# US 6,726,113 B2 · Temperature control strategy utilizing neural network processing of occupancy and activity level sensing

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

- Number: US 6,726,113 B2
- Document ID: US6726113B2
- Status: Granted
- Kind: B2
- Application number: 10/082,792
- Filed: 2002-02-25
- Published: 2003-08-28
- Granted: 2004-04-27
- Earliest claimed: 2002-02-25
- Assignees: Carrier Corporation
- Inventors: Charles Guo
- Web page: https://patentgrep.com/patent/US6726113B2

## CPC

- G05D 23/1917: {using digital means}
- F24F 11/30: for purposes related to the operation of the system, e.g. for safety or monitoring
- F24F 11/46: Improving electric energy efficiency or saving
- F24F 11/63: Electronic processing
- F24F 2120/10: Occupancy
- F24F 2120/14: Activity of occupants
- G05D: SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- F24F: AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING (removing dirt or fumes from areas where they are produced B08B15/00; vertical ducts for carrying away waste gases from buildings E04F17/02; tops for chimneys or ventilating shafts, terminals for flues F23L17/02)

## Abstract

A temperature control system utilizes detected occupancy and activity levels to automatically condition a response of an HVAC system to a difference between a setpoint temperature and an actual temperature within a zone. The illustrated example includes an infrared sensor that provides at least one signal indicating the activity and occupancy levels in the zone. A neural network processes the sensor signal to provide an indication of the occupancy and activity levels to a controller. The controller automatically adjusts at least one control parameter of the HVAC system to compensate for changes in the occupancy or activity levels that would affect the temperature comfort setting in the zone.

## Claims

### Claim 1

A temperature control assembly, comprising:
a thermostat device that allows an individual to select a setpoint temperature for a zone;
a sensor that detects whether at least one individual occupies the zone and a level of activity within the zone, the sensor providing at least one signal indicating the detected occupation and activity level;
a neural network that receives the sensor signal and provides an output classifying the sensed activity and occupation levels; and
a controller that receives the neural network network output and automatically conditions a response to a difference between a current temperature in the zone and the selected setpoint temperature.

### Claim 2 (depends on 1)

The assembly of claim 1, including a sensor signal modifier that provides a digital and an analog output to the neural network that are indicative of a variance and mean of the sensor signal, respectively.

### Claim 3 (depends on 1)

The assembly of claim 1, wherein the neural network comprises an input layer that processes a mean and variance of the sensor signal, a hidden layer coupled to the input layer through first selectively adjustable strengths and an output layer coupled to the hidden layer through second selectively adjustable strengths, the output layer providing the network output.

### Claim 4 (depends on 3)

The assembly of claim 3, wherein the output layer comprises neurons that each provide a one bit output value and wherein the combined output values indicate one of four possible conditioned responses.

### Claim 5 (depends on 3)

The assembly of claim 3, wherein the network is trainable to adjust the strengths responsive to a selected sensor input such that the network output has a selected value.

### Claim 6 (depends on 5)

The assembly of claim 5, wherein the network is trainable in a manner that allows a previously selected strength to be adjusted to a new strength.

### Claim 7 (depends on 1)

The assembly of claim 1, wherein the controller causes there to be no response to the temperature difference when the network output indicates that the zone is not occupied.

### Claim 8 (depends on 1)

The assembly of claim 1, wherein the controller automatically adjusts the setpoint temperature responsive to at least one selected network output.

### Claim 9 (depends on 1)

The assembly of claim 1, wherein the controller automatically adjusts a performance of a selected portion of a HVAC system associated with the thermostat responsive to at least one selected network output.

### Claim 10 (depends on 1)

The assembly of claim 1, wherein the sensor comprises a passive infrared pyroelectric sensor.

### Claim 11

A method of controlling a temperature in a zone where a thermostat provides an indication of a selected setpoint temperature, comprising the steps of:
detecting whether the zone is occupied by an individual;
determining an activity level within the zone;
using a neural network to generate an output indicating the detected occupancy and the determined activity level; and
automatically responding to the neural network output by adjusting a response to a difference between the setpoint temperature and a current temperature in the zone.

### Claim 12 (depends on 11)

The method of claim 11, including training the neural network by providing a selected input indicating a known occupancy level and a known activity level and adjusting strengths of the neural network to condition the output of the network to a desired value responsive to the selected input.

### Claim 13 (depends on 12)

The method of claim 12, including changing at least one of the adjusted strengths responsive to a second selected input indicating the known occupancy and activity levels.

### Claim 14 (depends on 11)

The method of claim 11, including preventing any response to the temperature difference when the network output indicates that the zone is not occupied.

### Claim 15 (depends on 11)

The method of claim 11, including automatically changing the setpoint temperature responsive to at least one network output.

### Claim 16 (depends on 11)

The method of claim 11, including automatically adjusting a performance parameter of a HVAC system associated with the thermostat responsive to at least one network output.

### Claim 17 (depends on 11)

The method of claim 11, including using a sensor to provide an indication of the occupation level and activity level and using a mean and variance of the sensor output in the neural network to generate the network output.

### Claim 18

An air conditioning system, comprising:
a compressor that selectively compresses a refrigerant fluid;
an condenser in fluid communication with the compressor;
an evaporator in fluid communication with the condenser and the compressor;
a fan that moves air across at least portions of the evaporator to provide the air to a selected zone;
a thermostat that allows an individual to select a setpoint temperature for the zone and provides an indication of a current temperature in the zone;
at least one sensor that provides at least one signal responsive to a detected occupancy level and activity level in the zone;
a neural network that processes the sensor signal and provides an output indicating the level of occupancy and activity in the zone; and
a controller that automatically adjusts a response of the system to a difference between the selected setpoint temperature and the actual temperature in the zone responsive to the network output.

### Claim 19 (depends on 18)

The system of claim 18, wherein the controller automatically does at least one of turning off the system, changing the setpoint temperature, changing the compressor frequency, or changing the fan speed responsive to a corresponding network output.

### Claim 20 (depends on 18)

The system of claim 18, including a sensor signal conditioner that provides a mean and variance of the sensor signal to the neural network.
