Note: Descriptions are shown in the official language in which they were submitted.
CA 02589635 2007-05-18
301-01 CA Patent
Power Restoration System for Electrical Power Network
Background
[001] A wide variety of services depend upon the reliable delivery of
electrical energy in
order to operate efficiently. Computers, traffic lights and a wide variety of
appliances all
rely upon external electrical energy provided by electrical power networks.
Consequently,
when the network in unable to provide electrical power a variety of
infrastructure problems
result.
[002] A wide variety of problems lead to a failure to deliver electrical power
in a
network. In order to provide a flexible solution that supports the bypassing
of non-
functional power lines it is beneficial to employ an electrical power network
with
substantial redundancy. Unfortunately, this redundancy often leads to
extremely complex
network topologies. The complexity of these topologies in turn leads to
difficulty in
identifying failed components within electrical power network as well as
difficulties in
returning power to customers that experience power failures.
[003] It would be beneficial to provide a simple solution that provides
alternative
network topologies to configurable medium voltage electrical mesh networks in
which a
medium voltage is typically in the range of 3 5kilovolts (kv) to lkv. Ideally,
such a simple
solution would be easily implemented and run on conventional computing
devices.
Further, it would be beneficial if the solution provided a suitable response
very quickly as
even brief disruptions to the electrical power systems in most cities
represent a significant
loss in productivity and a potential danger to its inhabitants.
Summary of Invention
[004] The invention supports a simple method of configuring mesh networks in a
robust
way that supports fault location and power restoration.
[005] In accordance with an embodiment of the invention there is taught a
method for
configuring an electrical power network comprising: providing an electrical
network, the
electrical network comprising: at least a first electrical power source; a
second other
1
CA 02589635 2007-05-18
301-01 CA Patent
electrical power source and, a set of electrical bridges, each of the
electrical bridges
supporting a conductive state and an other than conductive state; receiving
data associated
with the electrical network; determining a number of independent virtual
paths, C, in
dependence upon the received data; determining C different virtual paths; and,
when C> 1,
determining a first location for an electrical bridge in an open state;
determining a second
location for an electrical bridge in an open state; setting first and second
electrical bridges
of the set of electrical bridges to an open state in dependence upon the
determined first and
second locations for an electrical bridge in an open state.
10061 Additionally, the invention describes, a method for configuring an
electrical power
network comprising: providing an electrical network, the electrical network
comprising: at
least a first electrical power source; a second other electrical power source
and, a set of
electrical bridges, each of the electrical bridges supporting a conductive
state and an other
than conductive state; receiving data associated with the electrical network;
[007] mapping of some nodes having a node configuration matching a
predetermined
node configuration into other predetermined node configurations; determining a
number of
independent virtual paths, C, in dependence upon the received data and the
predetermined
node configurations; determining C different virtual paths; and, when C> 1,
determining a
first location for an electrical bridge in an open state; determining a second
location for an
electrical bridge in an open state; setting first and second electrical
bridges of the set of
electrical bridges to an open state in dependence upon the determined first
and second
locations for an electrical bridge in an open state.
Brief Description of the Drawings
[008] The invention is now described with reference to the drawings in which:
[009] Fig. 1 is a prior art electrical power network;
[0010] Fig. 2 is the prior art electrical power network of Fig. 1 with an
electrically isolated
faulty link;
[0011] Fig. 3 is a prior art network featuring a multiple branches;
2
CA 02589635 2007-05-18
301-01 CA Patent
[0012] Fig. 4 is a simple mesh network;
[0013] Fig. 5a is a node having more than three electrical contacts;
100141 Fig. 5b is a representation of Fig. 4a showing a plurality of nodes
each having three
electrical contacts;
[0015] Fig. 5c is an alternative representation of Fig. 4a different from that
of Fig. 5b;
[0016] Fig. 6 is a flowchart that outlines the method according to the first
embodiment of
the invention;
[0017] Fig. 7 shows the mesh network of Fig. 4 with virtual paths and
resulting open
bridge locations provided;
[0018] Fig. 8 shows the mesh network of Fig. 4 absent those links designated
as having
open bridges as described with reference to Fig. 7;
[0019] Fig. 9 shows the mesh network of Fig. 4 with an electrical isolated
faulty link;
[0020] Fig. 10. illustrates the mesh network of Fig.4 with a faulty segment
removed;
[0021] Fig. l la illustrates a simple mesh network;
100221 Fig. 11 b to 11 e illustrate mesh networks based upon the mesh network
of Fig. 11 a
after a failure of a link between two nodes.
