Abstract:
Adaptive protection of electric power systems is a new and a
revolutionary idea. It is still in its conceptual development stage. At
present, research work is going on to develop it as a protection philosophy
and to identify its vital components. A few research papers are available
in the protection literature on the subject, which highlight the importance
and the implementability of this protection philosophy. All the researchers
stress the need for this concept as it has much to offer in terms of
reliability, dependability, security, efficiency, innovation and economy.
Since the concept of adaptive protection is new, it is appropriate
to give it a definition as visualized by some authors. Some of the
definitions are :
(A) The adaptive protection is an on-line activity that modifies the
preferred protective response to a change in system conditions or
requirements. It is usually automatic, but can include timely
human intervention.
(B) The adaptive protection is a protection philosophy which permits
and seeks to make adjustments to various protection functions in
order to make them attuned to prevailing power system conditions.
(C) The adaptive protection refers to the ability of the protection
system to automatically alter its operating parameters in response
to changing network conditions to maintain optimal performance.
The key concept is to change something in a protection system in
response to changes in power system caused by operational or structural
disturbances.
-iThe
need for the development of adaptive protection arose because
of certain weaknesses and shortcomings in the existing protection
philosophy. Researchers have been realizing these shortcomings for some
time now and have boldly proposed a new way of thinking to tackle the
problems of protection. One way of thinking is, why not to make the
protection relays adaptable to the system changes so that the relay would
respond to new operating states in aself-adaptive manner, thereby reducing
the intervention of the operators to a minimum and increasing the efficacy
of the protection functions.
One of the most laborious and time-consuming tasks for a
protection engineer is to set and coordinate protective relays in an
interconnected power system. Ideally these settings are reviewed whenever
loads or other system conditions change enough to appreciably alter the
fault currents. Settings are also reviewed prior to temporarily taking a
line out of service and as a part of post-fault analysis. However, even a
review of settings is a troublesome task. Hence settings are not changed as
often as conditions warrant and protection system performance is degraded.
This problem was tackled by using computer aided approach in an
off-line mode. Based on certain coordination criteria, the settings of
relays are computed. In this way the system responds to faults and abnormal
conditions in a pre-determined manner. It is not only difficult to identify
and analyze all the operating conditions in advance, it is also impossible
to determine relay settings that would be optimum for all abnormal and
normal operating conditions.
There could be a better and efficient method to set and coordinate
the protective relays. The method is " Adaptive Coordination of Protective
Relays". In this method the relays could respond to the changing system
-11-
conditions and adapt according to the new conditions. The changing
conditions could be operational or structural. The requirements for such a
scheme to be implemented include the right kind of efficient algorithms,
appropriate digital relays and communication channels. The biggest
advantage of this scheme would be to liberate the protection engineer from
.the most tedious and time-consuming task of setting and coordinating
protective relays.
This dissertation aims at the development of the concept and the
philosophy of the adaptive coordination of relays in the general framework
of adaptive protection.
The adaptive coordination philosophy is implemented using the
concept of overcurrent relaying. The directional overcurrent relays in an
interconnected power system are suitable relays to be exploited for the
introduction of the adaptive concepts.
There are two coordination philosophies available for the
coordination of directional overcurrent relays. One is based on the
automation of the traditional interactive algorithms and the other is based
on the use of parameter optimization techniques. The former technique is
the result of traditional relay coordination philosophy implemented with
the help of computers. The protection engineer interacts with the computer
and tries to find out the solution to the coordination problem. His
experience and knowledge influences the coordination results. Whereas, in
the later method a performance function is optimized subject to certain
coordination criteria. The coordinated relay settings obtained are the
optimal relay settings which satisfy a wide variety of conditions which
arise due to different system conditions. The performance function chosen
reflects the desired operation of the relays.
In
the work reported in this dissertation, coordination problem is
formulated as a parameter optimization problem with a different
characterization which is suitable for adaptive coordination. The solution
methods employed efficiently fit in the overall coordination philosophy.
The traditional coordination algorithms need break point set. This
requires elaborate schemes for topological analysis of the network. The
break point set is needed to decrease the number of iterations in the
coordination process. It is established, in the dissertation, that if the
optimization techniques are used for relay coordination, this removes the
need of determining break points of the network. Thus eliminating the use of
elaborate topological analysis programs.
