State estimation based real-time markets - challenges and practical solutions

The main purpose of this presentation is to identify challenges posed by the state estimation (SE) based LMP markets relative to an efficient implementation of real-time energy market clearing and associated real-time dispatch. First, the problem of an accurate and reliable real-time state estimator solution is discussed. The output of the SE in real-time is the principal input to the real-time market clearing engine usually based on security constrained dynamic dispatch. Additionally the SE solution is used as one of the main inputs into ex-post LMP calculations whose results are critical for settlements of energy and ancillary services. Beyond normal reliability operations these two uses of SE results make SE a critical function in any real-time LMP market operation. LMP energy markets typically encompass a large geographical areas resulting in large network model and large SE problem with questionable observability. In situations like this the SE algorithm has to be robust and to provide reasonable solution even in cases with reduced observability and errors introduced with modeling and inaccurate measurements. To achieve this goal special techniques for filtering inaccurate measurements and carefully placing large weights on external measurement have been used in orthogonal factorization based state estimator. Various techniques such multi pass state estimation solution have been used to mitigate the effect of the inaccuracy of measurements in the external systems. In this presentation we show some alternatives to achieve the same goals in a single-pass state estimator solution.