# US 2019/0316799 A1 · Systems and methods for intelligent thermostat

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

- Number: US 2019/0316799 A1
- Document ID: US20190316799A1
- Status: Published application
- Kind: A1
- Application number: 16/384,674
- Filed: 2019-04-15
- Published: 2019-10-17
- Earliest claimed: 2018-04-17
- Inventors: Harshal V BHARATIA
- Web page: https://patentgrep.com/patent/US20190316799A1

## CPC

- F24F 11/46: Improving electric energy efficiency or saving
- F24F 11/54: using one central controller connected to several sub-controllers
- F24F 11/58: using Internet communication
- F24F 11/61: using timers
- F24F 11/62: characterised by the type of control or by internal processing, e.g. using fuzzy logic, adaptive control or estimation of values
- F24F 11/64: using pre-stored data
- F24F 11/65: for selecting an operating mode
- F24F 11/72: for controlling the supply of treated air, e.g. its pressure
- F24F 11/79: for controlling the direction of the supplied air
- F24F 11/80: for controlling the temperature of the supplied air
- G05B 19/042: using digital processors (G05B19/05 takes precedence)
- G06N 3/045: Combinations of networks
- G06N 3/0499: Feedforward networks
- G06N 3/08: Learning methods
- G06N 3/09: Supervised learning
- G06N 3/092: Reinforcement learning
- G06N 3/096: Transfer learning
- G06N 3/098: Distributed learning, e.g. federated learning
- G06N 20/00: Machine learning
- F24F 2110/10: Temperature
- F24F 2110/12: of the outside air
- F24F 2110/20: Humidity
- F24F 2110/22: of the outside air
- F24F 2120/12: Position of occupants
- F24F 2120/20: Feedback from users
- G05B 2219/2614: HVAC, heating, ventillation, climate control
- 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)
- G05B: CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS (systems for controlling or regulating non-electric variables G05D; systems for regulating electric or magnetic variables G05F; control devices or systems insofar as characterised by mechanical features only G05G)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS

## Abstract

The present invention provides systems and methods where an intelligent thermostat optimizes operation of HVAC system by learning based on thermal profile of a zone and improvising past performance with the goal of energy and cost savings while managing user comfort. The systems and methods further use adaptive clustering to dissect huge problem space, and dynamic collaboration strategy between thermostat units to rapidly enhance performance.

## Claims

### Claim 1

A system to regulate temperature for a zone of a dwelling, the system comprising a memory, a sensor and a processor configured to:
obtain a first system mode from a schedule for a first interval;
obtain a first comfort range of temperature and a first comfort level from said schedule for said first interval;
obtain a thermal profile for said zone; and
control an HVAC system to regulate a first temperature of said zone for said first interval, wherein to control said HVAC system, the processor is configured to:
obtain said first temperature of said zone and a plurality of zone conditions for said zone based on a sensor data item from the sensor for said first interval;
perform an adaptation processing for said first interval, wherein to perform said adaptation processing the processor is configured to:
identify a target processing to regulate said first temperature, wherein said target processing comprises:
a transit processing to transit said first temperature to a transit end temperature, said transit end temperature is above a minimum temperature of a transit comfort range of temperature for a heat system mode of said first system mode and is below a maximum temperature of said transit comfort range of temperature for a cool system mode of said first system mode, wherein to perform said transit processing the processor is configured to:
perform a first transition of said first temperature at beginning of said first interval with said transit comfort range of temperature as said first comfort range of temperature; or
perform a second transition of said first temperature at end of said first interval, wherein to perform said second transition, the processor is configured to obtain a second comfort range of temperature for a second interval in said schedule, said second interval following said first interval and requiring temperature at start of said second interval as a temperature within said second comfort range of temperature and with said transit comfort range of temperature as said second comfort range of temperature; or
a sustain processing to sustain said first temperature within said first comfort range of temperature; and
determine a plurality of parameters for said target processing based on a plurality of adaptation items, said plurality of parameters comprising a time, a duration and at least one parameter of said HVAC system for performing said target processing, said plurality of adaptation items comprising a representation of said thermal profile, a representation of said schedule for said target processing, said plurality of zone conditions and a plurality of target processing results, wherein selection of said plurality of parameters for said target processing regulates said first temperature based on said first comfort range of temperature while minimizing energy use and maximizing user comfort;
operate said HVAC system to perform said target processing at said time for said duration, whereby operating said HVAC system to perform said target processing regulates said first temperature of said zone as specified by said schedule; and
add a target processing result to said plurality of target processing results, wherein said target processing result comprises said plurality of parameters for said target processing, a representation of cost of said target processing and a change in a zone condition due to said target processing.

