
浏览全部资源
扫码关注微信
哈尔滨工业大学(深圳)机器人与先进制造学院,广东省深圳市 518055
翁浩文(2001—),男,硕士研究生,主要研究方向:电磁暂态实时仿真。E-mail:22s153261@stu.hit.edu.cn
王灿(1984—),男,通信作者,博士,副教授,IEEE Member,主要研究方向:柔性直流输电、交直流混合微电网、新能源汽车驱动电力电子系统。E-mail:can.wang@hit.edu.cn
潘学伟(1988—),男,博士,特聘研究员,主要研究方向:智能混合型微电网、可再生能源转换/分布式发电系统等。E-mail:davidpeterpan@163.com
收稿日期:2025-01-14,
修回日期:2025-04-06,
网络出版日期:2025-06-20,
纸质出版日期:2025-09-10
移动端阅览
翁浩文,王灿,潘学伟.基于联结神经网络的电力电子开关暂态模型及实时仿真方法[J].电力系统自动化,2025,49(17):198-207.
WENG Haowen,WANG Can,PAN Xuewei.Transient Model and Real-time Simulation Method for Power Electronic Switches Based on Connectionist Neural Networks[J].Automation of Electric Power Systems,2025,49(17):198-207.
翁浩文,王灿,潘学伟.基于联结神经网络的电力电子开关暂态模型及实时仿真方法[J].电力系统自动化,2025,49(17):198-207. DOI: 10.7500/AEPS20250114004.
WENG Haowen,WANG Can,PAN Xuewei.Transient Model and Real-time Simulation Method for Power Electronic Switches Based on Connectionist Neural Networks[J].Automation of Electric Power Systems,2025,49(17):198-207. DOI: 10.7500/AEPS20250114004.
随着电力电子技术在新能源和电动汽车领域的广泛应用,其动态特性对系统性能的影响日益显著,传统基于理想开关模型的实时仿真技术难以满足高精度需求。为此,文中提出一种基于联结神经网络的开关暂态建模与仿真方法,结合系统级与暂态模型的协同建模策略,通过引入时间节点联结机制增强模型的时间序列预测能力。系统级采用伴随离散电路模型与节点分析法进行电路网络求解,暂态模型利用数据驱动方式构建联结神经网络模型,以精确描述开关器件暂态过程。在自主搭建的现场可编程门阵列(FPGA)硬件平台上实现了实时仿真系统,通过双向升降压变换器和三相逆变器的实时仿真算例进行验证,实验分别采用不同的开关器件。通过与离线仿真结果对比表明,该暂态模型和实时仿真方法在多拓扑和多工况下均具有较高的精度和稳定性。
With the wide application of power electronic technologies in the fields of renewable energy and electric vehicles
their dynamic characteristics increasingly influence the system performance. Traditional real-time simulation technologies based on ideal switch models are difficult to meet high-precision demand. To this end
a switch transient modeling and simulation method based on connectionist neural networks is proposed. A co-modeling strategy combining both system-level and transient models is employed to enhance the time-series prediction capability of the model through the introduction of time-node coupling mechanism. At the system level
the circuit network is solved using the associated discrete circuit model and nodal analysis method
and the transient model utilizes a data-driven approach to construct a connectionist neural network model to accurately describe the transient process of the switching device. The real-time simulation system is implemented on a field programmable gate array (FPGA) hardware platform
and verified through real-time simulation cases of bidirectional buck-boost converters and three-phase inverters
in which various types of switching devices are used. The comparison with offline simulation results shows that the transient model and the real-time simulation method have high accuracy and stability under multi-topology and multi-case operating conditions.
刘文龙 , 吕志鹏 , 刘海涛 . 电力电子化配电台区形态发展以及运行控制技术综述 [J]. 中国电机工程学报 , 2023 , 43 ( 13 ): 4899 - 4922 .
LIU Wenlong , LYU Zhipeng , LIU Haitao . An overview of morphological development and operation control technology of power electronics dominated distribution area [J]. Proceedings of the CSEE , 2023 , 43 ( 13 ): 4899 - 4922 .
周斌 , 汪光森 , 李卫超 , 等 . 基于FPGA的电力电子系统电磁暂态实时仿真通用解算器 [J]. 电工技术学报 , 2023 , 38 ( 14 ): 3862 - 3874 .
ZHOU Bin , WANG Guangsen , LI Weichao , et al . An FPGA-based general solver for electromagnetic transient real-time simulation of power electronic systems [J]. Transactions of China Electrotechnical Society , 2023 , 38 ( 14 ): 3862 - 3874 .
赵浩然 , 孟铃涵 , 江艺宝 , 等 . 面向新型电力系统的实时仿真平台综述与展望 [J]. 高电压技术 , 2024 , 50 ( 10 ): 4611 - 4626 .
ZHAO Haoran , MENG Linghan , JIANG Yibao , et al . Review and prospects of real-time simulation platforms for new power systems [J]. High Voltage Engineering , 2024 , 50 ( 10 ): 4611 - 4626 .
VEKIĆ M S , GRABIĆ S U , MAJSTOROVIĆ D P , et al . Ultralow latency HIL platform for rapid development of complex power electronics systems [J]. IEEE Transactions on Power Electronics , 2012 , 27 ( 11 ): 4436 - 4444 .
徐晋 , 汪可友 , 李国杰 , 等 . 基于响应匹配的电力电子换流器恒导纳建模 [J]. 中国电机工程学报 , 2019 , 39 ( 13 ): 3879 - 3889 .
