查看: 1266|回复: 2

之前的初始数据可行解正确，换了个就没有可行解，求大神帮忙，急

字体大小: 正常放大

1 主题	2 听众	3 积分

升级 60%

该用户从未签到

自我介绍: adfa

电梯直达

1^#

发表于 2014-6-27 12:39 |只看该作者 |倒序浏览

|招呼Ta 关注Ta

本帖最后由 wujianjack2 于 2014-6-28 10:22 编辑

测试出来，第五行约束为@for(VIT(i): fvec(i)<=1023);时可行解正确，可改为@for(VIT(i): fvec(i)<=1024);就没有可行解了，调试信息我也看不明白。本来约束条件里应该是没有这行约束的，如果去掉程序一直运行。
但在以前的初始数据下是正确解。请大神帮忙分析下，看到底是哪里的问题：
data:
m0 = 100 0 0 0 2 2 0 0 100;
pFlag = 0 2 2 2 1 1 2 2 0;
bnd = 100 20 1 1 2 2 1 2 100;
itFlag = 1 0 0 0 1 0 0;
wpt = 1 0 0 0 0 0 0
0 1 0 0 0 0 0
0 0 1 0 0 0 0
0 0 0 1 0 0 0
1 0 0 0 0 2 0
0 1 1 0 1 0 0
0 0 0 0 0 0 1
0 0 0 0 0 1 0
0 0 0 0 1 0 0;
wtp = 0 1 0 0 0 0 0 0 0
0 0 1 0 0 0 0 0 0
0 0 0 1 1 0 0 0 0
1 0 0 0 0 2 0 0 0
0 0 0 0 0 0 0 1 0
0 0 0 0 0 1 1 0 0
0 0 0 0 2 0 0 0 1;
mvalue = -1 0 0 1 0 0 0
1 -1 0 0 0 0 0
0 1 -1 0 0 0 0
0 0 1 -1 0 0 0
-1 0 1 0 0 -2 2
0 -1 -1 2 -1 1 0
0 0 0 0 0 1 -1
0 0 0 0 1 -1 0
0 0 0 0 -1 0 1;
enddata
model:
sets:
!place m0 - initial marking;
!pFlag: 0==idle place, 1==resource, 2==working place;
!bnd - the place's boundness, m - reachable marking;
VIP:m0,pFlag,bnd,m,ps;
!transition itFlag: 1==initial transition, 0 == others;
!fvec - the firing vector of transitions;
VIT:itFlag, fvec;
!the arc from place to transition;
Pre(VIP,VIT):wpt;
!the arc from transition to place;
Post(VIT,VIP):wtp;
!incidence matirx;
Matrix(VIP,VIT):mvalue;

