Decomposed numerical observability investigation suitable for parallel processor implementation

A problem associated with numerical observability analysis in power systems is the computing time required. A new approach, to overcome this problem, is presented. The basis of the approach is the decomposition of the system into subsystems. The system Jacobian matrix is decomposed into reduced-order submatrices. It was found that the global rank test on the Jacobian matrix is equivalent to rank tests on these submatrices. When distributing the task of rank tests to parallel processors a considerable reduction in computing time may be achieved. This decomposed treatment may help in localising the source of unobservability