XU Jin , WANG Keyou , LI Guojie , et al . Fixed-admittance modeling of power electronic converters using response-matching technique [J]. Proceedings of the CSEE , 2019 , 39 ( 13 ): 3879 - 3889 .
WANG K Y , XU J , LI G J , et al . A generalized associated discrete circuit model of power converters in real-time simulation [J]. IEEE Transactions on Power Electronics , 2019 , 34 ( 3 ): 2220 - 2233 .
黄华震 , 仝涵 , 王宁燕 , 等 . 考虑寄生振荡的IGBT分段暂态模型对电磁干扰预测的影响分析 [J]. 电工技术学报 , 2021 , 36 ( 12 ): 2434 - 2445 .
HUANG Huazhen , TONG Han , WANG Ningyan , et al . Analysis of the influence of IGBT segmented transient model considering parasitic oscillations on electromagnetic interference prediction [J]. Transactions of China Electrotechnical Society , 2021 , 36 ( 12 ): 2434 - 2445 .
苏杭 , 徐晋 , 汪可友 , 等 . 考虑变换器损耗特性的小步长实时仿真方法 [J]. 中国电机工程学报 , 2021 , 41 ( 5 ): 1840 - 1851 .
SU Hang , XU Jin , WANG Keyou , et al . Small step real-time simulation method considering converter loss characteristics [J]. Proceedings of the CSEE , 2021 , 41 ( 5 ): 1840 - 1851 .
WANG W T , SHEN Z X , DINAVAHI V . Physics-based device-level power electronic circuit hardware emulation on FPGA [J]. IEEE Transactions on Industrial Informatics , 2014 , 10 ( 4 ): 2166 - 2179 .
BAI H , LUO H , LIU C , et al . A device-level transient modeling approach for the FPGA-based real-time simulation of power converters [J]. IEEE Transactions on Power Electronics , 2020 , 35 ( 2 ): 1282 - 1292 .
IDKHAJINE L , MONMASSON E . Embedded fully FPGA-based real-time simulators for static power converters with power switch characteristics approximated by identification [J]. IEEE Transactions on Industrial Electronics , 2022 , 69 ( 9 ): 9624 - 9633 .
BAI H , LIU C , BREAZ E , et al . A review on the device-level real-time simulation of power electronic converters: motivations for improving performance [J]. IEEE Industrial Electronics Magazine , 2021 , 15 ( 1 ): 12 - 27 .
SHEN Z X , DINAVAHI V . Real-time device-level transient electrothermal model for modular multilevel converter on FPGA [J]. IEEE Transactions on Power Electronics , 2016 , 31 ( 9 ): 6155 - 6168 .
HARI N , CHATTERJEE S , IYER A . Gallium nitride power device modeling using deep feed forward neural networks [C]// 2018 1st Workshop on Wide Bandgap Power Devices and Applications in Asia (WiPDA Asia) , May 16-18 , 2018 , <math id="M23"><mi mathvariant="normal">X</mi><mi mathvariant="normal">i</mi><mi mathvariant="normal">’</mi><mi mathvariant="normal">a</mi><mi mathvariant="normal">n</mi></math> https://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=88868459&type= 2.28600001 https://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=88868460&type= 6.77333355 , China : 164 - 168 .
KRISHNAMOORTHY H S , NARAYANAN AAYER T . Machine learning based modeling of power electronic converters [C]// 2019 IEEE Energy Conversion Congress and Exposition (ECCE) , September 29-October 3 , 2019 , Baltimore, USA : 666 - 672 .
LIANG T , HUANG Z , DINAVAHI V . Adaptive real-time hybrid neural network-based device-level modeling for DC traction HIL application [J]. IEEE Access , 2020 , 8 : 69543 - 69556 .
LI Q , BAI H , BREAZ E , et al . ANN-aided data-driven IGBT switching transient modeling approach for FPGA-based real-time simulation of power converters [J]. IEEE Transactions on Transportation Electrification , 2023 , 9 ( 1 ): 1166 - 1177 .
王钦盛 , 王灿 , 潘学伟 , 等 . 基于FPGA的电力电子恒导纳开关模型修正算法及实时仿真架构 [J]. 电力系统自动化 , 2024 , 48 ( 1 ): 150 - 159 .
WANG Qinsheng , WANG Can , PAN Xuewei , et al . Fixed-admittance switch model correction algorithm and real-time simulation architecture of power electronics based on field programmable gate array [J]. Automation of Electric Power Systems , 2024 , 48 ( 1 ): 150 - 159 .
Onsemi . Physically based, scalable SPICE modeling methodologies for modern power electronic devices: technical documentation [EB/OL]. [ 2024-08-01 ]. https://www.onsemi.com/pub/Collateral/TND6260-D.PDF https://www.onsemi.com/pub/Collateral/TND6260-D.PDF .
BAI H , LIU C , RATHORE A K , et al . An FPGA-based IGBT behavioral model with high transient resolution for real-time simulation of power electronic circuits [J]. IEEE Transactions on Industrial Electronics , 2019 , 66 ( 8 ): 6581 - 6591 .
ZAMANLOOY B , MIRHASSANI M . Efficient VLSI implementation of neural networks with hyperbolic tangent activation function [J]. IEEE Transactions on Very Large Scale Integration (VLSI) Systems , 2014 , 22 ( 1 ): 39 - 48 .
0
浏览量
0
下载量
0
CSCD
关联资源
相关文章
相关作者
相关机构
京公网安备11010802024621