1-s2.0-S0378779611000769-main

Electric Power Systems Research 81 (2011) 1514–152 4 Contents lists available at ScienceDirect Electric Po we r Sys tems Research j ournal h o me p a g e: www.elsevier.com/locate/epsr State estimation and bad data processing for systems including PMU and SCADA measurements George N. Korres ∗ , Nikolaos M. Manousakis School of Electrical and Computer Engineering, National Technical University of Athens 9, Iroon Polytechneiou Street, Zografou 15780, Athens, Greece a r t i c l e i n f o Article history: Received 30 November 2010 Received in revised form 3 March 2011 Accepted 6 March 2011 Available online 2 April 2011 Keywords: State estimation Phasor measurement unit Voltage phasor Current phasor Bad data processing a b s t r a c t Conventional state estimators (SE) are based on real-time measurements, consisting of bus voltages and active and reactive power flows and injections, and estimate the voltage phasors of the network buses. Until recently, these measurements were obtained only through SCADA. With the advent of GPS synchronized measurements obtained by phasor measurement units (PMU), effective techniques are required to incorporate the extremely accurate PMU measurements into state estimation, in order to improve its performance and observability. This paper develops a non-linear weighted least squares estimator by modeling the current phasor measurements either in rectangular or in polar coordinates and compares the two approaches. Any numerical problems arised at flat start or for lightly loaded lin es, are res olv ed. Theerror ampli fica tion, dueto thecurren t pha sormeasure men t transf ormati on from pol ar int o rec tan gul ar coordinates, is also inv est iga ted. The normalized residual test is used to effective ly ide nti fy any baddata in theconve nti ona l andphaso r mea sur eme nts. The propos ed tec hniques aretested with the IEEE 14-bus system. © 2011 Elsevier B.V. All rights reserved. 1. Intro ducti on Secure operation of a power system requires monitoring of its opera tingcondition s. This is trad ition allyaccompli shed by the state estimator, which provides optimal estimates of the bus voltage phasors, based on redundant measurements commonly provided by SCADA, including active and reactive power flows and injections andbus vol tag e mag nit ude s. Wit h the increasin g use of PMUs in recent years, the accuracy and reliability of state estimates and baddatadete cti oncan beenhanc ed [1–3]. Unliketraditi onalSCADA systems, the PMU is able to measure the voltage phasor of the installed bus and the current phasors of all the lines connected with that bus. In spite of their enormo us potential for improving the operation and real-time control of large transmission systems, PMUs wil l notmake sta te est ima tio n obs olete butwillhelp imp rov- ing its performance. When rec tan gul ar coo rdi nat es areused forthe pha sormeasurements and state variables, in a system completely observable by only PMUs, their relationship becomes linear and states can be obta ined by a linea r non-i terative algor ithm [4]. However, since the majority of measurements in existing estimators are of the tradi- tio naltype,it is har d to be ful ly replac ed byPMUs in thenear fut ure. As a conseq uence,state estimatorsincludingboth phasor and trad i- tional measurements have to be develo ped . In [5], state estimatio n ∗ Corresponding author. Tel.: +30 772 3621; fax: +30 772 3659. E-mail address: gkorres@sof tlab.ece.ntua.gr (G.N. Korres). problem is formulated without using any actual or virtual reference bus. The main advantage of this formulation is that it can successfull y detect andident ify erroneous PMUs.A mult i-are a state estimator is suggested in [6], where a central coordinator receives the results of individual area state estimators as well as measurements from the area boundaries and few globally synchronized phasor measurements from area buses, and computes the system widesolution. In [7] a two-s tep algor ithmis prop osed,which incor - por ate s the pha sormeasurements and the res ults of the tra dit ional SE in a post processing linear estimator and provides the same results as the nonlinear algorithm. In [8] a rectangular coordinate formulation is presented, by which numerical problems encoun- tered during flat start when using current phasors are avoided. Furthermore, no reference bus is used, which facilitates bad data processing for conventional as well as phasor measurements. A distr ibut ed stat e estimator, utili zing synch roniz ed phaso r measurements, is presented in [9]. The aggregated solution is obtained from the distributed solution using a sensitivity analysis based update at chosen boundary buses and an algorithm is developed to locate PMUs and determine the slack bus in each subsystem to coordinate the distributed SE solution. A hybrid state estimator, using conventi onal as well as PMU measurements in recta ngula r coord inat es, is prese nted in [10]. Its main disadvantageis tha t measurement transformation from polar into rectangular coordinates will amplify the errors of PMU measurements. In [11], the effect of measurement asynchronicity on the state estimation accuracy and the effect of using PMU as well as SCADA measurements are analyzed. In [12] an iterative least square state estimator is for- 0378-7796/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.epsr.2011.03.013

1-s2.0-S0378779611000769-main

Documents