Detecting actuation of electrical devices using electrical noise over a power line

Patent metadata

Field	Value
Publication number	US-9250275-B2
Application number	US-201213717195-A
Country	US
Kind code	B2
Filing date	Dec 17, 2012
Priority date	Sep 18, 2007
Publication date	Feb 2, 2016
Grant date	Feb 2, 2016

Abstract

Activity sensing in the home has a variety of important applications, including healthcare, entertainment, home automation, energy monitoring and post-occupancy research studies. Many existing systems for detecting occupant activity require large numbers of sensors, invasive vision systems, or extensive installation procedures. Disclosed is an approach that uses a single plug-in sensor to detect a variety of electrical events throughout the home. This sensor detects the electrical noise on residential power lines created by the abrupt switching of electrical devices and the noise created by certain devices while in operation. Machine learning techniques are used to recognize electrically noisy events such as turning on or off a particular light switch, a television set, or an electric stove. The system has been tested to evaluate system performance over time and in different types of houses. Results indicate that various electrical events can be learned and classified with accuracies ranging from 85-90%.

First claim

We claim:

1. An apparatus configured to detect a change in an electrical state of one or more electrical devices, the one or more electrical devices are coupled to an electrical power infrastructure of a structure and generate electrical noise on the electrical power infrastructure, the apparatus comprises:
a processing module configured to run on a computer processor; and
a sensing device configured to be removably coupled to an electrical outlet, the sensing device comprising:
a data acquisition receiver configured to receive the electrical noise via the electrical outlet when the sensing device is coupled to the electrical outlet, and further configured to convert the electrical noise into one or more first data signals, wherein:

the electrical outlet is electrically coupled to the electrical power infrastructure;
the sensing device is further configured to provide the one or more first data signals to the processing module; and
the processing module is further configured to determine electrical power consumed by the one or more electrical devices at least in part by using the one or more first data signals.

2. The apparatus of claim 1 , wherein:
the electrical noise comprises one or more electrical pulses on the electrical power infrastructure; and
the one or more electrical pulses are identifiable on the electrical power infrastructure for a length of time of less than ten microseconds.

3. The apparatus of claim 2 , wherein:
the one or more electrical pulses are identifiable on the electrical power infrastructure for the length of time of less than five microseconds.

4. The apparatus of claim 2 , wherein:
the electrical noise has a frequency between 10 hertz and 100,000 hertz.

5. A method of detecting and classifying electrical power usage by one or more electrical devices, the one or more electrical devices are coupled to an electrical power line of a structure, the method comprising:
measuring first electrical noise on the electrical power line;
after measuring the first electrical noise, using the first electrical noise to determine an occurrence of one or more electrical events on the electrical power line; and
using a computer processor to associate the occurrence of the one or more electrical events with a change in an electrical state of at least one device of the one or more electrical devices.

6. The method of claim 5 , wherein:
before measuring the first electrical noise, measuring second electrical noise on the electrical power line; and
before measuring the first electrical noise, using the second electrical noise to train the computer processor to associate the occurrence of the one or more electrical events with the change in the electrical state of at least one device of the one or more electrical devices.

7. The method of claim 5 , wherein:
using the computer processor to associate the occurrence of the one or more electrical events with the change in the electrical state comprises:
performing one or more transformations to the first electrical noise to separate out at least two component frequencies of the first electrical noise; and
performing a machine learning process to the at least two component frequencies of the first electrical noise to identify the change in the electrical state of the at least one device of the one or more electrical devices.

8. The method of claim 5 , wherein:
using the computer processor to associate the occurrence of the one or more electrical events with the change in the electrical state comprises:
associating the one or more electrical events with a switching-on or switching-off of the at least one device of the one or more electrical devices.

9. The method of claim 5 , wherein:
using the computer processor to associate the occurrence of the one or more electrical events with the change in the electrical state comprises:
associating the one or more electrical events with the change in the electrical state from a first state providing a first level of electrical power to the at least one device of the one or more electrical devices to a second state providing a second level of electrical power to the at least one device of the one or more electrical devices; and

the first level of electrical power is different from the second level of electrical power.

10. The method of claim 5 , further comprising:
coupling a sensing device to an electrical wall outlet of the structure, wherein:
the electrical wall outlet is electrically coupled to the electrical power line; and
measuring the first electrical noise comprises:
measuring the first electrical noise using the sensing device coupled to the electrical wall outlet of the structure.

11. The method of claim 5 , further comprising:
before measuring the first electrical noise, transmitting a first reference signal over the electrical power line, wherein:
the first reference signal interacts with a first device of the one or more electrical devices to create the first electrical noise.

12. The method of claim 5 , wherein:
the first electrical noise comprises one or more electrical pulses on the electrical power line; and
the one or more electrical pulses are detectable on the electrical power line for a length of time of less than ten microseconds.

13. The method of claim 12 , wherein:
the one or more electrical pulses are detectable on the electrical power line for the length of time of less than five microseconds.

14. An electrical state change detection device configured to detect a change in an electrical state of one or more electrical devices, the one or more electrical devices are coupled to an electrical power infrastructure of a structure, the electrical state change detection device comprises:
an interface module comprising:
a coupling mechanism configured to removably couple to the electrical power infrastructure; and
a receiver configured to capture electrical noise on the electrical power infrastructure via the coupling mechanism;

one or more filters coupled to the interface module and configured to pass one or more portions of the electrical noise;
a converter module coupled to the one or more filters and configured to convert the one or more portions of the electrical noise into one or more data signals; and
a processing module configured to run on a computer processor, the processing module is further configured to identify the change in the electrical state of the one or more electrical devices at least in part by using the one or more data signals.

15. The electrical state change detection device of claim 14 , wherein:
the processing module comprises:
an signature module configured to run on the computer processor and further configured to use the one or more data signals to create one or more noise signatures; and
an event classification module configured to run on the computer processor and further configured to use the one or more noise signatures to determine whether the change in the electrical state of the one or more electrical devices has occurred and to determine an event type of the change in the electrical state of the one or more electrical devices.

16. The electrical state change detection device of claim 14 , wherein:
the processing module is configured to identify when a first one of the one or more electrical devices is turned on or turned off at least in part by using the one or more data signals.

17. The electrical state change detection device of claim 14 , wherein: the one or more filters comprise:
a 60 hertz filter; and
a bandpass filter with a passband of:
100 hertz

Assignees

• Georgia Tech Res Inst

Inventors

• Patel Shwetak N
• Robertson Thomas M
• Abowd Gregory D
• Reynolds Matthew S

Classifications

• G01R21/006 Primary

  Measuring power factor · CPC title

• G05B2219/40408

  Intention learning · CPC title

• H04B2203/5458

  Monitor sensor; Alarm systems · CPC title

• Y04S20/38

  Cross-Sectional Technologies · mapped topic

• H04B2203/5425

  improving S/N by matching impedance, noise reduction, gain control · CPC title