Intel® Fortran Compiler
Build applications that can scale for the future with optimized code designed for Intel® Xeon® and compatible processors.

warning message from BCONF

yuezhanwei
Beginner
461 Views
Can someone please help? I have no clue what I can do in this case. The code I am using is the following:

integer, parameter :: dp = selected_real_kind(p = 8, r = 30)
REAL(dp)::cons, b_p, temp_s

INTEGER:: IPARAM(7), ITP, L, NOUT
REAL(dp) F, FSCALE, RPARAM(7), X(N), XGUESS(N), XLB(N), XSCALE(N), XUB(N)
EXTERNAL:: FirmValue

XGUESS = (/b_p, temp_s/)
XSCALE = (/1.E0,1.E0/)
XLB = (/B_min, S_min/)
XUB = (/B_max, S_max/)
FSCALE = 1.E0
ITP = 0;
CALL U4INF (IPARAM, RPARAM)
IPARAM(1) = 1;
IPARAM(4) = 1000;
IPARAM(5) = 1000;
CALL BCONF(FirmValue, 0, XLB, XUB, X, XGUESS=XGUESS, XSCALE=XSCALE, IPARAM=iparam, & RPARAM=rparam)
CALL ERSET (0, 1, 0)

I got the following waring messages:
*** WARNING ERROR 11 from DU5INF. The relative function tolarence must be
*** negative while RFTOL = -5.592394780117511D+29 is given. The
*** algorithm will use RFTOL = 1.00000000000D-10.
Here is a trace back of subprogram calls in reverse order:
Routine name Eror type Error code
DU5INF 6 11 (called internally)
DB3ONF 0 0 (called internally)
DB2ONF 0 0 (called internally)
DBCONF 0 0
USER 0 0
*** WARNING ERROR 14 from DU5INF. The maximum allowable scaled step length
*** must be positive with STEPMX = 0.00000000000D+00 is given. The
*** algorithm will use STEPMX= Infinity.
Here is a trace back of subprogram calls in reverse order:
Routine name Eror type Error code
DU5INF 6 14 (called internally)
DB3ONF 0 0 (called internally)
DB2ONF 0 0 (called internally)
DBCONF 0 0
USER 0 0

I also tried to change RPARAM(1) and RPARAM(2) to 1.0E-16. But the program exited without executing BCONF. I highly appreciate your help or any hint.
0 Kudos
1 Reply
yuezhanwei
Beginner
461 Views
I figured out a way to solve my problem. I am posting it in case someone else runs into similar problems.
I am solving a complex nonlinear two dimentional minimization problem with simple bound constraints. I first used BCONF and run into all kinds of problems. Now I changed to LCONF with NEQ=0, A=zeros(NCON,NVAR), B=zeros(NVAR). This works beautifully.
0 Kudos
Reply