小波神經網絡進行短期電力負荷預測的matlab代碼 　2008-05-10 21:24
分類：默認分類 字號： 大大  中中  小小 

clear;
P=[ 0.2286 0.2090 0.0442 0.3690 0.2603 0.0359 0.0724 0.1759 0.2634;
0.1292 0.0947 0.0880 0.2222 0.1715 0.1149 0.1909 0.2347 0.2258;
0.0720 0.1393 0.1147 0.0562 0.0702 0.1230 0.1340 0.1829 0.1165;
0.1529 0.8700 0.0563 0.5157 0.2711 0.5460 0.2409 0.1811 0.1124;
0.1335 0.2558 0.3347 0.1872 0.1491 0.1977 0.2842 0.2922 0.1074;
0.0733 0.0900 0.1100 0.1614 0.1330 0.1248 0.0450 0.0655 0.0657;
0.1159 0.0771 0.1453 0.1425 0.0968 0.0624 0.0824 0.0774 0.0610;
0.0940 0.0882 0.0429 0.1506 0.1911 0.0832 0.1064 0.2237 0.2623;
0.0522 0.0393 0.1818 0.1310 0.2545 0.1640 0.1909 0.2056 0.2588;
0.1345 0.1430 0.0378 0.0500 0.0871 0.1002 0.1586 0.0925 0.1155;
0.0090 0.0126 0.0092 0.0078 0.0060 0.0059 0.0116 0.0078 0.0050;
0.1260 0.1670 0.2251 0.0348 0.1793 0.1503 0.1698 0.1852 0.0978;
0.3619 0.2450 0.1560 0.0451 0.1001 0.1837 0.3644 0.3501 0.1511;
0.0690 0.0508 0.0852 0.0707 0.0789 0.1295 0.2718 0.1680 0.2273;
0.1828 0.1328 0.0670 0.0880 0.0909 0.0700 0.2494 0.2668 0.3220];
T=[1 1 1 0 0 0 0 0 0;0 0 0 1 1 1 0 0 0;0 0 0 0 0 0 1 1 1];
threshold=[0 1;0 1;0 1;0 1;0 1;0 1;0 1;0 1;0 1;0 1;0 1;0 1;0 1;0 1;0 1];
net=newff(threshold,[15,3],{'morlet','logsig'},'traingda');
net.trainParam.epochs=500;
net.trainParam.goal=0.001;
LP.Ir=0.01;
s=15;r=15;
[r,q] = size(P);
pmin = min(P')';
pmax = max(P')'
magw =0.1*s^(1/r);
w = magw*randnr(s,r);
b = magw*rands(s,1);
b1=rands(s,1);
rng = pmax-pmin;
mid = 0.5*(pmin+pmax);
w=2*w./(ones(s,1)*rng');
b = b - w*mid;
a=w*rng/2.45;
t=w*mid;
net.IW{1,1}=w;
net.b{1,1}=b;
net.LW{2,1}=0.01*randnr(3,s);
net.b{2,1}=0.1*randnc(3,1);
net=train(net,P,T);
w1=net.IW{1,1};
b1=net.b{1,1};
w2=net.LW{2,1};
b2=net.b{2,1};
%保存網絡訓練結果
fww1=fopen('w11.dat','w');
fwb1=fopen('b11.dat','w');
fww2=fopen('w22.dat','w');
fwb2=fopen('b22.dat','w');
fprintf(fww1,'%6.4f %6.4f\n',w1);
fprintf(fwb1,'%6.4f %6.4f\n',b1);
fprintf(fww2,'%6.4f %6.4f\n',w2);
fprintf(fwb2,'%6.4f %6.4f\n',b2);
fclose(fww1);
fclose(fwb1);
fclose(fww2);
fclose(fwb2);
g=[ 0.2101 0.2593 0.2599;0.095 0.18 0.2235;0.1298 0.0711 0.1201;0.1359 0.2801 0.1171;
0.2601 0.1501 0.1102;0.1001 0.1298 0.0683;0.0753 0.1001 0.0621;0.089 0.1891 0.2597;
0.0389 0.2531 0.2602;0.1451 0.0875 0.1167;0.0128 0.0058 0.0048;0.159 0.1803 0.1002;
0.2452 0.0992 0.1521;0.0512 0.0802 0.2281;0.1319 0.1002 0.3205];
y=sim(net,g)
 ====================================================