### Claim 2 (depends on 1)

The system of claim 1, wherein to obtain said schedule the processor is further configured to:
define said schedule by a user;
derive said schedule based on a behavior of a user; or
any combination thereof.

### Claim 3 (depends on 1)

The system of claim 1, wherein to obtain said thermal profile the processor is further configured to:
derive said thermal profile for said zone wherein, to derive said thermal profile the processor is configured to:
determine a probe range of temperature based on lowest temperature and highest temperature in said schedule for said first system mode;
operate said HVAC system for said first system mode using an HVAC system stage and determine a first time to change one degree for a unit temperature range in said probe range of temperature for said first system mode and said HVAC system stage;
stop said operation of said HVAC system and determine a second time to change one degree for a unit temperature range in said probe range of temperature; and
build said thermal profile based on said first time and said second time for said zone;
transfer said thermal profile of said zone from an external service wherein said external service is external to said system; or
any combination thereof.

### Claim 4 (depends on 1)

The system of claim 1, wherein the processor is further configured to obtain said sensor data item from the sensor for said first interval comprising:
an indoor humidity for said zone;
an indoor temperature for said zone;
a temperature outside the dwelling;
a humidity outside the dwelling; or
any combination thereof.

### Claim 5 (depends on 1)

The system of claim 1, wherein to perform said transit processing, the processor is further configured to determine a transit HVAC stage, a transit begin time and a transit duration based on:
using a minimum HVAC stage of available plurality of HVAC system stages; and
performing said transit processing that reaches said transit end temperature at a transit end time, wherein the processor minimizes interval between start of said first interval and said transit end time during said first transition and minimizes interval between start of said second interval and said transit end time during said second transition.

### Claim 6 (depends on 5)

The system of claim 5, wherein to determine said transit HVAC stage, said transit begin time and said transit duration the processor is further configured to:
compute a maximum transit duration allowed for said transit processing based on a total duration of said first interval;
find a matching transit entry from a plurality of transit entries that matches said first temperature and said plurality of zone conditions;
derive said transit HVAC stage, said transit begin time and said transit duration based on:
found said matching transit entry;
said thermal profile if said thermal profile contains a match for said first temperature and said plurality of zone conditions; or
said maximum transit duration and lowest HVAC stage of said HVAC system if there is no matching transit entry and said thermal profile contains no match; and
add a new transit entry to said plurality of transit entries upon completion of said transit processing, said new transit entry comprising said transit HVAC stage used for said transit processing and said transit duration as a duration for which said transit HVAC stage was operational during said transit processing.

### Claim 7 (depends on 1)

The system of claim 1, wherein to perform said sustain processing, the processor is further configured to determine a sustain HVAC stage and a sustain range of temperature based on:
using a minimum HVAC stage of available plurality of HVAC system stages;
operating said HVAC system with minimum startup and shutdown cycle during said duration for said sustain processing; and
controlling said first temperature during said duration for said sustain processing to maximize user comfort and maintaining cost of said sustain processing within a predetermined bump threshold cost, wherein to maximize comfort the processor is configured to:
move said sustain range of temperature towards maximum temperature of said comfort range of temperature for a heat system mode of said first system mode; and
move said sustain range of temperature towards minimum temperature of said comfort range of temperature for a cool system mode of said first system mode.

