1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
|
subroutine fexcd(jflag,nc,nd,t,y,ydp)
c input variables jflag, nc, nd, t, y
c jflag=0 or 1
c nc = # of continuous states yc
c nd = # of discrete states yd
c t = time
c y = state variable = [yc; yd]
c output variable = ydp
c
c iflag=0 >> external routine must
c load ydp(1:nc) with ydot=d/dt ( yc(t) )
c derivative of continuous state
c iflag=1 >> external routine must
c load ydp(1:nd) with yplus= yd(t+)
c update of discrete state
c here y=[yc;yd] has dimension nc+nd=3+2
c
c Example:
c 1/ call this fexcd:
c y0c=[1;2;3]; y0d=[1;-1]; nd=2; t0=0; t=1:10;
c y=odedc([y0c;y0d],nd,[5,0],t0,t,'fexcd')
c
c 2/ using dynamic link:
c link('fexcd.o','fexcd')
c y0c=[1;2;3]; y0d=[1;-1]; nd=2; t0=0; t=1:10;
c y=odedc([y0c;y0d],nd,[5,0],t0,t,'fexcd')
c
c 3/ passing a parameter to fexcd routine:
c y=odedc([y0c;y0d],nd,[5,0],t0,t,list('fexcd',param))
c param can be retrieved in fexcd by:
c param(1)=y(nc+nd+1) , param(2)=y(nc+nd+2) etc
c with this calling sequence y is a nc+nd+np vector
c where np=dimension of scilab variable param
c
c Copyright INRIA
double precision t, y, ydp
dimension y(*), ydp(*)
if(jflag.eq.0) then
ydp(1) = y(4)
ydp(2) = y(5)
ydp(3) = 0.d0
elseif(jflag.eq.1) then
ydp(1)=-y(4)
ydp(2)=-y(5)
endif
return
end
c The odedc example in the manual:
subroutine fcd(jflag,nc,nd,t,x,xdp)
c set dimensions of u,v,y
double precision u(1),v(1),y(1)
c
double precision t,x(*),xdp(*)
if (jflag.eq.0) then
c fc(t,xc,e(t)-hd(t,xd)
c u=e(t)-hd(t,xd)
call finput(t,v)
call hd(t,x(nc+1),u)
u(1)=v(1)-u(1)
c u(2)=v(2)-u(2) ....
call fc(t,x,u,xdp)
elseif(jflag.eq.1) then
call hc(t,x,y)
call fd(x(nc+1),y,xdp)
endif
return
end
subroutine fc(t,xc,u,xdot)
double precision t,xc(*),xdot(*),u(*)
c A=[-10,2,3;4,-10,6;7,8,-10];B=[1;1;1]
xdot(1)=-10*xc(1)+2*xc(2)+3*xc(3)+u(1)
xdot(2)=4*xc(1)-10*xc(2)+6*xc(3)+u(1)
xdot(3)=7*xc(1)+8*xc(2)-10*xc(3)+u(1)
return
end
subroutine hc(t,x,y)
double precision t,x(*),y(*)
y(1)=x(1)+x(2)+x(3)
return
end
subroutine fd(xd,y,xp)
c Ad=[0.5,1;0,0.05] Bd=[1;1]
double precision xd(*),y(*),xp(*)
xp(1)=0.5D0*xd(1)+xd(2)+y(1)
xp(2)=0*xd(1)+0.05D0*xd(2)+y(1)
return
end
subroutine hd(t,xd,u)
double precision t,xd(*),u(*)
u(1)=xd(1)+xd(2)
return
end
c The odedc example in the manual:
c It is assumed here that scilab variables
c A,B,C and Ad,Bd,Cd exist.
c
subroutine fcd1(jflag,nc,nd,t,x,xdp)
c
c iflag=0 --> returns in xcd(1:nc) dot(xc=x(1:nc))
c iflag=1 --> returns in xcd(1:nd) update(xd=x(nc+1:nc+nd))
c
c set here dimensions of u,v,y
double precision u(1),v(1),y(1)
c
double precision t,x(*),xdp(*)
if (jflag.eq.0) then
c xcd=fc1(t,xc,u)
c u=e(t)-hd1(t,xd)
call finput(t,v)
call hd1(t,x(nc+1),u)
u(1)=v(1)-u(1)
c u(2)=v(2)-u(2) ....