A simple but efficient method is proposed to determine the
backup/primary relay pairs by using the linked-list type of data structure
known as LINKNET. The backup/primary relay pairs are automatically
generated for the close-in as well as the far-bus fault currents.
Two integrated software packages OPCORD and OPCON have been
developed for coordinating the directional overcurrent relays of an
interconnected power system. Different software packages like load flow
analysis, fault analysis, determination of backup/primary relay pairs, main
coordination program, etc. have been integrated. The programs could be run
as interactive packages or as off-line 'batch' processing packages,
depending on the choice of the user. The information needed for the
coordination is stored and used by the software. No data base management
packages are needed to manage huge data generated by the program. An
efficient method has been developed to generate, store and retrieve the
data as per the requirements of the coordination. Data generation,
updating, retrieval and use are all combined in the same program.
The
problem of sympathy trips of relays in an interconnected power
system has been studied in detail. A sympathy trip of a relay is the
tendency to operate for a fault which is otherwise of no concern to this
relay. The various reasons which give rise to the sympathy trip tendency of
a relay have been classified into different categories. It is concluded
that the use of adaptive coordination would remove the sympathy trip
tendency of certain relays which are caused due to a change in the system
conditions.
A full system adaptive optimal coordination algorithm has been
developed and tested successfully for different test power transmission
systems. The program continuously monitors the system for any disturbance,
structural or operational. If a disturbance is detected, the program is
invoked immediately. Nothing is assumed apriori. The algorithm calculates
everything according to the present conditions of the network. No assumed
contingencies are considered.
The determination of the pickup current settings and the time
multiplier settings are the essence of any overcurrent coordination
process. The pickup current value is normally fixed before the coordination
is attempted. This procedure does not ensure correct setting of the relays.
Therefore, it is proposed to dynamically vary it as the load current
varies. This step would make the relays more sensitive to the
system changes. Therefore, an expression for pickup current setting has been
developed which dynamically incorporates any change in load current into
the new value of pickup current.
Due to large number of interconnections and ever growing demand,
the
size and complexity of the present day power systems, have increased
tremendously. Therefore, it is becoming difficult and time consuming to
solve the coordination problem of large and complex power networks. Hence
there is a need for an efficient decomposition technique for solving the
coordination problem of large power networks. Thus an efficient
decomposition technique for the optimal coordination of directional
overcurrent relays of large power networks has been developed and
successfully implemented. The large power network is decomposed into number
of sub-networks called blocks by identifying boundary buses, branches and
relays. The blocks are analyzed sequentially and the boundary data table is
updated during the solution process. All the components of the software are
fully integrated. It is observed that the larger the number of blocks,
lesser is the overall execution time for the solution of the coordination
problem.
An adaptive optimal coordination algorithm has been developed
which is based upon the decomposition approach. The relay settings are
determined in an adaptive manner, block-wise. Considerable execution time
and the memory requirements are saved if the decomposition approach is used
for the coordination process.
A new concept of local optimal coordination has been developed and
reported in this dissertation. A local disturbed region is identified along
with all the backup/primary relay pairs in response to a structural change
or an operational change in the power transmission system. The coordination
problem is solved only for this local region using local optimization. The
methods developed do not employ any data base management packages or any
elaborate topological analysis programs.
-viThe
concept of local optimal coordination has been used to develop
an adaptive coordination philosophy. This method is an ideal method to
solve the coordination problem of the affected part of the network only, in
an on-line manner. Only a small portion of the power network is identified
and its coordination problem is solved efficiently. The time taken for
execution is negligible in comparison with the full system coordination
method. The memory requirements are also drastically reduced as a small
portion of the network is dealt with.
During the progress of the present work, it was realizedlthat the
concepts of control theory can be easily applied to the problem of relay
coordination. In future coordinated control and protection schemes are
sure to be developed and implemented. Therefore, we have formulated the
classical coordination problem as an adaptive control systems engineering
problem. It is made clear that this is a closed-loop, non-linear,
multidimensional control system with variable parameters and intermittent
control. The concept of adaptive relay coordination control system is
presented and discussed. The structure of the main component of this
control system, i.e., the adaptive relay coordination controller has been
given. Other components of the control system are also presented and
discussed.