Matrix: YCheng/psse0

Description: Power system state simulation matrix, Yunzhi Cheng, UT Arlington

 (bipartite graph drawing)

• Matrix group: YCheng
• download as a MATLAB mat-file, file size: 817 KB. Use UFget(1870) or UFget('YCheng/psse0') in MATLAB.

 Matrix properties number of rows 26,722 number of columns 11,028 nonzeros 102,432 structural full rank? yes structural rank 11,028 # of blocks from dmperm 1 # strongly connected comp. 1 explicit zero entries 0 nonzero pattern symmetry 0% numeric value symmetry 0% type real structure rectangular Cholesky candidate? no positive definite? no

 author Y. Cheng editor T. Davis date 2007 kind power network problem 2D/3D problem? no

 Additional fields size and type b full 26722-by-1 guess full 11028-by-1

Notes:

```Power system state simulation matrix from Yunzhi Cheng, UT Arlington.
In MATLAB, the solution to x=A\b is desired, but this can be slow in
MATLAB 7.3 because of the speed of sparse QR as compared to sparse
Cholesky.  Using x=(A'*A)\(A'*b) is faster, but of course yields
slightly less accurate (but still acceptable) results.  Note that an
initial guess to the solution is provided, for use by an iterative
method.  However, sparse Cholesky with an AMD ordering is very fast
for this matrix and thus iterative methods are unlikely to be
competitive.  In MATLAB 7.3 on a 3.2 Ghz Pentium 4 desktop,
x=(A'*A)\(A'*b) takes 0.07 seconds.
```

 Ordering statistics: result nnz(V) for QR, upper bound nnz(L) for LU, with COLAMD 2,433,000 nnz(R) for QR, upper bound nnz(U) for LU, with COLAMD 84,587

 SVD-based statistics: norm(A) 200252 min(svd(A)) 0.185943 cond(A) 1.07695e+06 rank(A) 11,028 sprank(A)-rank(A) 0 null space dimension 0 full numerical rank? yes

 singular values (MAT file): click here SVD method used: s = svd (full (R)) ; where [~,R,E] = spqr (A) with droptol of zero status: ok