如何提高精度？
程序定义的变量是单精度，但是想提高计算精度。我把变量都定义成为REAL(KIND=8)之后，程序报实参与虚参不匹配，所以困惑了。给位大大，帮帮忙改下呗，谢谢谢谢！PROGRAM MAIN
IMPLICIT NONE
INTEGER I
REAL T
REALCCU,CNBSN
REAL RES
REAL*8 ACU,ASC
REAL B,RRR
REAL CAV
REALt1,t2
REAL*16Q,F
REAL y
REAL Iop
F=0.000
t1=0.0
Iop=8930.0
B=12.84
RRR=150
ACU=3.80233E-5
ASC=3.80233E-5
t2=2.0
y=5.0
Q=Iop**2*y*(EXP(-2.0*y*t1/2.0)-EXP(-2.0*y*t2/2.0)) &
/(2.0*Acu**2.0)
WRITE(*,*) "Q=",Q
DO I=1,11
T=89.0+(i-1)*0.1
CALL CNB3SN(T,B,CNBSN)
CALL CCCU(T,CCU)
CALL DIANZULV(T,B,RRR,RES)
CAV=(CCU*ACU+CNBSN*ASC)/(ACU+ASC)
F=CAV*0.001/RES+F
END DO
STOP
END
SUBROUTINE CNB3SN(T,B,CNBSN)
IMPLICIT NONE
REAL b4,b3,b2,b1,b0
REALT,B
REALCNBSN
IF(T>0.0.AND.T<10.4 ) THEN
b4=0.0
b3=38.8-1.8*B+0.0634*B**2
b2=-110*exp(-0.434*B)
b1=207.0-3.83*B+2.86*B**2
b0=0
ELSEIF(T>10.4.AND.T<26.113 ) THEN
b4=0.0
b3=7.42
b2=0.0
b1=1522.0
b0=0.0
ELSEIF(T>=26.113 .AND. T<169.416) THEN
b4=0.0
b3=0.0
b2=-61.635
b1=19902.0
b0=-305807.0
ELSEIF(T>=169.416 .AND. T<300.0) THEN
b4=0.0
b3=0.0
b2=-7.4636
b1=4411.0
b0=763801.0
ELSE
b4=0.0
b3=0.0
b2=0.0
b1=0.0
b0=1415377.0
ENDIF
CNBSN=b4*(T**4)+b3*(T**3)+b2*(T**2)+b1*T+b0
RETURN
END
SUBROUTINE CCCU(T,CCU)
IMPLICIT NONE
REAL a0,a1,a2,a3,a4,a5,a6,a7
REAL dens,aa1,aa2,aa3,aa4,aa
REAL T
REAL CCU
a0=-1.91844
a1=-0.15973
a2=8.61013
a3=-18.996
a4=21.9661
a5=-12.7328
a6=3.54322
a7=-0.3797
aa1=a0*((log10(T))**0.0)+a1*((log10(T))**1.0)
aa2=a2*((log10(T))**2.0)+a3*((log10(T))**3.0)
aa3=a4*((log10(T))**4.0)+a5*((log10(T))**5.0)
aa4=a6*((log10(T))**6.0)+a7*((log10(T))**7.0)
aa=aa1+aa2+aa3+aa4
dens=8960.0
CCU=dens*(10.0**aa)*1.0
WRITE(*,*) "CCU=",CCU
RETURN
END
SUBROUTINE DIANZULV(T,B,RRR,RES)
REAL RES1,RES2,RES3,RES
RES1=1.7E-8/RRR
RES2=1.0E-8/(2.33*10**9/T**5+9.57*10**5/T**3+163.0/T)
RES3=(0.37E-10+0.0005E-10*RRR)*B
RES=RES1+RES2+RES3
RETURN
END 不想修改程序的话，可以通过添加编译器的编译指令实现
譬如gfortran可以参考 https://gcc.gnu.org/onlinedocs/gcc-7.2.0/gfortran/Fortran-Dialect-Options.html#Fortran-Dialect-Options
-fdefault-real-8
Set the default real type to an 8 byte wide type. This option also affects the kind of non-double real constants like 1.0, and does promote the default width of DOUBLE PRECISION to 16 bytes if possible, unless -fdefault-double-8 is given, too. Unlike -freal-4-real-8, it does not promote variables with explicit kind declaration.
或者
-freal-4-real-8
-freal-4-real-10
-freal-4-real-16
-freal-8-real-4
-freal-8-real-10
-freal-8-real-16
Promote all REAL(KIND=M) entities to REAL(KIND=N) entities. If REAL(KIND=N) is unavailable, then an error will be issued. All other real kind types are unaffected by this option. These options should be used with care and may not be suitable for your codes. Areas of possible concern include calls to external procedures, alignment in EQUIVALENCE and/or COMMON, generic interfaces, BOZ literal constant conversion, and I/O. Inspection of the intermediate representation of the translated Fortran code, produced by -fdump-tree-original, is suggested.
已解决，谢谢！
页:
[1]