Detailed Description of the Invention
[0023] It is well known and understood in the art that a short circuit to an
electrical ground
will act to absorb electrical power. When a consumer is provided electricity
from a same
source via two different but connected paths a short circuit in either path
will prevent the
delivery of electricity via either of the two paths. Thus, while it is
beneficial to have
redundant paths available, it is frequently not beneficial to make use of
redundant paths
until a conventional path that is experiencing a fault is electrically
isolated from the rest of
the network.
3
CA 02589635 2007-05-18
301-01 CA Patent
[0024] Referring to Fig. 1, an electrical power network 100 according to the
prior art is
shown. The network transfers electrical energy from a sources lOla and 101b to
consumers 102a to 102f. Each of the consumers 102a to 102f is connected to the
electrical
power network 100 via a set of breakers 103. The breakers 103 selectively
electrically
couple the consumers 102a to 102f to power lines 104a to 104g. In addition,
breakers 105a
and 105b are provided electrically proximate the sources lOla and lOlb. Since
the breaker
103c is shown in a non-conducting state it is clear that electrical energy
propagating along
power line 104c will not be permitted to propagate in power line 104d and vice
versa. In
this way, a ring topology is broken into two electrically isolated paths 106a
and 106b.
When a fault occurs in electrical path 106a the consumers 102d to 102f
associated with
electrical path 106b continue to receive electrical power. In addition, once
the fault in
electrical path 106a has occurred it is relatively easy matter to determine a
relative location
of the fault by opening the breakers and selectably closing the breakers. Such
techniques
are well understood in the art. In this instance, this technique benefits from
the fact that
the electrical energy propagates to any specific location within the network
via one and
only one path when the network is suitably configured. A person of skill in
the art will
appreciate that once the fault has been isolated, it is a relatively simple
matter to dispatch
technical professionals to reset some of the breakers 103 to provide power to
all the
consumers 102a to 102f while electrically isolating the fault.
[0025] Referring to Fig. 2, the electrical power network 100 of Fig. 1 is
shown with power
line 104b electrically isolated from the remainder of the electrical power
network 100. The
breakers 103 electrically adjacent power line 104b are shown in a non-
conducting state. In
order to ensure that consumers 102b and 102c receive power, the breaker 103c
is in the
closed position. Thus, consumers 102b and 102c are receiving power from source
101 b.
This change in state of the breakers 103a and 103c results in two new
electrical paths 106c
and 106d.
[0026] A person of skill in the art will also appreciate that other topologies
of electrical
grids are sufficiently simple that isolating faults within them is trivial.
For example,
4
CA 02589635 2007-05-18
301-01 CA Patent
referring to Fig. 3, a simple electrical network 300 in accordance with the
prior art is
shown. The network comprises a source 301, breakers 302, and consumers 303. As
will
be apparent to one of skill in the art, to the extent that electrical power is
only able to flow
in one direction from a specific source to a specific consumer it is a
relatively simple task
to determine a relative location of a fault between that source and that
consumer despite
the presence of multiple branches in the electrical network 300. Thus, a
direction of
electrical energy propagation is associated with each of the links. Arrows 304
indicate this
direction of energy propagation within the network 300.
[0027] Modern electrical power networks are typically designed as mesh
networks. Like
the simple network of Fig. 1, within a mesh network it is desirable to ensure
that a given
consumer is supplied by only one power source via an electrical path with a
clear direction
of electrical energy flow absent any redundant electrical paths.
Unfortunately, when a
fault occurs within a complicated mesh network, it is often very difficult to
generate a new
set of electrical paths that provides electrical power to all consumers while
avoiding the
detected fault within the mesh network. Some embodiments of the invention
support both
easy identification of faulty elements of a complex mesh network and
determining suitable
paths within complex suitably designed mesh networks. Embodiments of the
invention
support generating alternative electrical paths that bypass known faulty
network elements.
[0028] Referring to Fig. 4, a simple mesh network suitable for control in
accordance with a
first embodiment of the invention is shown. The network comprises: sources 401
a to 401 e
that are electrically coupled to the remainder of the network via nodes of
degree one and
junctions 402a to 402i that are described as nodes of degree three. The mesh
network
provides electrical energy to loads (not shown) electrically coupled to the
nodes. For the
purposes of the method nodes of degree two are not addressed initially. In
generating the
mesh, a person of skill in the art will appreciate that it is often the case
that a given
location, represented by a node, is often served by more than three links of
the mesh. For
example in Fig. 4, nodes 402b and 402e each have four links. Such a case is
shown in Fig.
