A computationally efficient method for identifying network parameter errors

A recently proposed method that is based on the normalized Lagrange multipliers for identifying network parameter errors has shown to be very effective yet also carries a high computational burden prohibiting its efficient use for large transmission grids. This paper presents an efficient solution that greatly reduces this burden facilitating the method's implementation in large practical power systems. The reduction in the computational cost is achieved by exploiting the sparse structure of the jacobian and gain matrices. The necessary subset of entries of the inverse matrix is determined by a two-step back stepping logic, and they are computed by a modified version of the well-known 'sparse inverse' algorithm. Simulation results show that the proposed approach drastically reduces the computational load. Scenarios with different types of parameter errors are simulated to illustrate the application of the proposed implementation for very large size power systems.