Dong Wenkai,Wang Haifeng,Cao Jun.Dynamic Aggregation of Same Wind Turbine Generators in Parallel Connection for Studying Oscillation Stability of a Wind Farm[J].IEEE Transactions on Power Systems,2019.

Jin Yuqing,Lu Chongjiang,Ju Ping,Rehtanz Christian,Wu Feng,Pan Xueping.Probabilistic Preassessment Method of Parameter Identification Accuracy with an Application to Identifying the Drive Train Parameters of DFIG[J].IEEE Transactions on Power Systems,2019.

Ke Jia,Chenjie Gu,Lun Li,Zhengwen Xuan,Tianshu Bi,David Thomas.Sparse voltage amplitude measurement based fault location in large-scale photovoltaic power plants[J].Applied Energy,2018.

薛飞,李宏强,李旭涛,徐恒山.基于LSTM神经网络的双馈风机控制参数辨识方法[J].中国电力,2023(06).

徐岩,向益锋,马天祥.基于粒子群算法优化参数的VMD-GRU短期电力负荷预测模型[J].华北电力大学学报(自然科学版),2023(01).

李立,郑天悦,黄世楼,邓俊,王彤,夏楠.基于扩展卡尔曼滤波的DFIG变流器控制系统参数辨识方法[J].电网与清洁能源,2022(12).

王凇瑶,张智晟.基于量子加权多层级GRU神经网络的综合能源系统多元负荷短期预测[J].电力系统保护与控制,2022(23).

姜惠兰,王绍辉,李希钰,肖瑞.考虑动态电压区间无功支撑的双馈风机连锁故障穿越控制策略[J].高电压技术,2022(01).

许饶琪,彭晓涛,秦世耀,陈人杰,王瑞明,王靖然.基于M序列的双馈风机变流器参数辨识方法研究[J].电网技术,2022(02).

李辉,吴优,谢翔杰,柴兆森,龚立娇,鲁敏.基于改进PSO的双馈风电机组传动链参数辨识[J].太阳能学报,2021(12).

郭强,王鹤,聂永辉,高磊.考虑恢复暂态过程的直驱发电系统低电压穿越模型参数解耦辨识方法[J].高电压技术,2021(10).

徐岩,靳伟佳,朱晓荣.基于遗传粒子群算法的光伏并网逆变器参数辨识[J].太阳能学报,2021(07).

郭强,孙华东,高磊,息梦,聂永辉,宋瑞华.计及风电场随机特性的SVG模型参数智能辨识方法研究[J].中国电机工程学报,2020(24).

秦继朔,贾科,孔繁哲,杨彬,武文强,毕天姝.基于寻优算法的永磁风机并网逆变器故障穿越控制参数分步辨识[J].中国电机工程学报,2021(S1).

徐岩,张建浩.基于AEPSO-BPNN的光伏阵列多场景参数辨识[J].智慧电力,2020(10).

李建文,吴滨源,李永刚,李戎,严丽,孙伟.分析谐波劣化机理的逆变器灰箱宽频域矩阵模型及求解方法[J].电力系统自动化,2020(18).

夏天华,马骏超,黄弘扬,彭琰,肖修林,陈皓,郭瑞鹏.基于RTDS硬件在环测试的SVG控制器参数辨识[J].电力系统保护与控制,2020(13).

孟建辉,赵彭辉,王毅,张赟.基于虚拟惯性控制的多光储单元协同运行研究[J].电网技术,2021(05).

贾科,杨哲,魏超,郑黎明,李彦宾,毕天姝.基于斯皮尔曼等级相关系数的新能源送出线路纵联保护[J].电力系统自动化,2020(15).

潘学萍,鞠平,温荣超,吴峰,金宇清.解耦辨识双馈风电机组转子侧控制器参数的频域方法[J].电力系统自动化,2015(20).

潘学萍,温荣超,鞠平,金宇清,张彦涛.双馈风电机组网侧控制器参数辨识的频域方法[J].电网技术,2015(03).

Views

440

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Network security situation forecast based on improved general regression neural network

Research on the dynamic equivalence of synchronous generator based on WAMS

A Method for Model Reference Adaptive Identification of Continuous System

Temperature monitoring for asynchronous motors based on parameter identification technique

Online identification method of rotor resistance for slip vector-controlledinduction motors

Related Author

SHEN Hong-yan2

WANG Yu-fei1

HE Zi-dong1

LIU Guo-ping3

ZHANG Ning2

MI Zeng-qiang1

张国梁

张兆亮

Related Institution

College of Information Science & Technology,Agricultural University of Hebei

School of Control and Computer Engineering,North China Electric Power University

河北石家庄050041

河北石家庄050071

华北电力大学电气与电子工程学院河北保定071003

AI问答

⁰