基于神经网络的预测控制模型仿?- 百度文库

Ԥģͷ

ժҪһȨֵߵĶ̬BPģԤϲԤģһ𹹳ɶ̬Ԥڴ˻γɶģж̬Ԥ㷨㷨Ԥ⾫ȣǿԤ㷨³ԡ

ؼʣԤ ̬ 

1.

̬(DMC)һڽȶԻԶԤ㷨 Ŀǰѹ㷺Ӧڹҵ̿ơڶԾӦϵԤģͣ ˽ģ򵥣 ͬʱöಽŻ뷴УϣֱӴʱͶ󣬲õĸܺͽǿ³ԡ

ǣDMC㷨ʵʿдһϵ⣬ģʧձڵһ⣬᲻̶ͬȵӰϵͳܡDMCʵʿвģʧԭҪ2һ罨ģŵأʵʿƵȫ̷ΧDMCģʧ䣻ʵϵͳķԣһʹñضģͷ仯ʱһ̶ĽԾӦƿȫ̷ΧĿƣȻʹʵʿڶķǽģڳģʧ䡣DMCģʧ⣬ѧ߽˴оȡ˷ḻоɹлDMCƲ߱ʶܿ㷨ģ߱ʶУ㷨ԼԪģͱʶڱʶĻٽж̬ƵȡЩ㷨ܽ߱ʶģͲԶԶ󽵽׽ģ(ṹԽģ)³ԲãŽС⣬˻УĶ̬㷨ЩûʱԤѧģԭģõδԤ⡣˻Ԥⷽ붯̬ƣӦʵʡַȻʹϵͳֳõȶԣȷѧģͻһѡ

ͨѵѧϰܱƽн纯ص㣬һֲBPԤDMCԤģ͡ԤԤģδܰһвȷϢԹΪBPһϵйȥϢԤδΪģԤҪ䣬ͽѧģ͵ĸһֲڶģѼ򻯻޷Կǵһء

ͨз棬֤˻Ԥ㷨ЧԼԽԣ

Ӷǿ˶̬㷨³ԡ

2.BPģ

BPģʹϢĻԭǣźXiͨмڵ㣨㣩ڵ㣬α任źYkѵÿXtֵYֵt֮ƫͨڵڵǿȡֵWijڵڵ֮ǿTjkԼֵʹݶȷ½ѧϰѵȷСӦȨֵֵѵֹͣʱѵ缴ܶϢдСľתϢ 2.1 BPģͽ

BPģͰģ͡úģ͡ģͺѧϰģ͡ 1ڵģ

ڵģͣOj=f(WijXi-q j) (1)

ڵģͣYk=f(TjkOj-q k) (2)

f-úq -񾭵Ԫֵ

2úģ

úǷӳ²ϲڵ̼ǿȵĺֳƴ̼һȡΪ(0,1)ȡֵSigmoid f(x)= 1/(1+e) 3 3ģ

ģǷӳ֮Сĺ

Ep=1/2(tpi-Opi) tpi- iڵֵOpi-iڵֵ (4)

4ѧϰģ

ѧϰ̣²ڵϲڵ֮ȨؾWij趨̡BPʦѧϰʽ-Ҫ趨ֵʦѧϰʽ-ֻģʽ֮֡ѧϰģΪ

Wij(n+1)= hiOj+aWij(n) (5)

h-ѧϰӣi-ڵiļOj-ڵjļa-ӡ 2.2 BPģ͵ȱݷŻ 1ѧϰh Ż

ñ䲽СԶѧϰӣٵͼӿٶȡ

h =h+a(Ep(n)- Ep(n-1))/ Ep(n)

aΪ0~1֮ȡֵ 6

2ڵŻ

2

-x

ڵĶٶܵӰϴ󣬵ڵ̫ʱᵼѧϰʱڵСʱݴ𲽻ع вԼ̬ɾһЩصڵ㣬ڵɾ׼ɸýڵָһڵȨֵֵͨȡ0.10.05䣩֮УýڵɾڵLɲο湫ʽ㣺

L=(m+n)1/2+c (7) m-ڵn-ڵc-110ij

3Ԫȷ

öԪعдɾǿڵ 4㷨Ż

BP㷨õ½ھֲСѵʱϳû߻ƵؼȫܱδŴ㷨IGAȡͳBP㷨˷ȱ㡣

2.3 BP

豸ضĽѧģΪ yout(k)? 8

ʽУϵͳakʱģa(k)?1.2(1?0.8e?0.1ka(k)yout(k?1)?u(k?1)1?yout2(k?1)

)

Ľṹѡ2-6-1ѧϰٶ =0.5͹ϵ=0.05ָźΪ

rin(k)=0.5*sin(6**t) clear all; x=[0,0]'; close all; u_1=0; xite=0.50; y_1=0; alfa=0.05; I=[0,0,0,0,0,0]'; w2=rand(6,1); Iout=[0,0,0,0,0,0]'; w2_1=w2;w2_2=w2; FI=[0,0,0,0,0,0]'; w1=rand(2,6); ts=0.001; w1_1=w1;w1_2=w1; for k=1:1:1000 dw1=0*w1; time(k)=k*ts; u(k)=0.50*sin(3*2*pi*k*ts);

a(k)=1.2*(1-0.8*exp(-0.1*k)); y(k)=a(k)*y_1/(1+y_1^2)+u_1;

for j=1:1:6

I(j)=x'*w1(:,j);

Iout(j)=1/(1+exp(-I(j))); end

yn(k)=w2'*Iout; e(k)=y(k)-yn(k);

w2=w2_1+(xite*e(k))*Iout+alfa*(w2_1-w2_2);

for j=1:1:6

FI(j)=exp(-I(j))/(1+exp(-I(j)))^2; end

for i=1:1:2 for j=1:1:6

dw1(i,j)=e(k)*xite*FI(j)*w2(j)*x(i); end end

w1=w1_1+dw1+alfa*(w1_1-w1_2);

x(1)=u(k); x(2)=y(k);

w1_2=w1_1;w1_1=w1; w2_2=w2_1;w2_1=w2; u_1=u(k); y_1=y(k); end

subplot(2,1,1);

plot(time,y,'r',time,yn,'b');

xlabel('times');ylabel('ʵͷ'); grid on

subplot(2,1,2);

plot(time,y-yn,'r');

xlabel('times');ylabel('error'); grid on

ϵͷ779662525#qq.com(#滻Ϊ@)