!;
BadMarkedSiphon(VIP):vp;
EnabledTrans(VIT):et;
EnabledResourse(VIP,VIT): ert;
endsets
data:
m0 = 5 0 0 0 0 5 0 0 0 2 2 2;
pFlag = 0 2 2 2 2 0 2 2 2 1 1 1 ;
bnd = 5 5 5 5 5 5 5 5 5 5 5 5 ;
itFlag = 1 0 0 0 0 0 1 0 0 0 ;
wpt = 1 0 0 0 0 0 0 0 0 0
0 1 0 1 0 0 0 0 0 0
0 0 1 0 0 0 0 0 0 0
0 0 0 0 1 0 0 0 0 0
0 0 0 0 0 1 0 0 0 0
0 0 0 0 0 0 1 0 0 0
0 0 0 0 0 0 0 1 0 0
0 0 0 0 0 0 0 0 1 0
0 0 0 0 0 0 0 0 0 1
1 0 0 0 0 0 0 0 1 0
2 0 0 0 0 0 0 1 0 0
0 0 1 0 1 0 1 0 0 0 ;
wtp = 0 1 0 0 0 0 0 0 0 0 0 0
0 0 1 0 0 0 0 0 0 1 1 0
0 0 0 0 1 0 0 0 0 0 1 0
0 0 0 1 0 0 0 0 0 1 0 0
0 0 0 0 1 0 0 0 0 0 2 0
1 0 0 0 0 0 0 0 0 0 0 1
0 0 0 0 0 0 1 0 0 0 0 0
0 0 0 0 0 0 0 1 0 0 0 1
0 0 0 0 0 0 0 0 1 0 1 0
0 0 0 0 0 1 0 0 0 1 0 0 ;
mvalue = -1 0 0 0 0 1 0 0 0 0
1 -1 0 -1 0 0 0 0 0 0
0 1 -1 0 0 0 0 0 0 0
0 0 0 1 -1 0 0 0 0 0
0 0 1 0 1 -1 0 0 0 0
0 0 0 0 0 0 -1 0 0 1
0 0 0 0 0 0 1 -1 0 0
0 0 0 0 0 0 0 1 -1 0
0 0 0 0 0 0 0 0 1 -1
-1 1 0 1 0 0 0 0 -1 1
-2 1 1 0 2 0 0 -1 1 0
0 0 -1 0 -1 1 -1 1 0 0 ;
@text(Minimize_Bad_Siphon.dat)= @write('the place whose value is 0 is included in a bad siphon ');
@text(Minimize_Bad_Siphon.dat)= @writefor(VIP(i)|pFlag#NE#0:@newline(1),'p',VIP(i),'=',vp(i));
enddata
max = @sum(BadMarkedSiphon(i)|pFlag(i)#NE#0:vp(i));
!constrants (1);
@for(VIP(i):
@for(VIT(j)| pFlag(i)#NE#0 #AND# wtp(j,i)#NE#0 : vp(i) >= @sum(VIP(z):@if(wpt(z,j)#NE#0,1,0)*vp(z))-
@sum(VIP(z):@if(wpt(z,j)#NE#0,1,0))+1);
);
!constrants (2);
@sum(VIP(i)|pFlag(i)#NE#0 : vp(i)) < @sum(VIP(i) : @if(pFlag(i)#NE#0,1,0))-0.00001;
!constrants (3);
@for(VIT(i) :
@for(VIP(j) | itFlag(i)#NE#1 #AND# wpt(j,i)#NE#0 #AND# pFlag(j)#EQ#2 : m(j) >= et(i));
);
!constrants (4);
@for(VIT(i) :
@for(VIP(j) | itFlag(i)#NE#1 #AND# wpt(j,i)#NE#0 #AND# pFlag(j)#EQ#2 : et(i) >= m(j)/bnd(j));
);
!constrants (5);
@for(VIP(i) :
@for(VIT(j) | pFlag(i)#EQ#1 #AND# itFlag(j)#NE#1 #AND# wpt(i,j)#NE#0: (m(i)/wpt(i,j)+vp(i)) >= ert(i,j));
);
!constrants (6);
@for(VIP(i) :
@for(VIT(j) | pFlag(i)#EQ#1 #AND# itFlag(j)#NE#1 #AND# wpt(i,j)#NE#0: ert(i,j) >= (m(i)-wpt(i,j)+1)/(m0(i)-wpt(i,j)+1));
);
!constrants (7);
@for(VIP(i) :
@for(VIT(j) | pFlag(i)#EQ#1 #AND# itFlag(j)#NE#1 #AND# wpt(i,j)#NE#0: ert(i,j) >= vp(i));
);
!constrants (8);
@for(VIT(i) | itFlag(i)#NE#1: @sum(EnabledResourse(j,i)| pFlag(j)#EQ#1 #AND# wpt(j,i)#NE#0: ert(j,i))+0.00001 < @sum(VIP(z)| pFlag(z)#EQ#1: @if(wpt(z,i)#NE#0,1,0))+1-et(i) );
!constrants (9);
@for(VIP(i) | pFlag(i)#NE#0 : @bin(vp(i)));
!constrants (10);
@for(VIT(i) | itFlag(i)#NE#1 : @bin(et(i)));
!constrants (11);
@for(VIP(i) :
@for(VIT(j)| pFlag(i)#EQ#1 #AND# itFlag(j)#NE#1 #AND# wpt(i,j)#NE#0 :@bin(ert(i,j)));
);
!constrants (12);
@for(VIP(i)|pFlag(i)#EQ#2: ps(i)=@if(m(i)#gt#0,1,0));
@sum(VIP(i)|pFlag(i)#EQ#2: ps(i))>1;
@for(VIP(i): m(i)>=0);
@for(VIT(i): fvec(i)>=0 );
@for(VIT(i): fvec(i)<=1024);
@for(VIP(i):m(i)= m0(i)+@sum(VIT(j):mvalue(i,j)*fvec(j)));
@for(VIP(i):@gin(m(i)));
@for(VIT(i):@gin(fvec(i)));
@for(VIP(i)|pFlag(i)#EQ#2: @bin(ps(i)););
end