### Claim 8 (depends on 7)

The system of claim 7, wherein the processor is further configured to determine said sustain HVAC stage and said sustain range of temperature based on:
a matching learned sustain entry from a plurality of learned sustain entries if said matching learned sustain entry belongs to said plurality of learned sustain entries and matches said first comfort range of temperature, said first comfort level and said plurality of zone conditions;
a no-matching sustain entry, wherein said no-matching sustain entry is created and added to a plurality of learning sustain entries when no learning sustain entry matches said first comfort range of temperature, said first comfort level, and said plurality of zone conditions, creating said no-matching widen sustain entry comprising:
computing a first target temperature based on said first comfort range of temperature and said first comfort level;
determining said sustain range of temperature based on said first target temperature;
determining a minimum HVAC stage to perform said sustain processing based on said thermal profile, said first target temperature and said plurality of zone conditions;
setting a learning mode to widen; and
setting said no-matching sustain entry as said learning sustain entry;
a matching widen sustain entry, wherein said matching widen sustain is obtained by:
finding a candidate widen sustain entry from said plurality of learning sustain entries with said learning mode of widen and matching said first comfort range of temperature, said first comfort level, and said plurality of zone conditions;
performing a widening of said sustain range of temperature by a widen amount if said sustain range of temperature remains within bounds of said first comfort range of temperature for said found candidate widen sustain entry and does not increase cost of said sustain processing beyond a predetermined widen threshold cost;
updating said learning mode for said candidate widen sustain entry to bump comfort if it is not possible to widen said sustain range of temperature by said widen amount for said found candidate widen sustain entry or said widening of said sustain range of temperature increases cost of said sustain processing beyond said predetermined widen threshold cost; and
using said candidate widen sustain entry found as said matching widen sustain entry;
a matching bump comfort sustain entry, wherein said matching bump comfort sustain entry is obtained by:
finding a candidate bump comfort sustain entry from said plurality of learning sustain entries with said learning mode of bump comfort and matching said first comfort range of temperature, said first comfort level, and said plurality of zone conditions;
performing an increase in comfort level of said candidate bump comfort sustain entry for said found candidate finish learning sustain entry provided cost of said sustain processing does not increase beyond a predetermined bump threshold cost;
updating said learning mode for said candidate bump comfort sustain entry to finish learning for said found candidate finish learning sustain entry if increase in comfort level increases cost of said sustain processing beyond said predetermined widen threshold cost; and
using said candidate bump comfort sustain entry found as said matching bump comfort sustain entry; or
a matching finish learning sustain entry, wherein said matching finish learning sustain entry is obtained by:
finding a candidate finish learning sustain entry from said plurality of learning sustain entries with said learning mode of finish learning and matching said first comfort range of temperature, said first comfort level, and said plurality of zone conditions;
removing said candidate finish learning sustain entry from said plurality of learning sustain entries and adding said candidate finish learning sustain entry to said plurality of learned sustain entries; and
using said candidate finish learning sustain entry found as said matching finish learning sustain entry.

### Claim 9 (depends on 1)

The system of claim 1, wherein to determine said plurality of parameters for said target processing, the processor is further configured to use:
a functional logic;
a fuzzy logic;
a decision tree;
a neural network based learning;
a heuristics based learning;
a reinforcement learning; or
any combination thereof.

### Claim 10 (depends on 1)

The system of claim 1, wherein the processor is further configured to determine said plurality of parameters for said target processing using:
said target processing result of a past target processing in said system;
said target processing result of a past target processing in a different system different than said system;
a software program that interfaces with said system;
an additional processing implementation that can be loaded into said system;
a computing platform that directly or indirectly interfaces with said system; or
any combination thereof.

### Claim 11 (depends on 1)

The system of claim 1, wherein the processor is configured to request a server processing performed by a serving processor based on a plurality of routing parameters, said serving processor is configured to perform said server processing to determine said plurality of parameters for said target processing, said plurality of routing parameters comprising said first temperature, said first comfort level, and said plurality of zone conditions.

