A Data-driven Model of Peak Current Control Boost Converter Based on Neural Network Method

doi:10.11985/2022.02.016

Abstract

Abstract:

Switching power supply is a kind of high frequency electric energy conversion device and is widely applied to many fields. However, how to model the converter quickly and accurately remains to be solved. A data-driven modeling method is proposed to analyze the power loss of a peak current controlled Boost converter by UC3842 control chip based on neural network(NN). In order to select the key input and output parameters for data-driven model, the operation mechanism of the Boost converter is studied. Then, data-driven modeling method based on neural network is introduced and applied to model the Boost converter. Finally, the input characteristic and the inner loss characteristic are analyzed by the proposed method. Compared with traditional mechanism modeling method, data-driven modeling can bypass the complex internal mechanism and the model established by using the mapping relationship between data has the advantages of high speed and high accuracy.

Key words: Data-driven model, neural network, peak current control Boost converter

CLC Number:

TM561

WANG Runnan, XIE Fan, ZHANG Bo. A Data-driven Model of Peak Current Control Boost Converter Based on Neural Network Method[J]. Journal of Electrical Engineering, 2022, 17(2): 142-150.

Figures/Tables 14

References 21

[1]	李志刚, 梅霜, 王少杰, 等. IGBT模块开关损耗计算方法综述[J]. 电子技术应用, 2016, 42(1):10-14,18.
	LI Zhigang, MEI Shuang, WANG Shaojie, et al. The review of IGBT module switching loss calculation method[J]. Application of Electronic Technique, 2016, 42(1):10-14,18.
[2]	张明元, 沈建清, 李卫超, 等. 一种快速IGBT损耗计算方法[J]. 船电技术, 2009, 29(1):33-36.
	ZHANG Mingyuan, SHEN Jianqing, LI Weichao, et al. A fast power loss for IGBT modules[J]. Marine Electrical Technology, 2009, 29(1):33-36.
[3]	马西奎, 李明, 戴栋, 等. 电力电子电路与系统中的复杂行为研究综述[J]. 电工技术学报, 2006, 21(12):1-11.
	MA Xikui, LI Ming, DAI Dong, et al. Review of research on complex behavior in power electronic circuits and systems[J]. Transactions of China Electrotechnical Society, 2006, 21(12):1-11.
[4]	PAULA W J de, TAVARES G H M, SOARES G M, et al. Switching losses prediction methods oriented to power MOSFETs—A review[J]. IET Power Electronics, 2020, 13(14):2960-2970. doi: 10.1049/iet-pel.2019.1003
[5]	SELHI H, CHRISTOPOULOS C. Generalized TLM switch model for power electronics applications[J]. IEEE Proceedings-Science Measurement and Technology, 1998, 145(3):101-104. doi: 10.1049/ip-smt:19981896
[6]	GOLE A M, KERI A, KWANKPA C, et al. Guidelines for modeling power electronics in electric power engineering applications[J]. IEEE Trans. Power Del., 1997, 12(1):505-514. doi: 10.1109/61.568278
[7]	WONG C. EMTP modeling of IGBT dynamic performance for power dissipation estimation[J]. IEEE Transactions on Industry Applications, 1997, 33(1):64-71. doi: 10.1109/28.567078
[8]	MUNK-NIELSEN S, TUNTELEA L N, JAEGER U. Simulation with ideal switch models combined with measured loss data provides a good estimate of power loss[C]// Conference Record of the 2000 IEEE Industry Applications Conference. 2000, 5:2915-2922.
[9]	AZUMA S, KIMATA M, SETO M, et al. Research on the power loss and junction temperature of power semiconductor devices for inverter[J]. Proceedings of the IEEE International Vehicle Electronics Conference, 1999, 1:183-187.
[10]	程龙, 陈权, 李国丽, 等. 基于寄生参数的功率MOSFET数学建模及损耗分析[J]. 华东理工大学学报, 2014, 40(6):740-745.
	CHENG Long, CHEN Quan, LI Guoli, et al. Mathematic modeling and loss analysis of power MOSFET based on parasitic parameters[J]. Journal of East China University of Science and Technology, 2014, 40(6):740-745.
[11]	耿学锋, 何赟泽, 李孟川, 等. IGBT多物理场建模技术与应用研究概述[J]. 中国电机工程学报, 2022, 42(1):271-289.
	GENG Xuefeng, HE Yunze, LI Mengchuan, et al. An overview of IGBT multiphysics modeling technology and application[J]. Proceedings of the CSEE, 2022, 42(1):271-289.
[12]	施博辰, 赵争鸣, 蒋烨, 等. 功率开关器件多时间尺度瞬态模型(Ⅰ)——开关特性与瞬态建模[J]. 电工技术学报, 2017, 32(12):16-24.
	SHI Bochen, ZHAO Zhengming, JIANG Ye, et al. Multi-time scale transient models for power semiconductor devices[J]. Transactions of China Electrotechnical Society, 2017, 32(12):16-24.
[13]	刘运华, 张波, 谢帆, 等. 多尺度和多物理场的电力电子变换器建模方法初探[J]. 电力系统自动化, 2020, 44(16):61-69.
	LIU Yunhua, ZHANG Bo, XIE Fan, et al. Preliminary study on modeling methods with multiscale and multiphysics for power electronic converters[J]. Automation of Electric Power Systems, 2020, 44(16):61-69.
[14]	张驰, 郭媛, 黎明. 人工神经网络模型发展及应用综述[J]. 计算机工程与应用, 2021, 57(11):57-69.
	ZHANG Chi, GUO Yuan, LI Ming. Review of development and application of artificial neural network models[J]. Computer Engineering and Applications, 2021, 57(11):57-69.
[15]	AGATONOVIC-KUSTRIN S, BERESFORD R. Basic concepts of artificial neural network (ANN) modeling and its application in pharmaceutical research[J]. Journal of Pharmaceutical and Biomedical Analysis, 2000, 22(5):717-727. doi: 10.1016/S0731-7085(99)00272-1
[16]	PHAM D T, LIU X. Identification of linear and nonlinear dynamic systems using recurrent neural networks[J]. Artificial Intelligence in Engineering, 1993, 8(1):67-75. doi: 10.1016/0954-1810(93)90032-B
[17]	CHEN H L, YE S Z. Modeling and optimization of EMI filter by using artificial neural network[J]. IEEE Transactions on Electromagnetic Compatibility, 2020, 99:1-9.
[18]	WANG Jianpeng, XU Meng, ZHANG Jin, et al. A data-based IGBT model for efficient and accurate electro-thermal analysis[C]// 2020 IEEE Energy Conversion Congress and Exposition,IEEE, 2020:3442-3448.
[19]	曹建安, 裴云庆, 王兆安. Boost PFC电路中开关器件的损耗分析与计算[J]. 电工电能新技术, 2002(1):41-44.
	CAO Jianan, PEI Yunqing, WANG Zhaoan. Loss analysis and calculation of switching devices in Boost PFC circuit[J]. New Technology of Electrical Energy, 2002(1):41-44.
[20]	HURLEY W G, WOLFLE W H. Transformers and inductors for power electronics: Theory, design and applications[M]. London:Wiley, 2013.
[21]	张卫平. 开关变换器的建模和控制[M]. 北京: 机械工业出版社, 2019.
	ZHANG Weiping. Modeling and control of switching converter[M]. Beijing: China Machine Press, 2019.

