基于神经网络的LCL型单相并网逆变器稳定性分析

doi:10.16183/j.cnki.jsjtu.2023.653

上海交通大学学报 ›› 2025, Vol. 59 ›› Issue (12): 1805-1814.doi: 10.16183/j.cnki.jsjtu.2023.653

• 新型电力系统与综合能源 • 上一篇下一篇

基于神经网络的LCL型单相并网逆变器稳定性分析

韩华玲¹, 贾一超¹, 马子涵²^,³(), 邓俊⁴, 黄萌²^,³

¹ 中国电力科学院研究院有限公司南京分院, 南京 210009
² 武汉大学电气与自动化学院, 武汉 430072
³ 综合能源电力装备及系统安全湖北省重点实验室, 武汉 430072
⁴ 国网陕西省电力有限公司电力科学研究院, 西安 710100

收稿日期:2023-12-29 修回日期:2024-03-10 接受日期:2024-03-20 出版日期:2025-12-28 发布日期:2025-12-30
通讯作者: 马子涵 E-mail:zihanma1108@whu.edu.cn
作者简介:韩华玲(1985—),助理工程师,从事电力系统接入及方法的研究.
基金资助:
国家电网有限公司科技项目(4000-202355078A-1-1-ZN)

Stability Analysis of LCL Grid-Connected Inverter Based on Neural Network

HAN Hualing¹, JIA Yichao¹, MA Zihan²^,³(), DENG Jun⁴, HUANG Meng²^,³

¹ Nanjing Branch, China Electric Power Research Institute Co., Ltd., Nanjing 210009, China
² School of Electrical Engineering and Automation, Wuhan University, Wuhan 430072, China
³ Hubei Key Laboratory of Power Equipment and System Security for Integrated Energy, Wuhan 430072, China
⁴ Electric Power Research Institute, State Grid Shaanxi Electric Power Co., Ltd., Xi’an 710100, China

Received:2023-12-29 Revised:2024-03-10 Accepted:2024-03-20 Online:2025-12-28 Published:2025-12-30
Contact: MA Zihan E-mail:zihanma1108@whu.edu.cn

1. 2023-653-附录.pdf(154KB)

摘要/Abstract

摘要：

LCL型并网逆变器的参数不确定性严重影响新能源发电的电能质量,因此需要对参数摄动情况下逆变器的稳定情况进行分析.针对上述问题,首先建立LCL型单相并网逆变器状态空间模型,提出基于参数模型的神经网络建模方法.通过参数特征分析,得到基于参数分布的训练数据集,在神经网络中对数据集进行训练,输出稳定性判别结果.最后,通过MATLAB/Simulink仿真与实验平台验证所提基于神经网络分析逆变器稳定性方法的有效性.

关键词: LCL型单相并网逆变器, 稳定性分析, 神经网络, 反向传播, 二分类

Abstract:

The parameter uncertainty of LCL type grid-connected inverter can significantly affect the power quality of renewable energy, making it essential to analyze the stability of the inverter under the parameter inception. To solve these problems, this paper establishes a state space model of LCL type single-phase grid-connected inverter and proposes a neural network modeling method based on the parameter model. Through the parameter characterization, a training dataset based on the parameter distribution is obtained. This dataset is then trained in the neural network to produce the stability discrimination results. Finally, the effectiveness of the proposed method is verified by MATLAB/Simulink simulation and the experimental platform.

Key words: LCL type single-phase grid-connected inverter, stability analysis, neural network, back propagation, binary classification

中图分类号:

TM464

韩华玲, 贾一超, 马子涵, 邓俊, 黄萌. 基于神经网络的LCL型单相并网逆变器稳定性分析[J]. 上海交通大学学报, 2025, 59(12): 1805-1814.

HAN Hualing, JIA Yichao, MA Zihan, DENG Jun, HUANG Meng. Stability Analysis of LCL Grid-Connected Inverter Based on Neural Network[J]. Journal of Shanghai Jiao Tong University, 2025, 59(12): 1805-1814.

