Note: Descriptions are shown in the official language in which they were submitted.
CA 02549087 2006-05-31
1 "ELECTRICAL PROFILE MONITORING SYSTEM FOR DETECTION OF
2 ATYPICAL CONSUMPTION"
3
4 FIELD OF THE INVENTION
Embodiments of the invention relate to systems for monitoring
6 usage of utilities, such as electrical, for alterations in normal patterns
of
7 consumption of utilities and, more specifically, to a system of detection of
8 patterns indicative of theft of electrical utilities, such as in the indoor
cultivation of
9 marijuana.
11 BACKGROUND OF THE INVENTION
12 It is estimated that electrical theft alone results in millions of dollars
13 of loss per year. Ontario, Canada reports an estimated cost of $500 million
14 dollars per year. It is believed that the most significant contributor to
electrical
theft is the indoor cultivation of marijuana known as grow operations or "grow-
16 ops". While recent case law has made the recovery of lost energy revenue
17 possible, the reality is that very little monetary recovery is made through
court
18 actions.
19 Along with the staggering costs related to the electrical theft, there
are additional costs to the community which include property damage, increased
21 potential for fires due to wiring required to tap into the grid, electrical
brown-outs
22 and power outages due to blown transformers.
23 Typically, relatively new single family residential properties having
24 underground power lines are targeted as sites for grow-ops. The electrical
power
lines are readily bypassed and are more suitable than older services to
provide
26 the sustained amperage, typically a 120 amp draw that is required for a
large
CA 02549087 2006-05-31
1 commercial operation. Older services typically provide only 60 amp or 100
amp
2 overhead lines which are more susceptible to malfunction.
3 Monitoring of electrical services at a location in the service which
4 would detect bypassing of conventional metering is uncommon. Applicant is
unaware of systems currently in use which are capable of economically
6 identifying atypical usage patterns at the primary level and thereafter
pinpointing
7 specific households which may be of interest to the utility providers and to
law
8 enforcement.
9 Systems are known to monitor consumption at secondary lines
which feed electricity from the transformer to the residence which are capable
of
11 detecting over-usage, being typically in excess of a predetermined value,
such
12 as about 50% of the expected for a single household. Over-usage due to a
grow
13 operation or the like at the primary level however becomes more difficult
to
14 detect as the over-usage is typically not seen as a significant alteration
in
measurement using conventional metering.
16 There is great interest in systems which can be used to identify
17 uncommon consumption patterns, at the primary level, which may be
indicative
18 of utility theft and which do not infringe upon existing laws which protect
19 individual rights and freedoms.
2
CA 02549087 2006-05-31
1 BRIEF DESCRIPTION OF THE DRAWINGS
2 Figure 1 illustrates a conventional electrical grid system which
3 provides electricity to a plurality of residential properties;
4 Figure 2 is a perspective view of an underground electrical pull box
which supplies electricity through a primary power line;
6 Figure 3 is a rolled-out elevation schematic according to Fig. 2;
7 Figure 4 is a rolled-out top view according to Fig. 2;
8 Figure 5 is a top view according to Fig. 2;
9 Figure 6 is a schematic of a system according to an embodiment of
the invention for monitoring consumption of electricity at the primary line;
11 Figure 7a is a graphical illustration of amperage results from a
12 primary line meter without a grow operation;
13 Figure 7b a graphical illustration of amperage results from a
14 primary line meter with a grow operation;
Figure 8 is a schematic of the system according Fig. 5 comprising
16 determination of heat patterns indicative of consumption of electricity at
a
17 plurality of transformers fed from the primary line and which, in turn,
each feed a
18 plurality of residences;
19 Figure 9 is a schematic of the system according to Figs. 5 and 6
further comprising identification of consumption patterns at each of a
plurality of
21 individual residences connected to each of the transformers to identify a
22 residence of interest;
23 Figure 10 is a schematic illustrating a meter suitable for measuring
24 amperage in increments of about 0.1 amp to about 0.01 amp and transmitting
said measurement via a digital modem to a central computer;
3
CA 02549087 2006-05-31
1 Figure 11 is a schematic illustrating an electrical grid system for a
2 specific residential community having 37 active residences monitored
according
3 to an embodiment of the invention as discussed in Example 3 herein for the
4 existence of a grow-op; and
Figure 12 is a graphical representation of digital recording ammeter
6 output from a primary line according to Example 3 disclosed herein.
