63文章解读与程序——电力系统保护与控制EI\CSCD\北大核心《基于混沌集成决策树的电能质量复合扰动识别》已提供下载资源

👆👆👆👆👆👆👆👆👆👆👆👆👆👆👆👆👆👆👆👆👆👆👆👆👆下载资源链接👆👆👆👆👆👆👆👆👆👆👆👆👆👆👆👆👆👆👆👆👆👆👆👆👆👆

《《《《《《《《更多资源还请持续关注本专栏》》》》》》》

论文与完整源程序_电网论文源程序的博客-CSDN博客https://blog.csdn.net/liang674027206/category_12531414.html

电网论文源程序-CSDN博客电网论文源程序擅长文章解读,论文与完整源程序,等方面的知识,电网论文源程序关注python,机器学习,计算机视觉,深度学习,神经网络,数据挖掘领域.https://blog.csdn.net/LIANG674027206?type=download

针对电能质量复合扰动类别多、特征关联性强及识别错误率较高的问题，提出了一种基于混沌集成决策树的电能质量复合扰动识别方法。首先参考IEEE标准，给出了常见的7种单一电能质量扰动和16种电能质量复合扰动的信号模型，并批量生成扰动波形样本。然后针对上述扰动的特性差异，通过S变换时频域分析，设计和提取出9种扰动时频域特征。最后利用集成学习集体能力和混沌搜索优势，构建混沌集成决策树，并有效完成了电能质量复合扰动识别。仿真实验和142组实测数据验证结果表明，该方法对于23 种扰动的分类准确率高于基本决策树、复杂决策树及加权最近邻法等方法，具有良好的应用前景。

部分代码展示：

%对电能质量扰动进行S变换时频分析

%%  初始值
clear
clc 
%区间(a，b)内产生随机数r = a + (b-a).*rand([m n]));
w=100*pi;       x=0:0.001:0.4;
a_3=0.15*rand(1);a_5=0.15*rand(1);a_7=0.15*rand(1);
b_3=2*pi*rand(1);b_5=2*pi*rand(1);b_7=2*pi*rand(1);  %谐波参数 a=0-0.15   b=0-2*pi

a2=0.1+0.8*rand(1);         %电压暂降/暂升参数0.1-0.9   t2-t1=4T-9T
a4=0.9+0.1*rand(1);        %电压中断参数  a4=0.9-1     t2-t1=4T-9T
a_f=0.3+0.2*rand(1);    b=0.1+0.3*rand(1);      %电压闪变参数     a_f=0.3-0.5   b=0.1-0.4 
a6=0.1+0.7*rand(1); tao=0.008+0.032*rand(1); Wn=6+12*rand(1);       %电压振荡参数a6=0.1-0.8 tao=0.008-0.04  Wn=6-18 t2-t1=0.05T-3T
a7=1+9*rand(1);   tao=0.008+0.032*rand(1);    %暂态脉冲参数a7=1-10   tao=0.008-0.04 t2-t1=0.05T-3T
 
%%  电能质量扰动函数
V1=sin(w*x)+a_3*sin(3*w*x+b_3)+a_5*sin(5*w*x+b_5)+a_7*sin(7*w*x+b_7);
% plot(t,V1);    title('1:谐波'); 谐波
V2=(1-a2.*(stepfun(x,0.08)-stepfun(x,0.16))).*sin(w*x);
%   plot(t,V2)  电压暂降   0.1-0.9
V3=(1+a2.*(stepfun(x,0.08)-stepfun(x,0.16))).*sin(w*x);
  % 电压暂升    
V4=(1-a4.*(stepfun(x,0.08)-stepfun(x,0.16))).*sin(w*x);
%   plot(t,V4)      %0.9-1  电压中断
V5=(1+a_f*sin(b*w*x)).*sin(w*x);
%subplot(2,1,2);  plot(t,V5);   title('5:闪变');
V6=sin(w*x)+a6*exp(-(x-0.02)/tao).*sin(Wn*w*x).*(stepfun(x,0.02)-stepfun(x,0.03));
%plot(t,V6)   暂态振荡
V7=sin(w*x)+a7*exp(-(x-0.02)/tao).*(stepfun(x,0.02)-stepfun(x,0.06)); 
%plot(t,V7)   暂态脉冲

V8=(1-a2.*(stepfun(x,0.18)-stepfun(x,0.26))).*sin(w*x)+a_3*sin(3*w*x+b_3)+a_5*sin(5*w*x+b_5)+a_7*sin(7*w*x+b_7);
%电压暂降+谐波
V9=(1+a2.*(stepfun(x,0.18)-stepfun(x,0.26))).*sin(w*x)+a_3*sin(3*w*x+b_3)+a_5*sin(5*w*x+b_5)+a_7*sin(7*w*x+b_7);
%电压暂升+谐波
V10=(1-a4.*(stepfun(x,0.08)-stepfun(x,0.16))).*sin(w*x)+a_3*sin(3*w*x+b_3)+a_5*sin(5*w*x+b_5)+a_7*sin(7*w*x+b_7);
%电压中断+谐波
V11=(1+a_f*sin(b*w*x)).*sin(w*x)+a_3*sin(3*w*x+b_3)+a_5*sin(5*w*x+b_5)+a_7*sin(7*w*x+b_7);
%电压闪变+谐波

V12=(1-a2.*(stepfun(x,0.08)-stepfun(x,0.16))).*sin(w*x)+a6*exp(-(x-0.02)/tao).*sin(Wn*w*x).*(stepfun(x,0.02)-stepfun(x,0.03));
%电压暂降+暂态振荡
V13=(1+a2.*(stepfun(x,0.08)-stepfun(x,0.16))).*sin(w*x)+a6*exp(-(x-0.02)/tao).*sin(Wn*w*x).*(stepfun(x,0.02)-stepfun(x,0.03));
%电压暂升+暂态振荡
V14=(1+a_f*sin(b*w*x)).*sin(w*x)+a6*exp(-(x-0.02)/tao).*sin(Wn*w*x).*(stepfun(x,0.02)-stepfun(x,0.03));
%电压闪变+暂态振荡
V15=sin(w*x)+a_3*sin(3*w*x+b_3)+a_5*sin(5*w*x+b_5)+a_7*sin(7*w*x+b_7)+a6*exp(-(x-0.02)/tao).*sin(Wn*w*x).*(stepfun(x,0.02)-stepfun(x,0.03));
%谐波+暂态振荡

效果展示：