5a. In such cases, the node is treated as multiple instances of a plurality of
nodes of degree
three as shown in Fig. 5b. Clearly, nodes degrees higher than four are also
possible
however such nodes are easily reduced to multiple nodes of degree three. A
person of skill
CA 02589635 2007-05-18
301-01 CA Patent
in the art will appreciate that it is often important to accurately represent
the configuration
of the nodes that are used in producing a node having a degree of four or
more.
Specifically, in many cases a node of degree four is designed by coupling two
nodes of
degree three. Thus, the case of Fig. 5a may be more accurately represented by
Fig. 5c than
Fig. 5b depending on how the connections within the node are disposed.
[0029] Referring to Fig. 6, a flowchart according to the first embodiment of
the invention
is provided. The method according to the first embodiment of the invention
involves
configuring a mesh network to produce a set of independent virtual paths where
the
individual independent virtual paths are used to determine a set of open
bridge locations.
The open bridge locations are then used to determine a set of electrical
circuits. When the
network is configured to support these electrical circuits, the operation of
the network is
simplified allowing fault location and power restoration schemes analogous to
electrical
networks described with reference to Fig. 1, Fig. 2 and Fig. 3. whose
operation is very
simple and well understood in the art. The method relies upon determining a
number of
independent virtual paths 601. The number of independent paths is given by the
formula:
[0030] Independent paths: C = (N+M) / 2
a. where N is the number of nodes of degree 1, and;
b. M is the number of nodes of degree 3.
[0031] Having determined the number of independent paths, the nodes of degree
1 are
arbitrarily chosen as being one of a virtual source and a virtual sink 602. In
accordance
with the method, each power network has at least one virtual source and one
virtual sink.
Clearly, complex mesh networks are likely to comprise a set of virtual sources
and a set of
virtual sinks. A set of C virtual paths are defined as flowing from virtual
sources of the set
of virtual sources to virtual sinks of the set of virtual sinks 603. Each of
the C virtual paths
is different from the other virtual paths and each of the virtual paths makes
use of a
segment that is not used by any other virtual paths. Further, the method
specifies that all of
the segments support at least one of the virtual paths. A single open bridge,
such as an
open circuit breaker or open electrical switch, is then provided for each of
the virtual paths
6
CA 02589635 2007-05-18
301-01 CA Patent
along a segment that is not used by the other virtual paths. A person of skill
in the art will
appreciate that in some cases the arbitrary selection of the virtual sources
and virtual sinks
has an effect on the location of the open bridges within the network. Thus, in
some cases,
should the method not provide a suitable result due to, for example, load
balancing
constraints, the method is optionally applied again with a different selection
of sources and
sinks.
[0032] In the case of the network of Fig. 4, nodes 402b and 402e are shown as
nodes of
degree four but treated as two nodes of degree three. Thus, in the network of
Fig. 4, there
are five electrical sources, n=5, and there are 11 nodes of degree three, m=3.
Thus the
number of virtual paths is (5+11)/2 or 8. Referring to Fig. 7, nodes 401 a and
401 c are
designated as virtual sources while nodes 401 b, 401 d and 401 e are virtual
sinks. Each of
the virtual paths 701 a to 701 h is shown by an arrow. The virtual paths 701 a
to 701 h begin
at a virtual source and end at a virtual sink. Each of the virtual paths 701a
to 701h makes
use of a segment that is suitable for an open electrical bridge. Suitable
locations for the
open electrical bridges 702a to 702h are shown as ellipses in Fig. 7.
[0033] Once the segments that support open bridges are determined, the mesh is
optionally
drawn as a set of simple, independent electrical power networks. The simple
rules
described by the first embodiment of the invention serve to generate a set of
simple circuits
in the mesh network. Referring to Fig. 8, the mesh network of Fig. 4 is drawn
with no
links shown between those nodes that are designated to have open bridges
therebetween.
When open bridges are disposed on those links that have been designated to
have open
bridges it is apparent that each of the nodes 402a to 402i receives electrical
power from a
single source. As previously described with reference to Fig. 1, locating a
fault within a
simple circuit is a relatively simple task and therefore, the method according
to the first
embodiment of the invention supports well known methods and systems for
detecting
faults. In the event that a portion of the network should fail, the failed
portion is easily
identified. Once identified, these portions are electrically isolated from the
electrical
power network by opening the appropriate electrical bridges. A person of skill
in art will
appreciate that fault location and power restoration methods described with
reference to
7
CA 02589635 2007-05-18
301-01 CA Patent
Fig 1 are applicable to nodes of degree two as well as portions of a network
that feature a
set nodes of degree two disposed sequentially.