7
8 SUMMARY OF THE INVENTION
9 Embodiments of the invention provide a method of detecting
atypical consumption patterns which when compared to known patterns of
11 consumption are useful in identifying electrical losses or theft, such as
by
12 marijuana grow-ops. Use of a meter having a resolution capable of detecting
13 suspect usage patterns on the primary line permits monitoring of
consumption
14 patterns without the need to access private property and which cannot be
bypassed which is typically the case with individual residence metering and
16 grow-ops.
17 Therefore in a broad aspect of the invention a method for detection
18 of atypical electrical consumption patterns in an electrical system having
a
19 primary supply line supplying electricity to a plurality of transformers
and wherein
each transformer supplies electricity to a plurality of consumers through a
21 plurality of secondary lines, comprises: metering the primary supply line
at
22 predetermined time intervals for establishing data indicative of patterns
of
23 consumption; comparing the patterns of consumption to known consumption
24 patterns for identifying suspect consumption patterns, and when a suspect
consumption pattern is identified, monitoring characteristics of the plurality
of
4
CA 02549087 2006-05-31
1 transformers for identifying a suspect transformer from the plurality of
2 transformers; and load testing at least one of the plurality of secondary
lines from
3 the suspect transformer to each of the plurality of consumers.
4 In one embodiment the meter is a digital recording ammeter having
a resolution of less than 1 amp and preferably from about 0.01 amp to about
0.1
6 amp.
7 In one embodiment, once the suspect consumption pattern has
8 been identified at the primary line, the utility or other monitoring agency
is
9 notified and the responsible agency monitors the characteristics of the
plurality of
transformers. Preferably, the characteristic monitored is the heat signature
of the
11 transformer which is typically measured using infrared laser technology
such as
12 with an infrared laser.
13 The suspect consumption pattern is compared to known
14 consumption patterns from data collected over a period of time sufficient
to avoid
detecting only peak consumption levels. In one embodiment, a threshold
16 consumption level is established, such as about 25% above peak consumption
17 levels, to avoid falsely identifying peak consumption periods as being
suspect.
18
5
CA 02549087 2006-05-31
1 DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT
2 As shown in Fig. 1, in residential electrical services, electricity is
3 typically provided by a primary electrical line 1 extending from a pull box
6 and
4 which provides power to a plurality of transformers 2 which in turn feed
electricity
through pairs of secondary lines 3 to each of a plurality of consumers 4,
typically
6 structures or residences, connected thereto. Consumption is typically
measured
7 at conventional meters 7 which are connected to the secondary lines 3
feeding
8 each residence. Conventional meters 7 are readily bypassed 8 such for
electrical
9 theft (Fig. 6). In newer neighbourhoods, the primary line 1 is an
underground line
which is connected from a source (not shown) to a plurality of underground
pull
11 boxes 6 which each feed a number of transformers 2. Typically, each pull
box 6
12 might feed 8 to 10 separate transformers 2 which in turn each feed 8 to 12
13 residences 4. In older neighbourhoods, the primary line 1 is an overhead
line.
14 For the purposes of the following description, embodiments of the invention
are
discussed in the context of an underground system. As one of skill would
16 understand, the embodiments of the invention are also equally applicable to
17 overhead services.