图/表 11

图1

图2

表1

图3

表2

表3

图4

图5

表4

图6

图7

参考文献 36

[1]	朱城昊, 王晗, 孙国歧, 等. 一种并网逆变器直流电容容值辨识方法[J]. 上海交通大学学报, 2022, 56(6): 693-700. doi: 10.16183/j.cnki.jsjtu.2021.515
	ZHU Chenghao, WANG Han, SUN Guoqi, et al. An identification for DC-link capacitor capacitance of grid connected inverter[J]. Journal of Shanghai Jiao Tong University, 2022, 56(6): 693-700.
[2]	胡长斌, 程麟舒, 罗珊娜, 等. 计及多种不确定因素的逆变器在线补偿控制[J]. 控制工程, 2023(1): 1-12.
	HU Changbin, CHENG Linshu, LUO Shanna, et al. Online compensation control of inverters considering multiple uncertain factors[J]. Control Engineering, 2023(1): 1-12.
[3]	全少理, 于昊正, 马杰, 等. 交/直流混合配网多逆变器分布式协同抗扰控制[J]. 上海交通大学学报, 2025, 59(5): 605-616. doi: 10.16183/j.cnki.jsjtu.2023.492
	QUAN Shaoli, YU Haozheng, MA Jie, et al. Distributed cooperative disturbance-rejection control of hybrid AC/DC distribution grids with multiple inverters[J]. Journal of Shanghai Jiao Tong University, 2025, 59(5): 605-616.
[4]	贤燕华, 江明轩, 王世强. 参数不确定并网逆变器的H_∞控制算法[J]. 水电能源科学, 2019, 37(11): 185-188.
	XIAN Yanhua, JIANG Mingxuan, WANG Shiqiang. H_∞ of grid connected inverters with uncertain parameters control algorithm[J]. Hydroelectric Energy Science, 2019, 37(11): 185-188.
[5]	LU M H, AL-DURRA A, MUYEEN S M, et al. Benchmarking of stability and robustness against grid impedance variation for LCL-filtered grid-interfacing inverters[J]. IEEE Transactions on Power Electronics, 2018, 33(10): 9033-9046. doi: 10.1109/TPEL.2017.2784685 URL
[6]	LIU T, LIU J J, LIU Z, et al. A study of virtual resistor-based active damping alternatives for LCL resonance in grid-connected voltage source inverters[J]. IEEE Transactions on Power Electronics, 2020, 5(1): 247-262.
[7]	GUAN Y P, WANG Y, XIE Y X, et al. The dual-current control strategy of grid-connected inverter with LCL filter[J]. IEEE Transactions on Power Electronics, 2019, 34(6): 5940-5952. doi: 10.1109/TPEL.63 URL
[8]	慕昆, 何国锋. 基于虚拟电阻的并网逆变器谐振抑制措施的研究[J]. 可再生能源, 2016, 34(6): 815-820.
	MU Kun, HE Guofeng. Research on method of resonance suppression for grid-connected inverter based on virtual resistor[J]. Renewable Energy Resources, 2016, 34(6): 815-820.
[9]	刘吉宏, 刘鹏飞, 张树新, 等. LCL型光伏并网逆变器全局鲁棒滑模变结构双闭环控制[J]. 可再生能源, 2020, 38(11): 1495-1499.
	LIU Jihong, LIU Pengfei, ZHANG Shuxin, et al. Global robust sliding mode variable structure dual closed-loop control for grid-connected PV inverters with LCL filters[J]. Renewable Energy Resources, 2020, 38(11): 1495-1499.
[10]	HUSSAIN M N, MELATH G, AGARWAL V. An active damping technique for PI and predictive controllers of an interlinking converter in an islanded hybrid microgrid[J]. IEEE Transactions on Power Electronics, 2021, 36(5): 5521-5529. doi: 10.1109/TPEL.63 URL
[11]	MOHAMED I S, ROVETTA S, DO T D, et al. A neural-network-based model predictive control of three-phase inverter with an output-filter[J]. IEEE Access, 2019, 7: 124737-124749. doi: 10.1109/Access.6287639 URL
[12]	BASIT B A, REHMAN A, CHOI H H, et al. A robust iterative learning control technique to efficiently mitigate disturbances for three-phase standalone inverters[J]. IEEE Transactions on Industrial Electronics, 2022, 69(4): 3233-3244. doi: 10.1109/TIE.2021.3071695 URL
[13]	曹永锋, 武玉衡, 叶永强, 等. 基于微分前馈自抗扰的逆变器控制策略[J]. 电力系统自动化, 2019, 43(5): 136-142.
