!
! This file is part of SACAMOS, State of the Art CAble MOdels for Spice.
! It was developed by the University of Nottingham and the Netherlands Aerospace
! Centre (NLR) for ESA under contract number 4000112765/14/NL/HK.
!
! Copyright (C) 2016-2018 University of Nottingham
!
! SACAMOS is free software: you can redistribute it and/or modify it under the
! terms of the GNU General Public License as published by the Free Software
! Foundation, either version 3 of the License, or (at your option) any later
! version.
!
! SACAMOS is distributed in the hope that it will be useful, but
! WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
! or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
! for more details.
!
! A copy of the GNU General Public License version 3 can be found in the
! file GNU_GPL_v3 in the root or at .
!
! SACAMOS uses the EISPACK library (in /SRC/EISPACK). EISPACK is subject to
! the GNU Lesser General Public License. A copy of the GNU Lesser General Public
! License version can be found in the file GNU_LGPL in the root of EISPACK
! (/SRC/EISPACK ) or at .
!
! The University of Nottingham can be contacted at: ggiemr@nottingham.ac.uk
!
!
! File Contents:
! SUBROUTINE modal_decomposition_LC
! SUBROUTINE orthogonal_transformation
! SUBROUTINE normalise_matrix_columns
!
! NAME
! modal_decomposition_LC
!
! AUTHORS
! Chris Smartt
!
! DESCRIPTION
! This subroutine calculates the modal decomposition give the inductance and capacitance matrices L and C
! for an inhomogeneous, lossless configuration.
! The theory is in the Theory Manual_Section 3.3 however there is more detailed discussion in
! C. R. Paul,'Analasys of Multiconductor transmission lines,'
!
! Inputs: 1. L, the inductance matrix
! 2. C, the capacitance matrix
!
! Outputs 1. TI, TII, TV, TVI, the modal decomposition matrices
! 2. The mode velocities
! 3. The mode impedances
!
! COMMENTS
!
!
! HISTORY
!
! started 20/6/2015 CJS: Looking forward towards inhomogeneous transmission lines
! 8/5/2017 CJS: Include references to Theory_Manual
!
SUBROUTINE modal_decomposition_LC(dim,L,C,TI,TII,TV,TVI,mode_velocity,mode_impedance)
USE type_specifications
USE general_module
USE constants
USE eispack
USE maths
IMPLICIT NONE
! variables passed to the subroutine
integer,intent(IN) :: dim ! matrix dimension
real(dp),intent(IN) :: L(dim,dim) ! inductance matrix
real(dp),intent(IN) :: C(dim,dim) ! capacitance matrix
real(dp),intent(OUT) :: TI(dim,dim) ! modal decomposition matrix [TI]
real(dp),intent(OUT) :: TII(dim,dim) ! modal decomposition matrix inverse of [TI]
real(dp),intent(OUT) :: TV(dim,dim) ! modal decomposition matrix [TV]
real(dp),intent(OUT) :: TVI(dim,dim) ! modal decomposition matrix inverse of [TV]
real(dp),intent(OUT) :: mode_velocity(dim) ! mode velocity list
real(dp),intent(OUT) :: mode_impedance(dim) ! mode impedance list
! local variables
! orthonormal diagonalisation of C
real(dp) :: UT(dim,dim)
real(dp) :: thetasqr(dim,dim)
real(dp) :: U(dim,dim)
real(dp) :: theta(dim,dim)
! [M] and the orthonormal diagonalisation of [M]=[theta][UT][L][U][theta]
real(dp) :: M(dim,dim)
real(dp) :: ST(dim,dim)
real(dp) :: gammasqr(dim,dim)
real(dp) :: S(dim,dim)
! un-normalised mode transformation matrix, T
real(dp) :: T(dim,dim)
! normalised mode transformation matrix, Tnorm and the normalisation matrix, alpha
real(dp) :: Tnorm(dim,dim)
real(dp) :: alpha(dim,dim)
! mode inductance and capacitance
real(dp) ::lmode
real(dp) ::cmode
! other matrices
real(dp) :: TM1(dim,dim)
real(dp) :: TM2(dim,dim)
! other variables
integer :: row
integer :: ierr
! START
! Calculate the product ofthe transmission line impedance (Z) and admittance (Y) matrices
if(verbose) then
write(*,*)'CALLED modal_decomposition_LC'
write(*,*)'L'
CALL dwrite_matrix(L,dim,dim,dim,0)
write(*,*)'C'
CALL dwrite_matrix(C,dim,dim,dim,0)
TM1=matmul(L,C)
write(*,*)'LC'
CALL dwrite_matrix(TM1,dim,dim,dim,0)
end if
! Stage 1: calculate the orthogonal transformation to diagonalise the symmetric real matrix C, [UT][C][U]=[thetasqr]
! Theory_Manual_Equation 3.37
CALL orthogonal_transformation(C,UT,thetasqr,U,dim)
! calculate the square root of the diagonal matrix thetasqr
theta(:,:)=0d0
do row=1,dim
theta(row,row)=dsqrt(thetasqr(row,row))
end do
! form the product [M]=[theta][UT][L][U][theta]
! Theory_Manual_Equation 3.39
TM1=matmul(U,theta)
TM2=matmul(L,TM1)
TM1=matmul(UT,TM2)
M=matmul(theta,TM1)
! STAGE 2:
! Theory_Manual_Equation 3.40
CALL orthogonal_transformation(M,ST,gammasqr,S,dim)
! form the matrix [T]=[U][theta][S] then normalise the matrix columns
! Theory_Manual_Equation 3.42
TM1=matmul(theta,S)
T=matmul(U,TM1)
CALL normalise_matrix_columns(T,Tnorm,alpha,dim)
! set the modal decomposition matrices TI,TV,TII,TVI
! Theory_Manual_Equation 3.43,3,44,3.45
TI=Tnorm
TM1=transpose(Tnorm)
ierr=0 ! set ierr=0 on input to matrix inverse to cause the program to stop if we have a singular matrix
CALL dinvert_Gauss_Jordan(TM1,dim,TV,dim,ierr)
TVI=transpose(TI)
TII=transpose(TV)
! calculate the mode inductance and capacitance and hence the mode velocities and impedances
! Theory_Manual_Equation 3.46,3,47
do row=1,dim
lmode=(alpha(row,row)**2)*gammasqr(row,row)
cmode=1d0/(alpha(row,row)**2)
mode_velocity(row)=1d0/sqrt(lmode*cmode)
mode_impedance(row)=sqrt(lmode/cmode)
end do
if (verbose) then
! do some checks
write(*,*)' mode alpha gammasqr'
do row=1,dim
write(*,*)row,alpha(row,row),gammasqr(row,row)
end do
write(*,*)' mode lmode cmode mode_velocity mode_impedance'
do row=1,dim
write(*,*)row,(alpha(row,row)**2)*gammasqr(row,row),(alpha(row,row)**(-2)),mode_velocity(row),mode_impedance(row)
end do
end if
RETURN
END SUBROUTINE modal_decomposition_LC
!
