What is a State ?

 The complete “solution” of the power system is

known if all voltages and angles are identified at
each bus. These quantities are the “state
variables” of the system.

 The operation of the power system can be either

static or dynamic.
 Static state of operation means if the system is
operating under steady state or slowly varying
conditions.
 Dynamic state of operation considers major
changes in the system like, a fault, failure of any
equipment, sudden change in load, loss of
 Why state estimation ?
2

 Meters aren’t perfect.

 Meters aren’t everywhere.

 Very few phase measurements?

 SE suppresses bad measurements and

uses the measurement set to the fullest
extent.
 What is state estimation
3

 State Estimation is the process of assigning a value to an

unknown system state variable based on measurements
from that system according to some criteria.
 The process involves imperfect measurements that are
redundant and the process of estimating the system states
is based on a statistical criterion that estimates the true
value of the state variables to minimize or maximize the
selected criterion.
 Most Commonly used criterion for State Estimator in Power
System is the Weighted Least Square Criteria.
 State Estimation
4

 It originated in the aerospace industry where the basic problem

have involved the location of an aerospace vehicle (i.e. missile ,
airplane, or space vehicle) and the estimation of its trajectory
given redundant and imperfect measurements of its position and
velocity vector.
 In many applications, these measurements are based on optical
observations and/or radar signals that may be contaminated with
noise and may contain system measurement errors.
 The state estimators came to be of interest to power engineers
in1960s. Since then , state estimators have been installed on a
regular basis in a new energy control centers and have proved
quite useful.
 State Estimation contd.,
5

 In the Power System, The State Variables are the

voltage Magnitudes and Relative Phase Angles at
the System Nodes.

 The inputs to an estimator are imperfect power

system measurements of voltage magnitude and
power, VAR, or ampere flow quantities.

 The Estimator is designed to produce the “best

estimate” of the system voltage and phase angles,
recognizing that there are errors in the measured
quantities and that they may be redundant
measurements.
 State Estimation contd.,
6

 To obtain the best estimate of the state of the

system based on a set of measurements of the
model of the system.

 The state estimator uses

 Set of measurements available from PMUs
 System configuration supplied by the topological
processor,
 Network parameters such as line impedances as
input.
 Execution parameters (dynamic weight-
adjustments…)
 State Estimation contd.,
7

 The state estimator provides

 Bus voltages, branch flows,…(state variables)

 Measurement error processing results
 Provide an estimate for all metered and
unmetered quantities.
 Filter out small errors due to model
approximations and measurement
inaccuracies;
 Detect and identify discordant measurements,
the so called bad data.
 Objectives of State
8
Estimation
 To provide diagnostics for modeling and
maintenance.
 To provide a consistent representation for power
system security analysis.
 On-line dispatch power flow
 Contingency analysis
 Load frequency control
 To provide a view of real-time power system
conditions.
 Real-time data come from SCADA
 SE supplements SCADA data: filter, fill & smooth
 Block diagram of State
9
estimation
 Classification of State
10
Estimation
 State Estimation also has different
approaches based on application of the
algorithms such as Conventional SE,
Distributed SE or Multi Area State
Estimation(MASE).

 SE algorithms are broadly classified into

Static State Estimator (SSE) and Dynamic
State Estimator (DSE).
 Static State Estimator
11

 Static state estimation assumes the power system to

be quasi-static, which implies the system state stays
constant between the two consecutive updates of the
state vector. Therefore, to improve real time
monitoring of the grid the updates need to be
performed within shorter intervals. The most widely
used form of SSE in industry is the weighted least
square (WLS) method.

 Static SE also includes the Sequential SE which has

the advantage of being able to perform updates with
partial measurement set. These sequential SE
addresses the problem with bad data and loss of data.
 Dynamic State Estimator
12

 Dynamic State Estimation (DSE) is a step closer to real

time monitoring when compared to SSE. There are two
parts to the DSE: Prediction and Filtering.
 The prediction of the state variables involves the
modeling of the power system behavior. The prediction
is calculated based on a mathematical model which
takes into account the nonlinearities of the
measurement functions unlike in SSE where a linear
model is developed for the ease of computation.
 The power system incorporating nonlinearity of the
power flow equations increases the computational
expense of the DSE and therefore is not widely
implemented in reality.
 Dynamic State Estimator
13
contd.,
 The second part of a DSE includes the filters
used to filter out the bad data by combining the
predicted information along with the actual
measurements on arrival.

 The Extended Kalman Filter is widely used to

perform this filtering. There are several other
classical methods which are provided in next
slides.
 Power system state
14
estimation
 The only thing we know about the power system comes
to us from the measurements so we must use the
measurements to estimate system conditions.
 Measurements were used to calculate the angles at
different buses by which all unmeasured power flows,
loads, and generations can be calculated.
 We call voltage angles as the state variables for the
three-bus system since knowing them allows all other
quantities to be calculated.
 If we can use measurements to estimate the “states” of
the power system, then we can go on to calculate any
power flows, generation, loads, and so forth that we
desire.
 State Estimator: determining the best state
15

 We need to generate the best guess for the state

given the noisy measurements we have available.
 This leads to the problem how to formulate a
“best” estimate of the unknown parameters given
the available measurement.
 The traditional methods most commonly
encountered criteria are
 The Maximum likelihood criterion
 The weighted least-squares criterion.
 Non traditional methods like
 Evolutionary optimization techniques like Genetic
Algorithms, Differential Evolution Algorithms etc.,
 Bad Data
16

 One of the essential function of state estimator is

to identify measurement errors, and to identify and
eliminate them if possible.

 The measurement errors are generally called as

Bad data which leads to wrong state estimate and
imperfect control actions

 The measurement errors are generally classified as

 Random Errors

 Large Measurement Errors

 Topological Errors
 Bad Data contd.,
17

 Random Errors usually exists in measurements due

to finite accuracy of measuring devices and
telecommunication medium.

 Large Measurement Errors can occur when the

measuring devices has biases, drifts and wrong
connections. Telecommunication system failures or
noise caused by unexpected interference also leads to
large deviation in measurements.

 Topological Errors occur when the information given

to operator at the control centre regarding topology of
the network and placement of measuring devices.
 Bad Data contd.,
18

 Depending on the severity of errors introduced into

tele-metered measurements, they have been classified
in to following three categories:

measured value and the true value exceeds 20𝜎; 𝜎 being the
 Extreme error: where absolute difference between

standard deviation of measurements.

measured value and the true value is between 5 𝜎 and 20 𝜎 .

 Gross error (Bad data): where absolute difference between

value and true value is less than 5𝜎.

 Normal error: where absolute difference between measured

 These errors are expected to be filtered by the state

estimator and the filtering action depends on the type
of estimator used.