問題來源：http://chinavib.com/forum/thread-48742-1-1.html

程序一：GA訓練BP權值的主函數

function net=GABPNET(XX,YY)

%--------------------------------------------------------------------------

%  GABPNET.m

%  使用遺傳算法對BP網絡權值閾值進行優化，再用BP算法訓練網絡

%--------------------------------------------------------------------------

%數據歸一化預處理

nntwarn off

XX=premnmx(XX);

YY=premnmx(YY);

%創建網絡

net=newff(minmax(XX),[19,25,1],{'tansig','tansig','purelin'},'trainlm');

%下面使用遺傳算法對網絡進行優化

P=XX;

T=YY;

R=size(P,1);

S2=size(T,1);

S1=25;%隱含層節點數

S=R*S1+S1*S2+S1+S2;%遺傳算法編碼長度

aa=ones(S,1)*[-1,1];

popu=50;%種群規模

initPpp=initializega(popu,aa,'gabpEval');%初始化種群

gen=100;%遺傳代數

%下面調用gaot工具箱，其中目標函數定義為gabpEval

[x,endPop,bPop,trace]=ga(aa,'gabpEval',[],initPpp,[1e-6 1 1],'maxGenTerm',gen,...

  'normGeomSelect',[0.09],['arithXover'],[2],'nonUnifMutation',[2 gen 3]);

%繪收斂曲線圖

figure(1)

plot(trace(:,1),1./trace(:,3),'r-');

hold on

plot(trace(:,1),1./trace(:,2),'b-');

xlabel('Generation');

ylabel('Sum-Squared Error');

figure(2)

plot(trace(:,1),trace(:,3),'r-');

hold on

plot(trace(:,1),trace(:,2),'b-');

xlabel('Generation');

ylabel('Fittness');

%下面將初步得到的權值矩陣賦給尚未開始訓練的BP網絡

[W1,B1,W2,B2,P,T,A1,A2,SE,val]=gadecod(x);

net.LW{2,1}=W1;

net.LW{3,2}=W2;

net.b{2,1}=B1;

net.b{3,1}=B2;

XX=P;

YY=T;

%設置訓練參數

net.trainParam.show=1;

net.trainParam.lr=1;

net.trainParam.epochs=50;

net.trainParam.goal=0.001;

%訓練網絡

net=train(net,XX,YY);

程序二：適應值函數

function [sol, val] = gabpEval(sol,options)

% val - the fittness of this individual

% sol - the individual, returned to allow for Lamarckian evolution

% options - [current_generation]

load data2

nntwarn off

XX=premnmx(XX);

YY=premnmx(YY);

P=XX;

T=YY;

R=size(P,1);

S2=size(T,1);

S1=25;%隱含層節點數

S=R*S1+S1*S2+S1+S2;%遺傳算法編碼長度

for i=1:S,

   x(i)=sol(i);

end;

[W1, B1, W2, B2, P, T, A1, A2, SE, val]=gadecod(x);

程序三：編解碼函數

function [W1, B1, W2, B2, P, T, A1, A2, SE, val]=gadecod(x)

load data2

nntwarn off

XX=premnmx(XX);

YY=premnmx(YY);

P=XX;

T=YY;

R=size(P,1);

S2=size(T,1);

S1=25;%隱含層節點數

S=R*S1+S1*S2+S1+S2;%遺傳算法編碼長度

% 前R*S1個編碼為W1

for i=1:S1,

    for k=1:R,

      W1(i,k)=x(R*(i-1)+k);

    end

end

% 接著的S1*S2個編碼（即第R*S1個后的編碼）為W2

for i=1:S2,

   for k=1:S1,

      W2(i,k)=x(S1*(i-1)+k+R*S1);

   end

end

% 接著的S1個編碼（即第R*S1+S1*S2個后的編碼）為B1

for i=1:S1,

   B1(i,1)=x((R*S1+S1*S2)+i);

end

% 接著的S2個編碼（即第R*S1+S1*S2+S1個后的編碼）為B2

for i=1:S2,

   B2(i,1)=x((R*S1+S1*S2+S1)+i);

end

% 計算S1與S2層的輸出

A1=tansig(W1*P,B1);