[0034] Having isolated the faulty section of the electrical network, there are
a variety of
ways to proceed in order to restore power to nodes that are currently not
receiving
electrical energy. In accordance with the first embodiment of the invention,
electrical
power is returned to the nodes that are not receiving power by simply closing
one electrical
bridge between the nodes that are not receiving power with an adjacent node
that is
receiving power with the exception of adjacent nodes that are optionally
electrically
coupled by links that are known to be faulty. Clearly, in many cases there are
other
constraints such as load balancing that restrict the choice or which open
bridge to close. A
person of skill in the art will appreciate that such considerations are easily
weighed and
considered when choosing a suitable electrical bridge to close. Referring
again to Fig. 8,
in the event that the link electrically coupling nodes 402d and 402e fails it
is a simple
matter to locate the faulty link. Referring to Fig. 9, once the faulty link
902de is identified,
electrical bridges 903de and 903ed are opened thereby electrically isolating
the faulty link
902de from the remainder of the network. Having electrically isolated the
faulty link
902de, it is now desirable to restore power to nodes 402e and 402h. Barring
further
electrical faults, it is apparent that closing any one of the open electrical
bridges adjacent
the unpowered nodes 402e and 402h other those open electrical bridges
electrical isolating
the faulty link 902de provide electrical power to nodes 402e and 402h. In this
example the
bridges 902ef and 902fe are set to a closed state (conducting) in order to
restore electrical
power to nodes 402e and 402h. A person of skill in the art will appreciate
that links
connecting the nodes optionally comprises nodes of degree two disposed
sequentially
between 402d and 402e. When this is the case, a method according to the prior
art
described with reference to Fig. 1 is optionally carried out to isolate the
faulty portion from
the remaining portion of the link. In addition each of the nodes optionally
corresponds to a
consumer, a group of consumers or another electrical load.
[0035] In an alternative to the first embodiment of the invention, once the
electrical fault is
located, it is designated as supporting an open bridge. The virtual paths are
then generated
in a way that ensures that the electrical fault corresponds to a link with an
open bridge. As
8
CA 02589635 2007-05-18
301-01 CA Patent
an open bridge corresponds to a link that is not in use it is then a simple
matter to ensure
that no power is directed to the faulty portion of the mesh network by
electrically isolating
the faulty link. It should be noted that in some cases, ie specific mesh
network topologies
that experience a specific fault, this alternative to the first embodiment of
the invention
does not always generate a suitable solution.
[0036] In a second alternative to the first embodiment of the invention, once
the electrical
fault is located, it is functionally removed from the mesh network. Once the
link is
"removed" the process in accordance with the first embodiment of the invention
is applied
again. The removal of the faulty portion of the mesh network is will reduce
the value of C.
In some cases, ie specific mesh network topologies that experience a specific
fault, this
alternative to the first embodiment of the invention does not always generate
a suitable
solution. Referring to Fig. 10, the mesh network Fig. 4 is shown absent the
link electrical
coupling nodes 402d and 402e. Absent this link, the node 402d is now a node of
degree 2
and, as per the method according to the first embodiment of the invention is
removed from
consideration. Referring to Fig. 11 a, a simplified electrical network 1100 is
shown. This
network 1100 is shown with nodes 1103a to 1103h with open bridges 1101 a to
1101d
generated in accordance with the method of the first embodiment of the
invention. Using
the method of the second alternative of the first embodiment of the invention
when the link
1102 fails, the network is redrawn absent this link. By removing this link
nodes 1103a and
1103b become nodes of degree two and, in accordance with the method of the
first
embodiment of the invention, these nodes are removed. The resulting network
1110a is
shown in Fig. 11 b. As drawn in Fig. 11 b, the network 1110a will not provide
a solution
when the method according to the first embodiment of the invention is applied
to it. The
network is optionally redrawn by combining nodes 1103c and 1103f to form node
1103cf
as shown in Fig. 11 c. Referring to Fig. 11 d, the node 1103cd, being a node
of degree four
is reduced to two nodes 1103c and 1103f of degree three, thereby producing a
new node
configuration. This new node configuration is solvable using the first
embodiment of the
invention however, due to change in the node configuration only some of the
solutions
generated for the network configuration of Fig. 11 d are applicable to the
network of Fig.
l la. Clearly, solutions that specify an open bridge on links electrically
coupling node
9
CA 02589635 2007-05-18
301-01 CA Patent
1103b to node 1103f are not suitable as no such electrical coupling is
provided in the
network of Fig. l la. Similarly, there is no link electrically coupling 1103c
to 1103g.