参数	数值
输入电压V_in/V	14~20
输出电压U_o/V	28~31
开关频率f_sw/kHz	80~120
负载电阻R_L/Ω	30
储能电感L/μH	120
输出电容C/μF	200

模型	固定变量	MSE	R
f (V_in, I_in, f_sw)	U_o=28 V	1.08×10^-6/1.08×10^-6/1.08×10^-6	0.999/0.999/0.999
f (V_in, I_in, f_sw)	U_o=31 V	1.33×10^-6/1.47×10^-6/1.48×10^-6	0.999/0.999/0.999
f (V_in, I_in, U_o)	f_sw=80 kHz	6.24×10^-6/6.45×10^-6/6.48×10^-6	0.999/0.999/0.999
f (V_in, I_in, U_o)	f_sw=120 kHz	6.07×10^-6/6.11×10^-6/6.37×10^-6	0.999/0.999/0.999
f (V_in, f_sw, P_loss)	U_o=28 V	2.25×10^-4/2.59×10^-4/2.49×10^-4	0.998/0.998/0.998
f (V_in, f_sw, P_loss)	U_o=31 V	2.35×10^-4/2.41×10^-4/2.48×10^-4	0.999/0.999/0.999
f (V_in, U_o, P_loss)	f_sw=80 kHz	1.39×10^-3/1.41×10^-3/1.43×10^-3	0.995/0.995/0.995
f (V_in, U_o, P_loss)	f_sw=120 kHz	6.39×10^-4/6.46×10^-4/6.51×10^-4	0.996/0.996/0.996