18 Having reference to Figs. 1-5, each pull box 6 provides electricity to
19 the plurality of transformers 2, which reduces the supplied voltage,
typically from
13000V or 25000V down to 220V and each transformer 2 then provides
21 electricity through the secondary lines 3 to the plurality of residential
properties
22 or structures 4. In a typical urban setting, each pull box 6 might service
about
23 100 residential homes 4, with about 10 homes 4 connected to each of 10
24 transformers 2.
6
CA 02549087 2006-05-31
1 As shown in Fig. 6, in an embodiment of the invention, a system for
2 detection of line losses or utility theft comprises a meter 10, such as
shown in
3 Fig. 10, connected to the primary electrical line 1 which feeds each
transformer 2
4 connected thereto to monitor for fluctuations in consumption (Figs. 7a and
7b)
relative to a known mean or average consumption which has been determined
6 for a particular area. Depending upon the type of transformer 2 and the
voltage
7 provided through the primary line 1, the average consumption can generally
be
8 predicted and can be verified using historical consumption records by the
utility
9 provider. Applicant is aware that an average consumption per household for a
commonly used transformer 2 in Calgary, Alberta, Canada might be about 1-2
11 amps at 110 VAC when measured on one line or about 2-4 amps at 110 VAC
12 when measured on both lines.
13 Preferably, the meter 10 provide data for determining measures
14 indicative of electrical consumption, typically electrical consumption in
amperage
(amps) and, most preferably, has a resolution capable of measuring
16 consumption at less than 1 amp increments. Preferably, the meter measures
17 consumption in about 0.1 amp increments or even at higher resolutions of
about
18 0.01 amp increments. Higher resolution monitoring assists in identifying
19 aberrations in consumption at the primary level where small changes may
result
where a single grow-op is tied into the system.
21 In a preferred embodiment, as shown in Fig. 10, the meter 10
22 comprises a digital recording ammeter M enclosed in a water resistant PVC
23 casing V which only actuates at predetermined time intervals, such as every
15
24 minutes, to take a reading and collects a number of interval readings over
a
predetermined time. Thus, a low-power battery system, such as a high capacity
7
CA 02549087 2006-05-31
1 lithium battery B can be used. The meter 10 is connected to the primary line
2 through a wire connection P. The collected data can be stored in the meter
10
3 for processing. Some processing can be performed on site or alternatively
the
4 data can be transmitted at predetermined intervals to a processor 5 for
comparison to the known consumption patterns. The meter 10 is preferably
6 equipped with a digital transmitter D such as a modem and a communication
7 antenna C. The processor 5 can be a central computer, such as at the
utility, and
8 the transmission is by means such as wireless technology. Primary lines 1
which
9 exhibit an increase in consumption which exceeds an established threshold
over
a specified period of time (Fig. 7b) are flagged as being of interest or
suspect.
11 Having reference to Fig. 8, once a suspect consumption pattern
12 has been identified at the primary line 1, characteristics, such as a heat
13 signature, are monitored at all of the transformers 2 fed by the suspect
primary
14 line 1. In the case of heat signatures, infrared laser technology is
typically used
on all of the transformers 2 which are fed by the primary line 1 from the pull
box
16 6 of interest. The transformers 2 are compared to other transformers 2
having a
17 similar draw from the same primary line 1 and the transformer 2 which
exhibits
18 an excessive heat signature, indicating an unexpected draw, is flagged as
being
19 suspect. The infrared laser is thus capable of narrowing the number of
residential properties or structures 4 of interest to only those which are fed
by
21 one or more suspect transformer(s) 2.
22 As shown in Fig. 9, having located a transformer 2 of interest, load
23 testing is performed on the plurality of secondary service lines 3 within
the
24 transformer 2 to identify the suspect service or structure 4s. Once the
suspect
structure 4s is identified, appropriate action can be taken.
8
CA 02549087 2006-05-31
1 A further advantage of the system, as described above, is
2 detection of power outages, which may occur for any reason, often unrelated
to
3 grow operations. Typically, utilities personnel are unaware of power outages
4 unless notified by the affected individuals or as a result of random sweeps
performed by Utility Trouble Services personnel who may notice an absence of
6 street lights or the like. Interval monitoring of all primary lines quickly
alerts the
7 system to a complete drop in draw from a primary line consistent with a
power
8 outage.
