Decentralized Dynamic State Estimation in Power Systems Using Unscented Transformation

Abstract

Abstract-This paper proposes a decentralized algorithm for real-time estimation of the dynamic states of a power system. The scheme employs phasor measurement units (PMUs) for the measurement of local signals at each generation unit; and subsequent state estimation using unscented Kalman filtering (UKF). The novelty of the scheme is that the state estimation at one generation unit is independent from the estimation at other units, and therefore the transmission of remote signals to a central estimator is not required. This in turn reduces the complexity of each distributed estimator; and makes the estimation process highly efficient, accurate and easily implementable. The applicability of the proposed algorithm has been thoroughly demonstrated on a representative model. Index Terms-dynamic state estimation, decentralized, phasor measurement units, unscented transformation, Kalman filter.

Decentralized control of oscillatory dynamics in power systems using an extended LQR

“…The dynamic equation of is given by (4) After incorporating in in (3), re-terming as , and replacing the pseudo-input with its time derivative in p.u., , (3) gets redefined to (5) The nonlinear equation given by (5) needs to be linearized before it can be used in a linear controller. Linearizing (5) about an operating point gives (6) where , , Appendix A gives the details of the DAEs in (5) and the matrices in (6). It should be noted that after introduction of the new state , neither nor is present in (5) (which can be verified in Appendix A).…”

“…As it is understood that (5) and (6) are in continuous form for the th generating unit for time and the partial derivatives are evaluated at time , the variables , , , and can be dropped from (5) and (6) without causing any ambiguity. Thus (5) and (6) are written in simple form as (25) The vector consists of eight functions corresponding to the eight states, A preliminary modification needs to be done in the system given by (8) for the derivation of Theorem 1, by adding a constant pseudo-input at the end of the column vector , as (26) Also, , where , and (27) …”

Controlling the renewable microgrid using semidefinite programming technique

“…As the NBP computes the mean and covariance of state variables in a distributed way, so it provides better estimation performance compared with the EKF. Interestingly, in order to avoid the calculation of Jacobians, the UKF based power system state estimation is explored in [16,22,23] and it shows that the UKF preserves high-order estimation accuracy compared with the EKF. This improvement is due to the fact that the UKF calculates the mean and covariance of state variables accurately that undergo a nonlinear transformation [24,23].…”

Sensor Transmission Power Schedule for Smart Grids

“…Case study and simulation IEEE 39-bus system, which is also know as the 10-generator New England system, is wildly applied to evaluate the estimation performance of smart grid [8,10]. Fig.…”

“…Amount of research on RTSE has been developed based on the observation data of PMU [7][8][9][10][11][12]. Due to the nonlinearity of power system, nonlinear filter is wildly applied in RTSE.…”