Measurement loss effect on power system state estimation

Abstract

The main objective of this research work is to classify measurements units in optimal measurement environments in the power system network. Firstly, the use of singular value decomposition is used to find the optimal measurements placement. This is the optimum environment where each measurement unit is given weight upon its effect on the state estimation accuracy. Loss of the measurement unit can be due to the bad data received and the measurement is discarded or not telemetered due to the communication link error. In this case the respective relative error in state estimation is obtained by using MATLAB simulation. The higher weighted as considered to be critical and less weighted are considered non critical measurements are identified in the power system. In this paper related quantities that are more affected are identified.Index Terms -Measurement placement, accuracy, measurement weight (index), state estimation, critical measurements.

State estimation using extended SEMM in power system

“…Under observability criterion, it is required to have number of measurements, m greater than or equal to the number of states (n). For N bus system, considering a slack bus where phase angle of the bus is known, state variables becomes (2N-1), therefore m • 2N-1 [1], [3], [4], [5], [6], [10] and [14].…”

“…In [1], [3], [4], [5], [6] and [14] authors use iterative approach such as Newton-Raphsons' method to obtain the state estimation solution. Jacobian matrix and gain matrix have to be recalculated during iterations.…”

“…The accuracy of the estimated states is proportional to the number of measurement units and their locations in the power system network. It is assumed that all have been addressed using measurement unit placement algorithms to obtain the optimum measurement locations [5], [6], [10] and [14].…”