9 An additional advantage of the system to the utility is the ability to
accumulate usage data to reconcile with retailer usage data from metered
sites.
11 Further, line losses can be identified and rectified, where reconciliation
data
12 illustrates a loss which is not necessarily related to a grow-op.
13 New smart meter technology is rapidly being introduced to the
14 industry to facilitate time-of-use metering at each residence, permitting
utilities to
charge for electrical usage dependent upon the time of use and for consumers
to
16 take advantage of times at which a lower cost is assessed to the use of
17 electricity. The combination of smart metering at each residence and
metering at
18 the primary line, using a system according to embodiments of the invention
19 disclosed herein, provides significant improvement in the collection of
data for
reconciliation and identification of losses, including the detection of line
loss such
21 as through faulty overhead or underground wires etc. Simply, the load
provided
22 at the primary line should be equal to the sum of all the consumptions
measured
23 at each residence, having consideration for known factors of line loss. A
24 discrepancy signals a problem with some part of the line which can be
located
using the present invention or other means.
9
CA 02549087 2006-05-31
1 IN USE
2 In use, as shown in Figs. 6-9, embodiments of the invention are
3 typically implemented by positioning a digital recording ammeter 10 on each
4 primary line 1 in an electrical grid. This minimizes the number of meters 10
by
about 10-fold over conventional transformer-based metering and 100-fold over
6 residence by residence metering. The meter 10 is typically powered by a high
7 capacity lithium battery and is programmed to actuate at specified
intervals, such
8 as every 15 minutes, to take a reading of the primary line amperage. Subsets
of
9 the ongoing data are stored at the meter 10 and after several readings are
taken,
such as readings collected over the period of an hour, the subset of data is
11 transmitted, such as by a wireless technology to the centralized processor
5 or
12 computer for comparison of the readings to predetermined thresholds for the
13 primary line consumption. The meter 10 is located off property such as in a
14 laneway in compliance with local, provincial and federal laws to avoid
violation of
the rights and privileges of the property owners.
16 The main computer system 5 analyzes and compares the
17 amperage readings for each meter 10 and identifies primary lines 1 having
18 unusually high draws. In a preferred embodiment of the invention, excessive
19 draws from the primary line 1, which exceed a predetermined threshold over
a
predetermined period of time sufficient to eliminate spikes due to non-
criminal
21 activities, are flagged as suspect. The predetermined threshold is
established
22 through historical consumption records maintained by the utility and is set
to
23 accommodate normal peak high usage periods. Preferably the predetermined
24 threshold is established having a buffer, such as about 25% over and above
the
peak high usage, to avoid incorrectly flagging high usage non-grow-op
services.
CA 02549087 2006-05-31
1 Further, typical expected consumption rates may also be predicted by experts
2 taking into consideration the number and size of the residences 4, the type
of
3 transformer 2 and the voltage supplied by the primary line 1.
4 As shown in Table A, examples of hypothetical comparisons or
ratios of amperage are made between average dwellings and grow operations
6 where the primary line exiting the pull box 6 supplies 10 transformers, each
of
7 which in turn supply 100 residential dwellings. The results are illustrative
of the
8 type of increase expected.
