matlab用RBF预测的程序,【求助】RBF数据预测建模仿真

该楼层疑似违规已被系统折叠 隐藏此楼查看此楼

%RBF预测模型

P1=[6801144.2880341263.8271361200.501512240251031.581092;

6801147.2724321266.914541196.2551931840251040.486211;

6801110.7859321257.1626041234.8733872440141063.831463;

6801102.5708871252.2413831234.4053843131131081.434287;

6821095.612791246.7743441235.2351793722141060.590935;

6871093.0398331245.1191121235.1499353820151056.767843;

6781090.7613511244.4589061234.3900963918151086.869741;

6801090.5520841243.1349341233.6835574017171062.05787;

6781111.3180651245.1358521202.5594863435221027.130695;

6801114.3433141244.2121451203.8472733434271031.403123;

6731114.8698711244.6112911204.4696063434241034.380123;

6851114.0594171243.8961611207.3840533434241035.012351;

6751114.0156321242.9546941211.1587883434261040.342943;

6821113.144671243.0473421214.26523434261042.897312;

6731112.6267961244.6325561219.4220543434261046.517504;

6801115.1223161245.9012051222.5275823434261039.554514;

6701118.9107531246.5785591219.8281313434261039.625884;

6721120.0983311247.4867321215.1880233434261039.834126;

6871120.3597881247.6193441208.7807293434261024.927316;

6821119.8265911247.5627431206.4018913434261033.284887;

6801117.8540281248.0771311202.0935473434271032.288727;

6791115.0567561247.981681202.6128643434271034.633368;

6731111.8100271247.6031205.6581223434281029.191511;

6801110.7040171248.6154371209.106343437261031.709948;

6751110.254331248.3086591212.8865413436301031.970216;

6801109.8525811245.8776841215.3953883037281046.323101;

6731099.1313371228.0517161222.8322773442171048.62035;

6881098.8183811226.8153281222.5451833241181039.707965;

6831098.6108481226.7472271222.122893437261056.062007;

6811095.6243131229.4128961220.5656233536271048.807404;

];

[t_data,ps]=mapminmax(P1,0,1);

%初始化数据

tt=t_data(:,8);

x=t_data(:,1:7);

tt=tt';

%随机选取中心

c=x;

%定义delta平方为样本各点的协方差之和

delta=cov(x');

% 计算协方差

% Covariance matrix

delta=sum(delta);

%隐含层输出R

for i=1:1:30

for j=1:1:30

R(i,j)=((x(i,:)-c(j,:)))*((x(i,:)-c(j,:))');

R(i,j)=exp(-R(i,j)./delta(j));

end

end

p=R;

%建模

%r=radbas(p);

err_goal=0.001;

sc=3;

net=newrb(p,tt,err_goal,sc,200,1);

%测试

ty=sim(net,p);

tE=tt-ty;

tSSE=sse(tE);

tMSE=mse(tE);

%预测(测试)曲线

figure;

plot(tt,'-+');

hold on;

plot(ty,'r:*');

legend('实际值','预测值');

title('RBF网络模型输出预测曲线');

% axis([1,30,0,1]);

NEWRB, neurons = 0, MSE = 0.000209463

NEWRB, neurons = 2, MSE = 3.88575e-05

>>

新手求助RBF神经网络数据预测问题-数据.xls 本程序用前六个数据预测下一数据,前200组数据用于训练,用后80组数据进行预测,可预测结果为一个值,请大家指教: 源程序为: clc clear close all %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %每五个数据整合, x=xlsread; n=length m=5; sum=0; j=1; x1=zeros); for k=1:m:n     for i=k:         sum=sum x;     end     x1=sum;     j=j 1;     sum=0; end j t=1:; figure plot;%,'d-m' hold on; grid on title; %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %归一化处理 [x2,mint,maxt] = premnmx figure plot;%,'d-m' hold on; grid on title; %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %输入数据 for k=1:1:200     p_train=[x2 x2 x2 x2 x2 x2]; 5*6     t_train=x2; 5*1 end     p_train=p_train'; %6*195     t_train=t_train'; %1*195 for z=1:1:81     p_test=[x2 x2 x2 x2 x2 x2];     t_test=x2; end      p_test=p_test';      t_test=t_test'; %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %建立 RBF 网络                      goal = 0.0001;                     % 训练误差的平方和 spread = 0.01;                     % 此值越大,需要的神经元就越少 MN = size;              % 最大神经元数 DF = 1;                            % 显示间隔 net = newrb; data_out=sim p_mse=mse figure plot hold on plot title; %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %预测并作图 y=sim y_mse=mse figure plot; hold on; plot title; legend;
新手求助RBF神经网络数据预测问题-历史开奖数据.xlsx 本程序用前六个数据预测下一数据,前200组数据用于训练,用后80组数据进行预测,可预测结果为一个值,请大家指教: 源程序为: clc clear close all %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %每五个数据整合, x=xlsread; n=length m=5; sum=0; j=1; x1=zeros); for k=1:m:n     for i=k:         sum=sum x;     end     x1=sum;     j=j 1;     sum=0; end j t=1:; figure plot;%,'d-m' hold on; grid on title; %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %归一化处理 [x2,mint,maxt] = premnmx figure plot;%,'d-m' hold on; grid on title; %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %输入数据 for k=1:1:200     p_train=[x2 x2 x2 x2 x2 x2]; 5*6     t_train=x2; 5*1 end     p_train=p_train'; %6*195     t_train=t_train'; %1*195 for z=1:1:81     p_test=[x2 x2 x2 x2 x2 x2];     t_test=x2; end      p_test=p_test';      t_test=t_test'; %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %建立 RBF 网络                      goal = 0.0001;                     % 训练误差的平方和 spread = 0.01;                     % 此值越大,需要的神经元就越少 MN = size;              % 最大神经元数 DF = 1;                            % 显示间隔 net = newrb; data_out=sim p_mse=mse figure plot hold on plot title; %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %预测并作图 y=sim y_mse=mse figure plot; hold on; plot title; legend;
评论
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值