A geometrical view for multiple gross errors detection, identification, and correction in power system state estimation

Abstract

In this paper it is described a geometrical approach to detect, identify, and recover multiple gross errors in power system state estimation. Using the classical WLS estimator the measurement residuals is computed, and then the error is composed. For the detection and identification of the measurements with gross errors the composed measurement error in the normalized for (CMEN) is used. The measurement magnitude corrections otherwise are performed using the composed normalized measurement error (CNE). To give support to the detection and identification of the measurements containing gross errors, a version of the largest normalized composed error test is provided. The IEEE-14 bus system as well as the reduced 45-bus Power System of the Brazil South is used to test the multiple gross error detection, identification and correction efficiency.

Ensemble CorrDet with adaptive statistics for bad data detection

“…Later on, the authors of [33] expand upon the chi‐square test, using the normalised residual within the test itself and showed significant improvement. Another significant improvement upon the chi‐square test came along when the geometrical interpretation of the measurement residuals was explored [34]. With this geometrical view of the PSSE process, it was discovered that errors and residuals were in different mathematical spaces.…”

Cyber‐physical robust control framework for enhancing transient stability of smart grids

“…To estimate the error, one needs also toestimate the error component . With that purpose, the innovation of ameasurement, related to the other measurements, is defined as the information itcontains, and not the others measurements of the measurement set [23]. This definition suggests that theinnovation is contained in the portion of the measurement that is independent ofthe other measurements of the system, i.e.…”

“…Since the residual and the other error component are orthogonal to each other, it ispossible to compose the ME vector; that is, for the i thmeasurement This error vector is called composed ME. To find the maskederror and compose the measurement's total error, it is used the innovation index(II), as proposed by [23] A measurement with low II indicates that a large component ofits error is not reflected in its residual as obtained by the classical WLS‐SE.Consequently, even when those measurements have gross errors, their residualswill be relatively small. Knowing that the vector spaces and are orthogonal to each other, then it ispossible to recalculate the composed error of the i thmeasurement.…”

Hybrid method for power system state estimation

“…As the measurements can have important deviations from real values, it is important to include a bad data detection procedure in a state estimation method. There are several mathematical methods to detect presence of bad data [11, 12]. Considering its simplicity, in this work this task is performed by the χ 2 ‐test [13, 14].…”