Identification and updating of minimally dependent sets of measurements in state estimation

A new algorithm for the identification and updating of minimally dependent sets of measurements is presented. The technique implemented in the algorithm is a mixed numerical-symbolic method based on a reduced model and graph theory. As a byproduct of the algorithm, critical measurements and error residual spread areas are provided. Emphasis is placed in the updating of the above quantities when one or more measurements are eliminated from the measurement set. Computational aspects of the proposed algorithm are discussed, and results from an illustrative example and test cases are reported. >