### Claim 12 (depends on 11)

The system of claim 11, wherein to request said server processing, the processor is configured to:
request to perform a route processing and obtain a routing map, wherein a route provider processor is configured to perform said route processing to create, update, and provide said routing map, said routing map associating a plurality of values for a routing parameter with a server processing;
identify said server processing based on matching a value for a routing parameter of said target processing with the plurality of values for the routing parameter associated with the server processing by said routing map;
request to perform the route processing for updating said routing map by adding a value to said plurality of values for a routing parameter associated with said server processing on learning a new association for said target processing with said serving processing; and
provide said plurality of adaptation items to perform said server processing.

### Claim 13 (depends on 11)

The system of claim 11, wherein to perform said server processing for said target processing, said serving processor is configured to:
obtain said plurality of adaptation items for said target processing;
identify a plurality of operation build implementations for determining said plurality of parameters for said target processing;
derive a decision knowledge item based on said plurality of target processing results for an operation build implementation in said plurality of operation build implementations;
estimate cost of said target processing for an operation build implementation in said plurality of operation build implementations based on said decision knowledge item and said plurality of adaptation items;
select a final operation build implementation in said plurality of operation build implementations, said final operation build implementation matches said decision knowledge item for said plurality of adaptation items and has least estimate of cost of said target processing;
use said final operation build implementation to determine said plurality of parameters for said target processing based on said plurality of adaptation items; and
obtain said target processing result and update said decision knowledge item for an operation build implementation in said plurality of operation build implementations.

### Claim 14 (depends on 1)

The system of claim 1, wherein the processor is further configured to:
regulate temperature of an elevated zone located in said dwelling at a higher elevation than said zone;
operate an upstream equalizer baffle to control a flow of air in an air duct from said zone to a return input of the HVAC system for said elevated zone;
operate a downstream equalizer baffle, to control a flow of air in an air duct from said elevated zone to a return input of the HVAC system for said zone; and
control the operation of said upstream equalizer baffle and said downstream equalizer baffle wherein to control said upstream equalizer baffle and said downstream equalizer baffle the processor is configured to:
activate the upstream equalizer baffle to move cool air from said zone to said elevated zone when said zone system mode is a cool system mode and said elevated zone system mode is a cool system mode; and
activate the downstream equalizer baffle to move hot air from said elevated zone to said zone when said zone system mode is a heat system mode and said elevated zone system mode is a heat system mode.

### Claim 15

A method to regulate temperature for a zone of a dwelling, the method executed by a processor to perform a plurality of operations, the method comprising:
obtaining a first system mode from a schedule for a first interval;
obtaining a first comfort range of temperature and a first comfort level from said schedule for said first interval;
obtaining a thermal profile for said zone; and
controlling an HVAC system to regulate a first temperature of said zone for said first interval, wherein controlling said HVAC system comprising:
obtaining said first temperature of said zone and a plurality of zone conditions for said zone based on a sensor data item from a sensor for said first interval;
performing an adaptation processing for said first interval, wherein said adaptation processing comprising:
identifying to perform a target processing to regulate said first temperature, wherein performing said target processing comprising:
performing a transit processing to transit said first temperature to a transit end temperature, said transit end temperature is above a minimum temperature of a transit comfort range of temperature for a heat system mode of said first system mode and is below a maximum temperature of said transit comfort range of temperature for a cool system mode of said first system mode, wherein said transit processing comprising:
performing a first transition of said first temperature at beginning of said first interval with said transit comfort range of temperature as said first comfort range of temperature; or
performing a second transition of said first temperature at end of said first interval, wherein performing said second transition comprises obtaining a second comfort range of temperature for a second interval in said schedule, said second interval following said first interval and requiring temperature at start of said second interval as a temperature within said second comfort range of temperature and with said transit comfort range of temperature as said second comfort range of temperature; or
performing a sustain processing to sustain said first temperature within said first comfort range of temperature; and
determining a plurality of parameters for said target processing based on a plurality of adaptation items, said plurality of parameters comprising a time, a duration and at least one parameter of said HVAC system for performing said target processing, said plurality of adaptation items comprising a representation of said thermal profile, a representation of said schedule for said target processing, said plurality of zone conditions and a plurality of target processing results, wherein selection of said plurality of parameters for said target processing regulates said first temperature based on said first comfort range of temperature while minimizing energy use and maximizing user comfort;
operating said HVAC system to perform said target processing at said time for said duration, whereby operating said HVAC system to perform said target processing regulates said first temperature of said zone as specified by said schedule; and
adding a target processing result to said plurality of target processing results, wherein said target processing result comprises said plurality of parameters for said target processing, a representation of cost of said target processing and a change in a zone condition due to said target processing.