A2=purelin(W2*A1,B2);

% 計算誤差平方和

SE=sumsqr(T-A2);

val=1/SE; % 遺傳算法的適應值

上述程序需要調用gaot工具箱

===============

initializega 函數的代碼:

function [pop] = initializega(num, bounds, evalFN,evalOps,options)

% function [pop]=initializega(populationSize, variableBounds,evalFN,

%                           evalOps,options)

%    initializega creates a matrix of random numbers with 

%    a number of rows equal to the populationSize and a number

%    columns equal to the number of rows in bounds plus 1 for

%    the f(x) value which is found by applying the evalFN.

%    This is used by the ga to create the population if it

%    is not supplied.

%

% pop            - the initial, evaluated, random population 

% populatoinSize - the size of the population, i.e. the number to create

% variableBounds - a matrix which contains the bounds of each variable, i.e.

%                  [var1_high var1_low; var2_high var2_low; ....]

% evalFN         - the evaluation fn, usually the name of the .m file for 

%                  evaluation

% evalOps        - any options to be passed to the eval function defaults []

% options        - options to the initialize function, ie. 

%                  [type prec] where eps is the epsilon value 

%                  and the second option is 1 for float and 0 for binary, 

%                  prec is the precision of the variables defaults [1e-6 1]

% Binary and Real-Valued Simulation Evolution for Matlab GAOT V2 

% Copyright (C) 1998 C.R. Houck, J.A. Joines, M.G. Kay 

%

% C.R. Houck, J.Joines, and M.Kay. A genetic algorithm for function

% optimization: A Matlab implementation. ACM Transactions on Mathmatical

% Software, Submitted 1996.

%

% This program is free software; you can redistribute it and/or modify

% it under the terms of the GNU General Public License as published by

% the Free Software Foundation; either version 1, or (at your option)

% any later version.

%

% This program is distributed in the hope that it will be useful,

% but WITHOUT ANY WARRANTY; without even the implied warranty of

% MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the

% GNU General Public License for more details. A copy of the GNU 

% General Public License can be obtained from the 

% Free Software Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.

if nargin<5

  options=[1e-6 1];

end

if nargin<4

  evalOps=[];

end

if any(evalFN<48) %Not a .m file

  if options(2)==1 %Float GA

    estr=['x=pop(i,1); pop(i,xZomeLength)=', evalFN ';'];  

  else %Binary GA

    estr=['x=b2f(pop(i,:),bounds,bits); pop(i,xZomeLength)=', evalFN ';']; 

  end

else %A .m file

  if options(2)==1 %Float GA

    estr=['[ pop(i,:) pop(i,xZomeLength)]=' evalFN '(pop(i,:),[0 evalOps]);']; 

  else %Binary GA

    estr=['x=b2f(pop(i,:),bounds,bits);[x v]=' evalFN ...

        '(x,[0 evalOps]); pop(i,:)=[f2b(x,bounds,bits) v];'];  

    end

end

numVars     = size(bounds,1);                 %Number of variables

rng         = (bounds(:,2)-bounds(:,1))'; %The variable ranges'

if options(2)==1 %Float GA

  xZomeLength = numVars+1;                 %Length of string is numVar + fit

  pop         = zeros(num,xZomeLength);         %Allocate the new population

  pop(:,1:numVars)=(ones(num,1)*rng).*(rand(num,numVars))+...

    (ones(num,1)*bounds(:,1)');

else %Binary GA

  bits=calcbits(bounds,options(1));

  xZomeLength = sum(bits)+1;                 %Length of string is numVar + fit

  pop = round(rand(num,sum(bits)+1));

end

for i=1:num

  eval(estr);

end