	CAO Yongfeng, WU Yuheng, YE Yongqiang, et al. Active disturbance rejection control strategy with differential feedforward for inverters[J]. Automation of Electric Power Systems, 2019, 43(5): 136-142.
[14]	杨林, 曾江, 马文杰, 等. 基于改进二阶线性自抗扰技术的微网逆变器电压控制[J]. 电力系统自动化, 2019, 43(4): 146-153.
	YANG Lin, ZENG Jiang, MA Wenjie, et al. Voltage control of microgrid inverter based on improved second-order linear active disturbance rejection control[J]. Automation of Electric Power Systems, 2019, 43(4): 146-153.
[15]	李志华, 曾江, 黄骏翅, 等. 基于线性自抗扰控制的微网逆变器时-频电压控制策略[J]. 电力系统自动化, 2020, 44(10): 145-154.
	LI Zhihua, ZENG Jiang, HUANG Junchi, et al. Time-frequency voltage control strategy of microgrid inverter based on linear active disturbance rejection control[J]. Automation of Electric Power Systems, 2020, 44(10): 145-154.
[16]	张雪妍, 付立军, 马凡, 等. 基于虚拟阻抗的逆变器输出电压动态性能优化[J]. 太阳能学报, 2020, 41(11): 38-45.
	ZHANG Xueyan, FU Lijun, MA Fan, et al. Dynamic performance optimization of inverter output voltage based on virtual impendance[J]. Acta Energiae Solaris Sinica, 2020, 41(11): 38-45.
[17]	曹文远, 韩民晓, 谢文强, 等. 基于扰动观测器的电压源型逆变器负载电流前馈控制及参数设计方法[J]. 电工技术学报, 2020, 35(4): 862-873.
	CAO Wenyuan, HAN Minxiao, XIE Wenqiang, et al. A disturbance-observer-based load current feedforward control and parameter design method for voltage-sourced inverter[J]. Transactions of China Electrotechnical Society, 2020, 35(4): 862-873.
[18]	MOHAMED S A Q, RAFAQ M S, CHOI H H, et al. A robust adaptive PI voltage controller to eliminate impact of disturbances and distorted model parameters for 3-phase CVCF inverters[J]. IEEE Transactions on Industrial Informatics, 2020, 16(4): 2168-2176. doi: 10.1109/TII.9424 URL
[19]	CHA H, VU T K, KIM J E. Design and control of proportional-resonant controller based photovoltaic power conditioning system[C]// 2009 IEEE Energy Conversion Congress and Exposition. San Jose, CA, USA:IEEE, 2009: 2198-2205.
[20]	赵清林, 郭小强, 邬伟扬. 单相逆变器并网控制技术研究[J]. 中国电机工程学报, 2007, 27(16): 60-64.
	ZHAO Qinglin, GUO Xiaoqiang, WU Weiyang. Research of control strategy for single-phase grid-connected inverter[J]. Proceeding of the CSEE, 2007, 27(16): 60-64.
[21]	程成, 谢少军, 谭玲娟, 等. 跟网型逆变器的非线性模型及稳定性分析方法[J]. 电力系统自动化, 2022, 46(6): 137-143.
	CHENG Cheng, XIE Shaojun, TAN Lingjuan, et al. Nonlinear model and stability analysis method for grid-following inverter[J]. Automation of Electric Power Systems, 2022, 46(6): 137-143.
[22]	陈新, 王赟程, 龚春英, 等. 采用阻抗分析方法的并网逆变器稳定性研究综述[J]. 中国电机工程学报, 2018, 38(7): 2082-2094.
	CHEN Xin, WANG Yuncheng, GONG Chunying, et al. Overview of stability research for grid-connected inverters based on impedance analysis method[J]. Proceedings of the CSEE, 2018, 38(7): 2082-2094.
[23]	洪芦诚, 徐佳裕, 唐润悦, 等. 三相LCL型逆变器序阻抗简化建模方法及并网稳定性分析[J]. 电力系统自动化, 2023, 47(7): 1-13.
	HONG Lucheng, XU Jiayu, TANG Runyue, et al. Simplified modeling method of sequence impedance and grid-connected stability analysis for three-phase LCL inverter[J]. Automation of Electric Power Systems, 2023, 47(7): 1-13.
[24]	QUAN H, SRINIVASAN D, KHOSRAVI A. Short-term load and wind power forecasting using neural network-based prediction intervals[J]. IEEE Transactions on Neural Networks and Learning Systems, 2014, 25(2): 303-315. doi: 10.1109/TNNLS.2013.2276053 pmid: 24807030