! __________________________________________________________
!
!
SUBROUTINE orthogonal_transformation(M_in,UT,thetasqr,U,dim)
!
! DESCRIPTION
! calculate an orthogonal transformation which diagonalises the symmetric matrix M_in
! [UT][M_in][U]=[thetasqr] where [thetasqr] is a diagonal matrix.
!
! The subroutine uses the eispack routine rg to calculate the eigenvalues and eigenvectors
! of a real symmetric matrix.
!
!
! COMMENTS
! The eigenvalues (and eigenvectors) are ordered in ascending order
!
! HISTORY
!
! 20/06/2016 CJS.
!
USE type_specifications
USE general_module
USE eispack
USE maths
IMPLICIT NONE
! vasriables passed to subroutine
integer,intent(IN) :: dim ! matrix dimension
real(dp),intent(IN) :: M_in(dim,dim) ! input symmetric matrix
! output matrices
real(dp),intent(OUT) :: UT(dim,dim)
real(dp),intent(OUT) :: thetasqr(dim,dim)
real(dp),intent(OUT) :: U(dim,dim)
! EISPACK variables
real(dp) :: M(dim,dim)
real(dp) :: w(dim) ! vector of eigenvalues
integer(i4):: matz
integer(i4):: ierr
! local matrices for checks
real(dp) :: UTU(dim,dim)
real(dp) :: Uthetasqr(dim,dim)
real(dp) :: UthetasqrUT(dim,dim)
! loop variables
integer :: row,col
! START
! copy the M_in matrix
M(:,:)=M_in(:,:)
! call the eispack routine rs
matz=1 ! calculate both eigenvalues and eigenvectors
CALL rs ( dim, M, w, matz, U, ierr )
if (ierr.NE.0) then
run_status='ERROR in eispack routine rs called from orthogonal_transformation'
CALL write_program_status()
STOP 1
end if
thetasqr(:,:)=0d0
do row=1,dim
thetasqr(row,row)=w(row)
end do
UT=TRANSPOSE(U)
! Some checks
if (verbose) then
write(*,*)'Eigenvalues'
do row=1,dim
write(*,*)row,w(row)
end do
write(*,*)'Eigenvectors:'
do col=1,dim
write(*,*)'Eigenvector ',col
do row=1,dim
write(*,*)row,UT(row,col)
end do
end do
! check that the diagonalisation is orthogonal
UTU=matmul(UT,U)
write(*,*)'Check [UT][U]=[I]'
CALL dwrite_matrix(UTU,dim,dim,dim,0)
write(*,*)'[M]'
CALL dwrite_matrix(M_in,dim,dim,dim,0)
write(*,*)'Check [U][thetasqr][UT]=[M]'
Uthetasqr=matmul(U,thetasqr)
UthetasqrUT=matmul(Uthetasqr,UT)
CALL dwrite_matrix(UthetasqrUT,dim,dim,dim,0)
end if ! verbose
END SUBROUTINE orthogonal_transformation
!
! __________________________________________________________
!
!
SUBROUTINE normalise_matrix_columns(T,Tnorm,alpha,dim)
!
! DESCRIPTION
! normalise the columns of matrix T to have length 1.
! Return the normalised matrix and the diagonal normalisation matrix, alpha
! such that [Tnorm]=[T][alpha]
!
!
! COMMENTS
!
!
! HISTORY
!
! 20/06/2016 CJS.
!
USE type_specifications
USE maths
IMPLICIT NONE
! vasriables passed to subroutine
integer,intent(IN) :: dim ! matrix dimension
real(dp),intent(IN) :: T(dim,dim) ! input matrix
! output matrices such that [Tnorm]=[T][alpha]
real(dp),intent(OUT) :: Tnorm(dim,dim)
real(dp),intent(OUT) :: alpha(dim,dim)
! loop variables
integer :: row,col
real(dp) :: length
! START
alpha(:,:)=0d0
do col=1,dim
length=0d0
do row=1,dim
length=length+T(row,col)*T(row,col)
end do
length=sqrt(length)
alpha(col,col)=1d0/length
do row=1,dim
Tnorm(row,col)=T(row,col)/length
end do
end do
END SUBROUTINE normalise_matrix_columns