# US 11,309,710 B2 · Method and system for assessing usage of devices of a property

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 11,309,710 B2
- Document ID: US11309710B2
- Status: Granted
- Kind: B2
- Application number: 15/768,387
- Filed: 2016-10-13
- Published: 2018-10-04
- Granted: 2022-04-19
- Earliest claimed: 2015-10-14
- Assignees: British Gas Trading Limited
- Applicants: British Gas Trading Limited
- Inventors: Alexandros Leontitsis; Oliver Parson
- Web page: https://patentgrep.com/patent/US11309710B2

## CPC

- G06Q 10/04: Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem" (market predictions or forecasting for commercial activities G06Q30/0202)
- G01D 4/002: {Remote reading of utility meters}
- G01D 4/004: {Remote reading of utility meters to a fixed location}
- G01D 4/006: {Remote reading of utility meters to a non-fixed location, i.e. mobile location}
- G01D 4/008: {Modifications to installed utility meters to enable remote reading}
- G06Q 50/06: Energy or water supply
- G06Q 10/06315: {Needs-based resource requirements planning or analysis}
- H02J 3/003: {Load forecast, e.g. methods or systems for forecasting future load demand}
- H02J 2105/42: Home appliances
- Y02B 70/30: Systems integrating technologies related to power network operation and communication or information technologies for improving the carbon footprint of the management of residential or tertiary loads, i.e. smart grids as climate change mitigation technology in the buildings sector, including also the last stages of power distribution and the control, monitoring or operating management systems at local level
- Y02B 70/3225: Demand response systems, e.g. load shedding, peak shaving
- Y04S 20/222: Demand response systems, e.g. load shedding, peak shaving
- Y04S 20/242: Home appliances
- Y04S 20/30: Smart metering, e.g. specially adapted for remote reading
- G06Q: INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G01D: MEASURING NOT SPECIALLY ADAPTED FOR A SPECIFIC VARIABLE; ARRANGEMENTS FOR MEASURING TWO OR MORE VARIABLES NOT COVERED IN A SINGLE OTHER SUBCLASS; TARIFF METERING APPARATUS; MEASURING OR TESTING NOT OTHERWISE PROVIDED FOR
- H02J: ELECTRIC POWER NETWORKS; CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- Y02B: CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO BUILDINGS, e.g. HOUSING, HOUSE APPLIANCES OR RELATED END-USER APPLICATIONS
- Y04S: SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS

## Abstract

In one aspect, there is provided a method comprising: receiving consumption data comprising readings from one or more utility meters associated with a property comprising one or more devices, the one or more devices comprising one or more devices of interest; determining, in the received consumption data, one or more positive consumption variations indicative of switching on of one or more of the devices and/or one or more negative consumption variations indicative of switching off of one or more of the devices; identifying one or more events associated with the one or more devices, based on the determined variations, by matching one or more positive variations with one or more negative variations; grouping the identified one or more events into one or more blocks, each block corresponding to an occurrence of usage of a device of the property; classifying the one or more blocks into one or more predetermined clusters, the one or more predetermined clusters comprising a respective predetermined cluster associated with each device of interest of the property; and determining an occurrence and/or an absence of usage of the one or more devices of interest of the property, based on the classification into the one or more predetermined clusters.