call fc1(t,x,u,xdp)
elseif(jflag.eq.1) then
c xcd=fd1(xd,y) xd=[x(nc+1),x(nc+2),...]
c y=hc1(t,xc)
call hc1(t,x,y)
call fd1(x(nc+1),y,xdp)
endif
return
end
subroutine fc1(t,xc,u,xdot)
c xdot=A*xc+B*u
c A and B real scilab matrices
double precision t,xc(*),xdot(*),u(*)
include '../stack.h'
call matptr('A'//char(0),m,n,la)
c call dset(m,0.0d0,xdot,1)
c call dgemm('n','n',m,1,m,1.0d0,stk(la),m,xc,m,1.0d0,xdot,m)
call brdmmul(stk(la),m,xc,m,xdot,m,m,m,1)
call matptr('B'//char(0),m,nb,lb)
call dgemm('n','n',m,1,nb,1.0d0,stk(lb),m,u,1,1.0d0,xdot,m)
end
subroutine hc1(t,x,y)
double precision t,x(*),y(*)
include '../stack.h'
c y=C*x
call matptr('C'//char(0),m,n,lc)
call brdmmul(stk(lc),m,x,m,y,m,m,n,1)
end
subroutine fd1(xd,y,xp)
c xp=Ad*xd + Bd*y
double precision xd(*),y(*),xp(*)
include '../stack.h'
call matptr('Ad'//char(0),m,n,la)
call brdmmul(stk(la),m,xd,m,xp,m,m,m,1)
call matptr('Bd'//char(0),m,nb,lb)
call dgemm('n','n',m,1,nb,1.0d0,stk(lb),m,y,1,1.0d0,xp,m)
end
subroutine hd1(t,xd,u)
double precision t,xd(*),u(*)
c u=Cd*xd
include '../stack.h'
c y=C*x
call matptr('Cd'//char(0),m,n,lc)
call brdmmul(stk(lc),m,xd,m,u,m,m,n,1)
end
subroutine finput(t,v)
double precision t,v(1)
v(1)=sin(3*t)
end
c dot(x)=A x + B u with u= (0,1) step function
subroutine phis(jflag,nc,nd,t,x,xdp)
c
c iflag=0 --> returns in xcd(1:nc) dot(xc=x(1:nc))
c iflag=1 --> returns in xcd(1:nd) update(xd=x(nc+1:nc+nd))
c
double precision t,x(*),xdp(*)
if (jflag.eq.0) then
c dot(x)=A*x+B*xd
call sbrc(t,x,xdp)
elseif(jflag.eq.1) then
c xd=1-xd
xdp(1)=1.d0-x(nc+1)
endif
end
c dot(x)=A x + B u with u= piecewise triangular function
subroutine phit(jflag,nc,nd,t,x,xdp)
c
c iflag=0 --> returns in xcd(1:nc) dot(xc=x(1:nc))
c iflag=1 --> returns in xcd(1:nd) update(xd=x(nc+1:nc+nd))
c
double precision t,x(*),xdp(*)
if (jflag.eq.0) then
c dot(x1c)=A*x1c+B*x2c
c dot(x2c)=xd
call sbrc(t,x,xdp)
xdp(nc)=x(nc+1)
elseif(jflag.eq.1) then
c xd=-xd
xdp(1)=-x(nc+1)
endif
end
subroutine sbrc(t,x,xdot)
c xdot=A*x1+B*x2
c A and B real scilab matrices
double precision t,x(*),xdot(*)
include '../stack.h'
call matptr('A'//char(0),m,n,la)
call brdmmul(stk(la),m,x,m,xdot,m,m,m,1)
call matptr('B'//char(0),m,nb,lb)
call dgemm('n','n',m,1,nb,1.0d0,stk(lb),m,x(m+1),1,1.0d0,xdot,m)
end
subroutine brdmmul(a,na,b,nb,c,nc,l,m,n)
c Copyright INRIA
double precision a(*),b(*),c(*)
double precision ddot
integer na,nb,nc,l,m,n
integer i,j,ib,ic
c
ib=1
ic=0
do 30 j=1,n
do 20 i=1,l
c(ic+i)=ddot(m,a(i),na,b(ib),1)
20 continue
ic=ic+nc
ib=ib+nb
30 continue
return
end
|