### Claim 16 (depends on 15)

The method of claim 15, wherein obtaining said schedule comprises:
defining said schedule by a user;
deriving said schedule based on a behavior of a user; or
any combination thereof.

### Claim 17 (depends on 15)

The method of claim 15, wherein obtaining said thermal profile further comprises:
deriving said thermal profile for said zone wherein, deriving said thermal profile comprises:
determining a probe range of temperature based on lowest temperature and highest temperature in said schedule for said first system mode;
operating said HVAC system for said first system mode using an HVAC system stage and determining a first time to change one degree for a unit temperature range in said probe range of temperature for said first system mode and said HVAC system stage;
stopping said operation of said HVAC system and determining a second time to change one degree for a unit temperature range in said probe range of temperature; and
dwelling said thermal profile based on said first time and said second time for said zone;
transferring said thermal profile of said zone using a service processor wherein said service processor is other than the processor; or
any combination thereof.

### Claim 18 (depends on 15)

The method of claim 15, wherein obtaining said sensor data item from the sensor for said first interval comprising:
an indoor humidity for said zone;
an indoor temperature for said zone;
a temperature outside the dwelling;
a humidity outside the dwelling; or
any combination thereof.

### Claim 19 (depends on 15)

The method of claim 15, wherein determining a transit HVAC stage, a transit begin time and a transit duration for said transit processing is based on:
using a minimum HVAC stage of available plurality of HVAC system stages; and
performing said transit processing to reach said transit end temperature at a transit end time, said transit processing minimizing interval between start of said first interval and said transit end time during said first transition and minimizing interval between start of said second interval and said transit end time during said second transition.

### Claim 20 (depends on 19)

The method of claim 19, wherein determining said transit HVAC stage, said transit begin time and said transit duration the processor comprises:
computing a maximum transit duration allowed for said transit processing based on a total duration of said first interval;
finding a matching transit entry from a plurality of transit entries that matches said first temperature and said plurality of zone conditions;
deriving said transit HVAC stage, said transit begin time and said transit duration based on:
found said matching transit entry;
said thermal profile if said thermal profile contains a match for said first temperature and said plurality of zone conditions; or
said maximum transit duration and lowest HVAC stage of said HVAC system if there is no matching transit entry and said thermal profile contains no match; and
adding a new transit entry to said plurality of transit entries upon completion of said transit processing, said new transit entry comprising said transit HVAC stage used for said transit processing and said transit duration as a duration for which said transit HVAC stage was operational during said transit processing.

### Claim 21 (depends on 15)

The method of claim 15, wherein determining a sustain HVAC stage and a sustain range of temperature for said sustain processing is based on:
using a minimum HVAC stage of available plurality of HVAC system stages;
operating said HVAC system with minimum startup and shutdown cycle during said duration for said sustain processing; and
controlling said first temperature during said duration for said sustain processing for maximizing user comfort and maintaining cost of said sustain processing within a predetermined bump threshold cost, wherein maximizing said user comfort comprises:
moving said sustain range of temperature towards maximum temperature of said comfort range of temperature for a heat system mode of said first system mode; and
moving said sustain range of temperature towards minimum temperature of said comfort range of temperature for a cool system mode of said first system mode.