[25]	ZHENG C, WANG S R, LIU Y L, et al. A novel equivalent model of active distribution networks based on LSTM[J]. IEEE Transactions on Neural Networks and Learning Systems, 2019, 30(9): 2611-2624. doi: 10.1109/TNNLS.2018.2885219 pmid: 30605108
[26]	XU Y, ZANG R, ZHAO J H, et al. Assessing short-term voltage stability of electric power systems by a hierarchical intelligent system[J]. IEEE Transactions on Neural Networks and Learning Systems, 2016, 27(8): 1686-1696. doi: 10.1109/TNNLS.2015.2441706 pmid: 26441430
[27]	TIAN G Y, ZHOU Q, BIRARI R, et al. A hybrid-learning algorithm for online dynamic state estimation in multimachine power systems[J]. IEEE Transactions on Neural Networks and Learning Systems, 2020, 31(12): 5497-5508. doi: 10.1109/TNNLS.5962385 URL
[28]	LI S H, FAIRBANK M, JOHNSON C, et al. Artificial neural networks for control of a grid-connected rectifier/inverter under disturbance, dynamic and power converter switching conditions[J]. IEEE Transactions on Neural Networks and Learning Systems, 2014, 25(4): 738-750. doi: 10.1109/TNNLS.2013.2280906 pmid: 24807951
[29]	FU X G, LI S H, FAIRBANK M, et al. Training recurrent neural networks with the Levenberg-Marquardt algorithm for optimal control of a grid-connected converter[J]. IEEE Transactions on Neural Networks and Learning Systems, 2015, 26(9): 1900-1912. doi: 10.1109/TNNLS.2014.2361267 pmid: 25330496
[30]	LIAO Y C, WANG X F. Small-signal modeling of AC power electronic systems: Critical review and unified modeling[J]. IEEE Open Journal of Power Electronics, 2021, 2: 424-439. doi: 10.1109/OJPEL.2021.3104522 URL
[31]	LIAO Y C, LI Y F, CHEN M J, et al. Neural network design for impedance modeling of power electronic systems based on latent features[J]. IEEE Transactions on Neural Networks and Learning Systems, 2023, 35(5): 5968-5980. doi: 10.1109/TNNLS.2023.3235806 URL
[32]	杜焕伦, 李哲帆, 石琼林, 等. 基于BP神经网络与DP等效电路的锂离子电池热电耦合模型构建[J/OL]. 电源学报, 2023: 1-12. [2025-05-28]. http://kns.cnki.net/kcms/detail/12.1420.tm.20231219.1348.014.html.
	DU Huanlun, LI Zhefan, SHI Qionglin, et al. Construction of a thermoelectric coupling model for llithium-ion batteries based on BP neural network and DP equivalent circuit[J/OL]. Journal of Power Supply, 2023: 1-12. [2025-05-28]. http://kns.cnki.net/kcms/detail/12.1420.tm.20231219.1348.014.html.
[33]	YAN C, LI M X, LIU W. Transformer fault diagnosis based on BP-Adaboost and PNN series connection[J]. Mathematical Problems in Engineering, 2019, 2019(1): 1019845. doi: 10.1155/mpe.v2019.1 URL
[34]	HAGAN M T, MENHAJ M B. Training feedforward networks with the Marquardt algorithm[J]. IEEE Transactions on Neural Networks and Learning Systems, 1994, 5(6): 989-993.
[35]	ZHOU D, WANG H, BLAABJERG F. Reactive power impacts on LCL filter capacitor lifetime in grid-connected inverter[J]. IEEE Open Journal of Power Electronics, 2020, 1: 139-148. doi: 10.1109/OJPEL URL
[36]	KAUFHOLD E, MEYER J, MULLER S, et al. Probabilistic stability analysis for commercial low power inverters based on measured grid impedances[C]//2019 9th International Conference on Power and Energy Systems (ICPES). Perth, Australia:IEEE, 2019: 1-6.