## Claims

### Claim 1

A computer-implemented method performed by at least one processor comprising:
receiving consumption data comprising readings from one or more utility meters associated with a property comprising one or more devices, the one or more devices comprising one or more devices of interest;
determining, in the received consumption data, one or more positive consumption variations indicative of switching on of one or more of the devices and/or one or more negative consumption variations indicative of switching off of one or more of the devices;
identifying a plurality of events associated with the one or more devices, based on the determined variations, each event identified by matching one or more positive variations with one or more negative variations;
grouping the identified events into one or more blocks, each block corresponding to an occurrence of usage of a device of the property, wherein the grouping comprises: grouping events of similar magnitudes occurring within a predetermined space of time, to generate a block associated with a device of interest of the property;
classifying the one or more blocks into one or more predetermined clusters, the one or more predetermined clusters comprising a respective predetermined cluster associated with each device of interest of the property, wherein the classifying comprises, for each block being classified, using a vector classifier using a model-based clustering and an expectation-maximization algorithm in an n-dimensional space to classify the block,
wherein, for a given one of said blocks formed in the grouping step, the given block comprising multiple events, each event corresponding to matched positive and negative power consumption variations as identified in the identifying step, the n dimensions input to the vector classifier comprise at least one of:
a time of the start of the given block of multiple events;
a time of the end of the given block of multiple events;
a duration of the given block of multiple events; and
a number of events in the given block of multiple events;
determining an occurrence and/or an absence of usage of the one or more devices of interest of the property, based on the classification into the one or more predetermined clusters;
comparing the determined occurrence and/or absence of usage of the one or more devices of interest of the property with one or more predetermined usage patterns associated with the one or more devices of interest of the property;
determining whether the determined occurrence and/or absence of usage of the one or more devices of interest of the property is associated with normal or abnormal usage, based on the comparison, wherein the comparing and determining whether the determined occurrence and/or absence of usage of the one or more devices steps comprise:
determining a detection time indicating one of: a time in an observation period at which occurrence of usage was detected; and a time period in the observation period during which an absence of usage was detected, and identifying the occurrence or absence of usage as normal or abnormal based on the detection time and probability information indicating probabilities of usage of the device over the observation period; and
based on the detection time and on whether the occurrence and/or absence of usage was determined as normal or abnormal;
causing output of a selected one of a plurality of predetermined alarm signal types when determined as abnormal, wherein the selected alarm signal type comprises one of: a first type of alarm signal to cause a communication to a relative; a second type of alarm signal to cause a communication to a professional caregiver; and a third type of alarm signal to cause a visit to the property by the professional caregiver; or
not causing output of any type of alarm signal when determined normal.

### Claim 2 (depends on 1)

The method of claim 1, further comprising:
receiving, from a user of the property and/or an operator, a cancellation signal overriding the one or more output alarm signals.

### Claim 3 (depends on 1)

The method of claim 1, wherein classifying the one or more blocks into one or more predetermined clusters further comprises using a classifier using at least one of:
a Support Vector Machine,
a k-nearest neighbours (k-NN) technique, and/or
a Bayesian inference.

### Claim 4 (depends on 3)

The method of claim 3, wherein the vector classifier using the expectation-maximization algorithm is configured to generate a single cluster associated with each device of interest of the property, wherein the vector classifier using expectation-maximization in an n-dimensional space is based on the consumption data and/or on a set of predetermined assumptions, wherein the vector and/or the assumptions are modified by parameters and/or modifiers based on the consumption data and/or on a set of predetermined assumptions.

### Claim 5 (depends on 1)

The method of claim 1, wherein grouping the identified one or more events into one or more blocks comprises:
selecting, based on the one or more devices of interest of the property, blocks comprising a number of events comprised within a predetermined range of numbers of events.

### Claim 6 (depends on 1)

The method of claim 1, further comprising determining one or more usage patterns, comprising:
receiving usage data associated with one or more determined occurrences of usage of the one or more devices of interest of the property, the usage data being received for a predetermined observation period of time of at least one half day, the observation period of time comprising one or more time divisions W divided into n time intervals ti shorter than an hour, with i being an integer between 1 and n;
retrospectively determining, in the received usage data, the probability of use of the one or more devices of interest within each of the n time intervals ti, by determining a usage ratio ri, with i being an integer between 1 and n, such that, for each device of interest of the property and for each time division W:
ri
=
number
⁢
⁢
of
⁢
⁢
times
⁢
⁢
device
⁢
⁢
of
⁢
⁢
interest
⁢
⁢
was
⁢
used
⁢
⁢
at
⁢
⁢
least
⁢
⁢
once
⁢
⁢
in
⁢
⁢
interval
⁢
⁢
t
i
number
⁢
⁢
of
⁢
⁢
intervals
⁢
⁢
t
i
⁢
⁢
in
⁢
⁢
time
⁢
⁢
division
⁢
⁢
W
.