### Claim 22 (depends on 21)

The method of claim 21, wherein determining said sustain HVAC stage and said sustain range of temperature is based on:
a matching learned sustain entry from a plurality of learned sustain entries if said matching learned sustain entry belongs to said plurality of learned sustain entries and matches said first comfort range of temperature, said first comfort level and said plurality of zone conditions;
a no-matching sustain entry, wherein said no-matching sustain entry is created and added to a plurality of learning sustain entries when no learning sustain entry matches said first comfort range of temperature, said first comfort level, and said plurality of zone conditions, creating said no-matching widen sustain entry comprising:
computing a first target temperature based on said first comfort range of temperature and said first comfort level;
determining said sustain range of temperature based on said first target temperature;
determining minimum HVAC stage to perform said sustain processing based on said thermal profile, said first target temperature and said plurality of zone conditions;
setting a learning mode to widen; and
setting said no-matching sustain entry as said learning sustain entry;
a matching widen sustain entry, wherein said matching widen sustain is obtained by:
finding a candidate widen sustain entry from said plurality of learning sustain entries with said learning mode of widen and matching said first comfort range of temperature, said first comfort level, and said plurality of zone conditions;
performing a widening of said sustain range of temperature by a widen amount if said sustain range of temperature remains within bounds of said first comfort range of temperature for said found candidate widen sustain entry and does not increase cost of said sustain processing beyond a predetermined widen threshold cost;
updating said learning mode for said candidate widen sustain entry to bump comfort if it is not possible to widen said sustain range of temperature by widen amount for said found candidate widen sustain entry or widening of said sustain range of temperature increases cost of said sustain processing beyond said predetermined widen threshold cost; and
using said candidate widen sustain entry found as said matching widen sustain entry;
a matching bump comfort sustain entry, wherein said matching bump comfort sustain entry is obtained by:
finding a candidate bump comfort sustain entry from said plurality of learning sustain entries with said learning mode of bump comfort and matching said first comfort range of temperature, said first comfort level, and said plurality of zone conditions;
performing an increase in comfort level of said candidate bump comfort sustain entry for said found candidate finish learning sustain entry provided cost of said sustain processing does not increase beyond a predetermined bump threshold cost;
updating said learning mode for said candidate bump comfort sustain entry to finish learning for said found candidate finish learning sustain entry if increase in comfort level increases cost of said sustain processing beyond said predetermined widen threshold cost; and
using said candidate bump comfort sustain entry found as said matching bump comfort sustain entry; or
a matching finish learning sustain entry, wherein said matching finish learning sustain entry is obtained by:
finding a candidate finish learning sustain entry from said plurality of learning sustain entries with said learning mode of finish learning and matching said first comfort range of temperature, said first comfort level, and said plurality of zone conditions;
removing said candidate finish learning sustain entry from said plurality of learning sustain entries and adding said candidate finish learning sustain entry to said plurality of learned sustain entries; and
using said candidate finish learning sustain entry found as said matching finish learning sustain entry.

### Claim 23 (depends on 15)

The method of claim 15, further comprising determining said plurality of parameters for said target processing using:
a functional logic;
a fuzzy logic;
a decision tree;
a neural network based learning;
a heuristics based learning;
a reinforcement learning; or
any combination thereof.

### Claim 24 (depends on 15)

The method of claim 15, further comprising determining said plurality of parameters for said target processing using:
said target processing result of a past target processing associated with said method;
said target processing result of a past target processing associated with a different method different than said method;
an additional processing that interacts with said method;
a computing platform that directly or indirectly interacts with said method; or
any combination thereof.

### Claim 25 (depends on 15)

The method of claim 15, further comprising requesting a server processing by a serving processor based on a plurality of routing parameters, said server processing determining said plurality of parameters for said target processing, said plurality of routing parameters comprising said first temperature, said first comfort level, and said plurality of zone conditions.