C/μF	L₁/mH	L₂/mH	特征值实部	期望输出
11.903 61	1.1	0.44	83.803 14	1
1.058 43	1.1	0.44	-52.298 20	0
11.293 47	1.1	0.44	25.883 93	1
11	1.066 56	0.44	-7.099 73	0
11	0.979 05	0.44	-12.863 66	0
11	0.992 98	0.44	-12.863 66	0
11	1.1	0.393 33	-8.982 97	0
11	1.1	0.414 27	-7.097 11	0
11	1.1	0.406 35	-7.799 62	0

系统参数	标称值
系统容量, S_inv/W	3 000
U_dc/V	400
U_g/V	220
直流电容, C_dc/mF	1
L₂/mH	0.2
L₁/mH	1
C/μF	10
R_g/Ω	0.1
L_g/mH	0.2
K_c	0.1
K_p/K_r	0.11/11.0

固定参数值	变化参数 C/μF	神经网络二分类输出值	是否稳定	固定参数值	变化参数 L₁/mH	神经网络二分类输出值	是否稳定	固定参数值	变化参数 L₂/mH	神经网络二分类输出值	是否稳定
L₁=1.1 mH L₂=0.44 mH	8	0.059 6	是	C=11 μF L₂=0.44 mH	0.8	0.046 8	是	C=11 μF L₁=1.1 mH	0.2	0.048 4	是
	9	0.041 3	是		0.9	0.051 0	是		0.3	0.046 6	是
	10	0.043 8	是		1.0	0.049 9	是		0.4	0.044 0	是
	11	0.042 4	是		1.1	0.063 4	是		0.5	0.042 2	是
	12	0.962 0	否		1.2	0.955 4	否		0.6	0.962 0	否

固定参数	变化参数	神经网络二分类输出值	是否稳定
C=9.902 μF	L₂=0.505 mH	0.042 436	是
L₁=1.02 mH	L₂=0.473 mH	0.962 769	否
L₁=1.02 mH	C=8.0 μF	0.044 810	是
L₂=0.26 mH	C=10.0 μF	0.961 398	否

基于神经网络的LCL型单相并网逆变器稳定性分析

Stability Analysis of LCL Grid-Connected Inverter Based on Neural Network

RichHTML

PDF (PC)

补充材料

可视化

摘要/Abstract

引用本文

使用本文

图/表 11

参考文献 36

相关文章 15

编辑推荐

Metrics

本文评价

[1]	李诗杰, 刘泰序, 刘佳仑, 董智霖, 徐诚祺. 基于GRU-MPC的双全回转推进拖轮轨迹跟踪控制[J]. 上海交通大学学报, 2026, 60(3): 418-426.
[2]	梁佳铭, 余音, 胡祎乐. 基于时空神经网络的金属疲劳裂纹扩展预测[J]. 上海交通大学学报, 2026, 60(3): 511-521.
[3]	. DSNet：用于语音情感识别的带有中性校准的解耦孪生网络[J]. J Shanghai Jiaotong Univ Sci, 2026, 31(2): 248-257.
[4]	. 类间隙滞后非线性系统复合双通道干扰估计自适应控制器设计[J]. J Shanghai Jiaotong Univ Sci, 2026, 31(1): 106-116.
[5]	夏筱彦, 张宇, 胡锡坤, 钟平. 基于扩散模型的无人机遥感目标检测物理对抗攻击方法研究[J]. 空天防御, 2026, 9(1): 52-62.
[6]	李湘, 陈思远, 张俊, 柯德平, 高杰迈, 杨欢欢. 基于物理信息嵌入的非固定长度电力系统暂态稳定快速评估[J]. 上海交通大学学报, 2025, 59(7): 962-970.
[7]	王晓倩, 周羽生, 毛源军, 李彬, 周文晴, 苏盛. 基于神经网络分位数的分布式光伏发电功率异常识别方法[J]. 上海交通大学学报, 2025, 59(6): 836-844.
[8]	曾金灿, 何耿生, 李姚旺, 杜尔顺, 张宁, 朱浩骏. 基于卷积神经网络与轻量级梯度提升树组合模型的电力行业短期以电折碳方法[J]. 上海交通大学学报, 2025, 59(6): 746-757.
[9]	. 基于深度学习序列方法的多人姿态估计用来检测人体与关键点位置[J]. J Shanghai Jiaotong Univ Sci, 2025, 30(6): 1103-1113.
[10]	潘美琪, 贺兴. 基于零样本学习的风力机故障诊断方法[J]. 上海交通大学学报, 2025, 59(5): 561-568.
[11]	黄逸翔, 窦迅, 李林溪, 杨函煜, 于建成, 霍现旭. 基于全局灵敏度分析的综合能源设备响应价值量化方法[J]. 上海交通大学学报, 2025, 59(5): 569-579.
[12]	. 基于基尼不纯度结构优化物理引导神经网络的薄膜型声学超材料传声损失预测[J]. J Shanghai Jiaotong Univ Sci, 2025, 30(3): 613-624.
[13]	. 基于毫米波雷达的智能心率提取方法[J]. J Shanghai Jiaotong Univ Sci, 2025, 30(3): 493-498.
[14]	. 基于表面肌电信号的BP-LSTM混合模型肘部运动实时预测[J]. J Shanghai Jiaotong Univ Sci, 2025, 30(3): 455-462.
[15]	周添龙, 姚方競, 饶卫雄. 部分已知状态空间模型下的目标状态估计算法[J]. 空天防御, 2025, 8(3): 111-122.