### Claim 7 (depends on 6)

The method of claim 6, further comprising at least one of, for one or more subdivisions D of one or more divisions W:
estimating, based on the determined probability of use, a usage probability density function of usage of the device of interest of the property, wherein estimating the probability density function comprises using a Kernel Density Estimation, or
defining, based on the estimated usage probability density function, one or more predefined moments in time corresponding to times at which a determined occurrence of usage of the one or more devices of interest of the property is associated with abnormal usage.

### Claim 8 (depends on 6)

The method of claim 6, further comprising, for one or more subdivisions D of one or more divisions W:
estimating a cumulative probability CPαi that the one or more devices of interest of the property is used within a given period Hαi of time within a subdivision D, wherein the period of time Hαi:
starting at a beginning of first time interval being a predetermined reference time interval tM associated with the start of the subdivision D, and
ending at an end of a second time interval ti, wherein M and i are respective indices identifying the first and second time intervals;
such that:
CP
⁢
⁢
α
⁢
⁢
i
=
1
-
∏
j
=
M
i
⁢
1
-
rj
.
wherein rj is the usage ratio for index j.

### Claim 9 (depends on 8)

The method of claim 8, further comprising at least one of:
estimating the cumulative probability that the one or more devices of interest of the property is not used within the given period Hαi of time within the subdivision D, based on the estimated cumulative probability, or defining, based on the given periods Hαi and one or more of the estimated cumulative probabilities, one or more predefined moments in time corresponding to times by which a determined absence of usage of the one or more devices of interest of the property is associated with abnormal usage, or comparing one or more moments in time at which the absence and/or occurrence of usage is determined with the one or more predefined moments in time.

### Claim 10 (depends on 1)

The method of claim 1, wherein determining whether the determined absence of usage of the one or more devices of interest of the property is associated with normal or abnormal usage comprises:
determining that the determined absence of usage is associated with abnormal usage if one or more moments in time by which the absence of usage is determined is later than at least one of a predefined moments in time.

### Claim 11 (depends on 1)

The method of claim 1, wherein determining whether the determined occurrence of usage of the one or more devices of interest of the property is associated with normal or abnormal usage comprises:
determining that the determined occurrence of usage is associated with abnormal usage if one or more moments in time at which the occurrence of usage is determined is similar to at least one of a predefined moments in time.

### Claim 12 (depends on 7)

The method of claim 7, wherein the one or more predefined moments comprise, for each device of interest of the property, different types of times causing output of different types of alarm signals.

### Claim 13 (depends on 1)

The method of claim 1, wherein the determining of the consumption variations comprises, at least one of:
defining edges between consecutive readings received from the utility meters, or merging a plurality of consecutive similar determined variations.

### Claim 14 (depends on 1)

The method of claim 1, wherein the identifying of the one or more events comprises at least one of:
ranking the identified variations based on their respective energy magnitude,
searching for matching variations within a predetermined search period,
searching for a match to the positive variations by iterations in successive turns, starting with the largest magnitude determined variation and continuing in decreasing magnitude of variation,
removing consumption data associated with an identified event from the received consumption data, or
performing one or more iterations on the consumption data.

### Claim 15 (depends on 1)

The method of claim 1, wherein the utility meter comprises at least one of:
a gas meter;
an electricity meter;
a water meter;
a bandwidth meter.

### Claim 16 (depends on 1)

The method of claim 1, wherein the one or more output alarm signals cause one or more of:
an intervention being triggered in the property;
a message being sent to a user of the property; and/or to an operator.

### Claim 17 (depends on 1)

The method of claim 1, wherein the one or more devices of interest comprises, one or more of the following:
a heating system,
a cooling system,
a lighting system,
an appliance, and/or
a water access point.

### Claim 18 (depends on 1)

A non-transitory computer readable medium comprising software code adapted, when executed on a data processing apparatus, to perform the method as set out in claim 1.

### Claim 19 (depends on 1)

A system comprising a processor and associated memory configured for performing the method as set out in claim 1.

### Claim 20 (depends on 1)

An apparatus comprising a processor and associated memory configured for performing the method as set out in claim 1.