基于多元线性回归的阻性和容性电流分解  

Decomposition method of resistive current and capacitive current based on multiple linear regression

在线阅读下载全文

作  者:韩永森[1] 李忠华[1] 郑欢[1] 郭文敏[1] 

机构地区:[1]哈尔滨理工大学教育部工程电介质及其应用技术重点实验室,黑龙江哈尔滨150080

出  处:《电机与控制学报》2016年第11期53-60,共8页Electric Machines and Control

基  金:国家重点基础研究发展计划项目(973计划)(2014CB239504)

摘  要:为了研究交流电压作用下非线性半导体器件和非线性绝缘电介质的绝缘状态和介电性能,提出一种阻性和容性电流分解算法。以非线性电阻和非线性电容构成的并联等效电路为研究对象,推导响应电流关于激励电压的非线性方程。通过坐标变换,将其转化成多元线性方程。利用多元线性回归方法,获得等效电路参数且实现了阻性和容性电流的分解。定性分析该算法的抗干扰能力和对非标准正弦波电压的适应能力。仿真结果表明:该算法可以准确地实现阻性和容性电流的分解;当响应电流含有55 d B的噪声时,电路参数的求解误差较小;激励电压谐波分量对电路参数的求解几乎没有影响。实验结果表明:该算法可以实现MOA阀片在交流电压作用下全泄露电流的分解。To study the insulation status and dielectric performance of nonlinear semiconductor equipment and nonlinear insulating dielectrics under AC voltage, a new decomposition method of resistive and ca- pacitive current was proposed. Nonlinear parallel equivalent circuit which consists of nonlinear resistance and nonlinear capacitance was chosen as the research object. The nonlinear expression of responding cur- rent on excitation voltage was deduced and then was transformed to multiple linear expression with coordi- nate transformation. The equivalent circuit parameters were got and the decomposition of the resistive and capacitive currents was achieved using multiple linear regression. The anti-interference performance and the adaptability to the nonstandard sinusoidal voltage were analyzed qualitatively. The simulation results show that the proposed method can decompose the resistive and capacitive current accurately. The rela- tive errors of circuit parameters are smaller when the signal to noise ratio of noise is 55 riB. The harmonics of excitation voltage have little effect on the solution of circuit parameters. The experimental results indi- cate that this method can decompose the total leakage current for MOA varistor under AC voltage.

关 键 词:电流分解 非线性等效电路 电路参数估计 多元线性回归 

分 类 号:TM85[电气工程—高电压与绝缘技术]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象