9 TABLE A
Residential Total Total amps
Community Residential at primary
Community line
Example 1
Average draw over a predetermined period 2 amps 200 amps 5 amps
of time per dwelling
Average draw over a predetermined period 100 amps 300 amps 7.5 amps
of time per dwelling with a single grow
operation
Percent increase 50%
Example 2
Average draw over a predetermined period 2 amps 200 amps 1.67 amps
of time per dwelling 3.15 amps
Average draw over a predetermined period 100 amps 300 amps 2.50 amps
of time per dwelling with a single grow 4.72 amps
operation
Percent increase 50%
11 Once a primary line I has been identified as being suspect,
12 notification is sent by the system 5 to a monitoring agency or the utility,
such as
13 to the Utility Trouble Services personnel, who are trained to monitor
14 characteristics of the transformers, such as in the use of infrared laser
technology and which typically carry such equipment for use in other routine
16 servicing and monitoring functions. The Utility Trouble Services personnel
11
CA 02549087 2006-05-31
1 perform an infrared laser scan of each transformer 2 connected to the
suspect
2 primary line 1. Measurements are often performed at night when temperature
3 differences are greatest and the transformers 2 are not affected by
sunlight. The
4 laser is typically directed to approximately the same position on each
transformer
2. The measurements can be performed without access to private property.
6 Should a transformer 2 have a heat signature which exceeds the
7 typical heat signature, particularly when compared to other transformers 2
of the
8 same type connected to the same primary line 1 or a primary line 1 having a
9 similar supply voltage, the structures 4 connected thereto are further
tested to
identify a suspect structure 4s.
11 Load testing of the suspect transformer 2 having an excessive heat
12 signature is also performed by the Utility Trouble Services personnel on
each of
13 the secondary service lines 3 within the transformer 2. Once a suspect
residence
14 4s is identified, notification of the appropriate authorities and measures
taken are
governed by individual Utility provider policies and are beyond the scope of
the
16 embodiments described herein.
17
18 Example 1
19 Following identification of a suspect primary line 1, heat signature
comparisons of the transformers 2 were made using a 3M Scotchtrak Infrared
21 IR-60L2 Series infrared laser, available from 3M Canada, London Ontario,
22 Canada. The results are shown in Table B. The emissivity values of the unit
was
23 operated at 0.92 (unit preset) and 0.80 (value for steel) without
appreciable
24 difference. The comparisons were performed on a sunny day (not preferred)
at
10:30 am at an ambient temperature of 8.4 C (47.1 F).
12
CA 02549087 2006-05-31
1 A digital recording ammeter placed on one of two incoming
2 services lines from a suspect transformer 2 gave a reading of 31 amps over a
3 predetermined period of time. Without load testing, a presumption was made
that
4 the amperage load for both lines was approximately double or 62 amps. Three
additional transformers 2 in the grid of ten transformers 2 being fed from the
6 primary line I were measured for comparison. The three additional
transformers
7 2 were selected to have approximately the same sun exposure as the
8 measurements were made during the day. Of note, all dwellings 4 in the
9 residential area were of approximately the same size.
Table B
Transformer # of services Heat reading Heat reading
(Celsius) (Fahrenheit)
Suspect 7 13.5 56.3
1 10 9.4 48.9
2 7 9.3 48.7
3 4 9.2 48.6
Margin from highest to second highest 4.1 7.4
11
12
13 Example 2
14 Following identification of a suspect primary line 1, heat
comparisons were made using the same equipment identified in the first
16 example and are shown in Table C. The comparisons were performed at dawn
17 at an ambient temperature of 8.9 C (48.0 F). Approximately three hours
later, a
18 load test was performed on both secondary lines 3 feeding into the now-
19 identified grow operation. The results were 60.3 amps and 71.8 amps. Once
inside the residence 4s, it was confirmed from the timers that the amperage
21 readings at the time the heat comparisons were performed were the same as
13
CA 02549087 2006-05-31
1 when the load testing was done. Seven additional transformers 2 in the grid
of
2 ten transformers 2 being fed from the primary line 1 were measured for
3 comparison. Of note, all dwellings 4 in the residential area were of
approximately
4 the same size.