[0037] A person of skill in the art will appreciate that there are certain
criteria typically
applied to generating suitable mesh network designs for electrical networks.
In order to
avoid difficulties associated certain network topologies, like the one
described with
reference to Fig. 11 b it is suggested that a library of predetermined problem
configurations
be generated. Referring to Fig. 11 e, a suitable solution to the network
configuration of Fig.
1 lb is shown with open bridges 1120a to 1120c. Thus, the node configuration
of Fig. 1 lb
is optionally associated with the predetermined open bridge solution of Fig.
11 e.
Similarly, other predetermined problem configurations are associated with at
least one
solution and stored. In this way, more complex network designs featuring
problem
configurations are optionally solved by reducing the problem configuration to
a
predetermined block with a predetermined solution, applying the method
according to the
first embodiment of the invention to the remainder of the mesh network, and
determining
suitable locations for open bridges by combining the solutions.
[0038] A person of skill in the art will appreciate that the method of the
first embodiment
of the invention is optionally carried out by a suitably programmed computer.
Further, the
mesh network provided with reference to the first embodiment of the invention
is intended
to support a simple example of the method. A person of skill in the art will
appreciate that
the method according to the first embodiment of the invention is applicable to
a wide range
of simple and complicated mesh network topologies. In addition, the method
according to
the first embodiment of the invention is sufficiently simple that it is
optionally executed by
a suitable processor disposed within a functioning node of the electrical mesh
grid. For
example, various companies produce remote terminal units (RTUs) that serve to
send and
receive information between a master computing system and the nodes of a mesh
network.
In some cases it is desirable to leave the configuration of a mesh network
under the control
of a master computing system however this does leave the mesh network
vulnerable to a
failure of the master computing system. In many cases the RTUs have a
processor and
memory that is suitable for carrying out a method according to the first
embodiment of the
invention. Thus, when an electrical grid network comprises RTUs with
sufficient
CA 02589635 2007-05-18
301-01 CA Patent
processing capability the network is optionally operated in an autonomous
fashion using an
automated fault detection and power restoration method according to the first
embodiment
of the invention. In this way, the configuration of the mesh network is
determined using
components of the mesh network thereby reducing the likelihood that a failure
of
equipment external to the mesh network will have an adverse effect upon the
network
itself. Clearly, different RTUs have different processing capabilities.
Although the
method according to first embodiment of the invention is believed to be easily
supported
by a variety of such processors the method has been tested using Motorola Tm
Moscad Tm
RTUs.
[0039] Embodiments of the invention presented herein are intended to support
medium
voltage networks, including medium voltage mesh networks. These networks
support the
transfer of electrical energy using voltage signals at 1 kvolt to 35 kvolts. A
person of skill
in the art will appreciate that various embodiments of the invention have
applications in
other fields. For example, the delivery of electrical power in a navel vessel
is often
critical. Even momentary disruptions of electrical power may leave a navel
vessel
vulnerable to enemy fire. The method according to the first embodiment of the
invention
is easily adapted by one of skill in the art to support power distribution
within a naval
vessel.
[0040] A person of skill in the art will also appreciate that the methods
according to the
invention are also useful in situations in which a node fails. Specifically,
if a node of
degree three should fail then it is recommended to isolate the node from the
remainder of
the mesh network by inhibiting a flow of electrical energy via any of the
three electrical
links associated with the failed node. Thus, each of the links is set to
inhibit electrical
signals thereby electrically isolating the failed node. A person of skill in
the art will
appreciate that a node of a degree that is higher than degree three is often
made up of a
plurality of nodes of degree three with a predetermined electrical
configuration. Clearly,
when a node of a degree higher than degree three fails it is often desirable
to understand
the actual configuration of the electrical interconnection within the node. In
some cases, it
is possible to continue operating a portion of the node while in others it is
not. Regardless
of the degree of the failed node it is desirable to electrically isolate the
failed portion of the
11
CA 02589635 2007-05-18
301-01 CA Patent
node from the remainder of the electrical network. Clearly, if a node of
degree one should
fail corresponding to a failure of an electrical power source then the failed
node is
electrically isolated from the remainder of the network. Thus, the failure of
a node of
degree one is very analogous to a failure of a link electrically coupled to
the node of degree
one.
[0041) Numerous other embodiments of the invention will be apparent to one of
skill in
the art without departing from the spirit and scope of the invention.
12
