|本期目录/Table of Contents|

[1]王青山,梁得亮,杜锦华.交流稳压电源的改进神经网络PID控制[J].电机与控制学报,2017,21(02):1-9.[doi:10.15938/j.emc.2017.02.001]
 WANG Qing-slian,LIANGDe-liang DU Jin-hua.Improved neural network PID controller for regulated power supply [J].,2017,21(02):1-9.[doi:10.15938/j.emc.2017.02.001]
点击复制

交流稳压电源的改进神经网络PID控制(PDF)
分享到:

《电机与控制学报》[ISSN:1007-449X/CN:23-1 408/TM]

卷:
21
期数:
2017年02
页码:
1-9
栏目:
出版日期:
2017-02-01

文章信息/Info

Title:
Improved neural network PID controller for regulated power supply


作者:
王青山梁得亮杜锦华
?(西安交通大学电力设备电气绝缘国家重点实验室,陕西西安710049)
Author(s):
?WANG Qing-slian LIANGDe-liang DU Jin-hua
?(State Key Laboratory of Electrical Insulation and Power Equipment,Xi’an Jiaotong University,Xi’an 710049,China)

关键词:
交流稳压电源PID控制器人工神经网络Levenberg-Marquardt算法连接权值
Keywords:
regulated power supply PIDcontroller artificial neural network Levenberg-Marquardt algo- rithm connection weight
分类号:
TM 464
DOI:
10.15938/j.emc.2017.02.001
文献标志码:
A
摘要:
?建立了交流稳压电源主电路数学模型并分析其闭环稳压控制原理。由于装置具有较强的非线性和变结构、变参数特性,采用经典PID控制器很难获得理想的控制效果。将人工神经网络与传统PID控制器相结合,构成一种不依赖于被控对象精确数学模型的神经网络=D控制器。为了提高神经网络的收敛速度,采用Levenberg-Marquardt算法计算连接权值更新量,并对当前解施加一个以一定概率保留的随机扰动,加快迭代过程跳出局部极小点。对装置主电路和改进神经网络PID控制器进行仿真,结果表明:系统动态响应快,鲁棒性强,调节平滑,具有较好的控制效果。最后,制造并测试了额定电压660 V、容量400 kVA的实验样机,对理论研究进行了实验验证。
Abstract:
The mathematical model of device’s main circuit is established and the closed-loop voltage sta-bilization control method is analyzed. With the strong non-linearity and variable structures and variable parameters,it is diicult to achieve ideal control effects using the classic PID controller. Artificial neural network was combined with conventional PID regulator to construct a neural network PID controller that did not rely on the precise mathematical model of controlled objects. To attain faster convergence speed of the neural network,the Levenberg-Marquardt algorithm was adopted to calculate the updating quantities of connection weights, to which random disturbances retained in certain probability were applied for speeding up the iterative process out of local minima. The device] main circuit together with neural net- work PIDcontroller was simulated and the results showthat the system has quick responses , strong ro-bustness and smooth adjustment. Testing and validation of such controller were also conducted experimentally using a prototype witii voltage rating 660 V and volume rating 400 kVA.

参考文献/References:

-

相似文献/References:

[1]杨益飞,骆敏舟,邢绍邦,等. D-分割技术确定磁轴承的PID参数鲁棒稳定域[J].电机与控制学报,2015,19(06):95.[doi:10. 15938/j. emc.2015.06.015]
 YANG Yi-fei,LUO Min-zhou,XING Shao-bang,et al. Robust stability regions of PID parameters for magneticbearing based on D-partition technique[J].,2015,19(02):95.[doi:10. 15938/j. emc.2015.06.015]

备注/Memo

备注/Memo:
收稿日期: 2014-12-14
基金项目:国家自然科学基金(51177125)
作者简介:王青山(1989—#,男,博士研究生,研究方向为新型智能配网变压器及其控制系统;
梁得亮(1965—),男,教授,博士生导师,研究方向为电磁器件及其系统的分析与实现;
杜锦华(1984—),女,博士,讲师,研究方向为变压器设计及参数分析技术。
更新日期/Last Update: 2017-03-30