On sparse matrices in C++ using templates and collection classes. I. Modelling power system matrices

Solving Ax=b, where A is a sparse matrix, is a frequent engineering problem; however, in power transmission system calculations A is among the sparsest matrices in all engineering problems. Over the years a large number of software packages have been developed to handle operations with large sparse matrices (BLAS, LAPAK, LINPAK, LAPAK++, SparseLib++, etc.). The development of matrix tools for extended real time application in power system operation and control centres is based on an object-oriented approach using C++. This research explores the usage of templates and collection classes in matrix algebra for power system calculations based on Visual C++ developed for Windows NT.