Skip to Main Content
In this paper it is shown that measurement errors or parameter errors appear on the measurement residual of the weighted least squares state estimation process in a very similar way. Based on this fact this paper proposes an approach to detect, identify and correct series branch parameter errors using a geometrical view. In the paper simulations only parameter errors are assumed, then firstly it is tested if the measurement set have a gross error, that is, if the transmission line parameters have an error, using for that purpose the measurement composed error (CME). The branches of a line i-j are suspicious of having errors if the measurements CMENs, directly related to that branch, are superior to the assumed gross error detection test threshold value (TV). Then among the measurements suspicious of having error, the one with largest CMEN will have more chance of having the parameter error. The correction to be made on the line parameter is the corresponding measurement CNE. In this proposed approach one measurement snap shots only is used. To test the efficiency of the proposed approach the IEEE 14-bus system has been used. Simulations results have demonstrated the reliability of the proposed approach to deal with parameter errors.