6 Table C
Transformer # of services Heat reading Heat reading
(Celsius) (Fahrenheit)
Suspect 10 16.8 62.2
1 9 10.6 51.1
2 10 10.5 50.9
3 11 10.6 51.1
4 11 10.5 50.9
5 11 10.5 50.9
6 9 8.5 47.3
7 8 8.9 48.0
Margin from highest to second highest 6.2 11.1
7
8
9 Example 3
As shown in Fig. 11, a residential community having 37 active
11 residences was monitored using an embodiment of the invention for the
12 presence of a grow-op. A meter 10 was placed on the primary line 1 feeding
13 transformers 2 (31W-3 to 31W-7). As shown in Fig. 12, a reading of 2.6 amps
14 was initially obtained at 1315 hours on Day 1. Subsequently, an atypical
consumption pattern was observed over a predetermined time period.
16 Heat signatures of the transformers 2 (31W-3 to 31W-7) were
17 measured, as shown in Table D, and it was noted that one transformer 2 (31W-
18 4) had a heat signature which exceeded that of the other transformers 2 fed
by
19 the primary line 1.
14
CA 02549087 2006-05-31
1 Table D
Transformer # of services Heat reading Heat reading
Celsius (Fahrenheit)
31W-7 6 8.7 47.7
31W-6 8 9.0 48.2
31W-5 8 8.8 47.8
31W-4 sus ect 8 12.9 55.2
31W-3 7 8.8 47.8
Margin from highest to second highest 3.9 7.0
2
3 Subsequent load testing identified the suspect residence 4s which
4 was entered on Day 2 at 1420 hours. An electrical bypass and 605 marijuana
plants were found within the residence of which 160 were starter plants under
2
6 active fluorescent lights, 150 were maturing plants (12hr) under 8 active
1000
7 watt lights, 150 were maturing plants (12hr) under 8 inactive 1000 watt
lights and
8 150 were vegetative plants (18hr) under 6 inactive 1000 watt lights. Three
160
9 watt exhaust fans were also found.
The sensitivity of a meter, having a resolution capable of
11 measuring amperage in 0.1 amp increments, connected to the primary line 1
12 was observed while load testing was performed under a variety of conditions
at
13 the pony panel in the residence. The results of the load testing are shown
in
14 Table E and can be compared to the primary line data shown in Fig. 12. It
is
clear that as the lights were activated, detection of the contribution from
the
16 grow-op was possible using the 0.1 amp meter. One of skill in the art would
17 appreciate that at higher primary line voltages corresponding amperage
would
18 be lower and thus meters capable of measuring at greater resolution are
19 required.
CA 02549087 2006-05-31
1 Table E
Conditions Time Secondary Secondary Primary Line
Line I Line 2 (Approx. -
(amps) (amps) amps)
8 x 1000 watt lights active 1500 33.8 44.6 3.0
Testing at bypass for entire 1510 42.9 60.9 3.5
consumption
All lights activated 1526 97.7 109.6 5.0
All lights activated 1540 98.4 108.9 5.0
All lights activated 1548 97.5 105.9 5.0
Blew fuse at pony panel 1550 ---- ----- 2.0
2
3 Having reference to Fig. 12, clear and distinct patterns of usage
4 were established and correspond with traditional consumption patterns. For
example one can note that the evening cycles are comparable however, the
6 consumption on Day 2 is lower. Both were mid-week and the external
7 temperature was comparable over the two day time period.
8 Clearly, once the marijuana grow-op electricity draw was
9 terminated mid-day on Day 2, a substantial decrease (about a 1.5-2 amp drop)
on the primary was observed. A 2 amp draw on the primary roughly equates to
11 about a 240 amp draw on the secondary. Factoring in line efficiency of
about
12 90%, the draw would theoretically be about 216 amps. The actual draw at the
13 time the grow-op was terminated was about 203.4 amps.
14 Clearly, the data from the meter connected to the primary line
correlates with the secondary power within the grid and illustrates that
electrical
16 usage patterns can be detailed from the primary line using a meter having
at
17 least a 0.1 amp resolution. One of skill in the art would understand that
it is likely
18 that a higher resolution meter would provide more detailed electrical use
patterns
19 and consumption details.
16