zan

转播0 淘帖0 分享0 收藏0 支持0 反对0 微信

使用道具举报

wujianjack2

21 主题	97 听众	3110 积分

TA的每日心情

	奋斗 2014-3-2 00:26

签到天数: 243 天

[LV.8]以坛为家I

2^#

发表于 2014-6-28 10:33 |只看该作者 |招呼Ta 关注Ta

你的问题，LINDO 14与LINGO 11求解出来的全局解不一样。
原始数据的解：
  Global optimal solution found.
  Objective value:                      9.00000000000
  Objective bound:                      9.00000000000
  Infeasibilities:                      0.00000000000
  Extended solver steps:                            0
  Total solver iterations:                            9
  Elapsed runtime seconds:                         0.17

                                                            Variable             Value
                                                               M( 1)       5.00000000000
                                                               M( 6)       5.00000000000
                                                            M( 10)       2.00000000000
                                                            M( 11)       2.00000000000
                                                            M( 12)       2.00000000000
                                                            PS( 1)       1.23456788063
                                                            PS( 2)       1.00000000000
                                                            PS( 5)       1.00000000000
                                                            PS( 6)       1.23456788063
                                                            PS( 9)       1.00000000000
                                                            PS( 10)       1.23456788063
                                                            PS( 11)       1.23456788063
                                                            PS( 12)       1.23456788063
                                                            FVEC( 7)       1019.00000000
                                                            FVEC( 8)       1019.00000000
                                                            FVEC( 9)       1019.00000000
                                                         FVEC( 10)       1019.00000000

修改数据后的解：
  Global optimal solution found.
  Objective value:                      6.00000000000
  Objective bound:                      6.00000000000
  Infeasibilities:                      0.00000000000
  Extended solver steps:                            0
  Total solver iterations:                            3
  Elapsed runtime seconds:                         0.06

                                                            Variable             Value
                                                               M( 1)       100.000000000
                                                               M( 5)       2.00000000000
                                                               M( 6)       2.00000000000
                                                               M( 9)       100.000000000
                                                            PS( 1)       1.23456788063
                                                            PS( 4)       1.00000000000
                                                            PS( 5)       1.23456788063
                                                            PS( 6)       1.23456788063
                                                            PS( 7)       1.00000000000
                                                            PS( 9)       1.23456788063
                                                            FVEC( 1)       1022.00000000
                                                            FVEC( 2)       1022.00000000
                                                            FVEC( 3)       1022.00000000
                                                            FVEC( 4)       1022.00000000
                                                            FVEC( 5)       1022.00000000
                                                            FVEC( 6)       1022.00000000
                                                            FVEC( 7)       1022.00000000

程序太长，逻辑判断太多，看你书写的格式估计是在做个大作业或者小项目，你很难让人花大把时间去仔细看你的程序的。
不排除软件计算有误，MINLP问题求解本身是非常困难的。首先，你应该仔细检查自己的模型，有时候或许是笔误，或者思考不周，导致浪费大量时间也是有可能的。