### Claim 26 (depends on 25)

The method of claim 25, wherein requesting said server processing comprises:
requesting to obtain a routing map from a route provider processor, wherein said route provider processor performs said route processing to create, update, and provide said routing map, said routing map associating a plurality of values for a routing parameter with a server processing;
identifying said server processing based on matching a value for a routing parameter of said target processing with the plurality of values for the routing parameter associated with the server processing by said routing map;
requesting to update said routing map by said router provider processor by adding a value to said plurality of values for a routing parameter associated with said server processing on learning a new association for said target processing with said serving processing; and
providing said plurality of adaptation items to perform said server processing.

### Claim 27 (depends on 25)

The method of claim 25, wherein performing said server processing for said target processing comprises:
obtaining said plurality of adaptation items for said target processing;
identifying a plurality of operation build implementations for determining said plurality of parameters for said target processing;
deriving a decision knowledge item based on said plurality of target processing results for an operation build implementation in said plurality of operation build implementations;
estimating cost of said target processing for an operation build implementation in said plurality of operation build implementations based on said decision knowledge item and said plurality of adaptation items;
selecting a final operation build implementation in said plurality of operation build implementations, said final operation build implementation matches said decision knowledge item for said plurality of adaptation items and has least estimate of cost of said target processing;
using said final operation build implementation to determine said plurality of parameter for said target processing based on said plurality of adaptation items; and
obtaining said target processing result for updating said decision knowledge item for an operation build implementation in said plurality of operation build implementations.

### Claim 28 (depends on 15)

The method of claim 15, further comprising:
regulating temperature of an elevated zone located in said dwelling at a higher elevation than said zone;
operating an upstream equalizer baffle for controlling a flow of air in an air duct from said zone to a return input of the HVAC system for said elevated zone;
operating a downstream equalizer baffle for controlling a flow of air in an air duct from said elevated zone to a return input of the HVAC system for said zone; and
controlling the operation of said upstream equalizer baffle and said downstream equalizer baffle comprising:
activating the upstream equalizer baffle to move cool air from said zone to said elevated zone when said zone system mode is a cool system mode and said elevated zone system mode is a cool system mode; and
activating the downstream equalizer baffle to move hot air from said elevated zone to said zone when said zone system mode is a heat system mode and said elevated zone system mode is a heat system mode.

### Claim 29

A computer program product comprising
a non-transitory computer readable storage medium comprising computer-readable program code bodied therewith executable by a processor to regulate temperature for a zone of a dwelling, the computer readable program code comprising:
computer readable program code configured to obtain a first system mode from a schedule for a first interval;
computer readable program code configured to obtain a first comfort range of temperature and a first comfort level from said schedule for said first interval;
computer readable program code configured to obtain a thermal profile for said zone; and
computer readable program code configured to control an HVAC system for regulating a first temperature of said zone for said first interval, wherein to control said HVAC system, the computer readable program code comprises:
computer readable program code configured to obtain said first temperature of said zone and a plurality of zone conditions for said zone based on a sensor data item from a sensor for said first interval;
computer readable program code configured to perform an adaptation processing for said first interval, wherein to perform said adaptation processing said computer readable program code is configured to:
identify a target processing to regulate said first temperature, wherein said target processing comprises:
a transit processing to transit said first temperature to a transit end temperature, said transit end temperature is above a minimum temperature of a transit comfort range of temperature for a heat system mode of said first system mode and is below a maximum temperature of said transit comfort range of temperature for a cool system mode of said first system mode, wherein to perform said transit processing the processor is configured to:
perform a first transition of said first temperature at beginning of said first interval with said transit comfort range of temperature as said first comfort range of temperature; or
perform a second transition of said first temperature at end of said first interval, wherein to perform said second transition, the processor is configured to obtain a second comfort range of temperature for a second interval in said schedule, said second interval following said first interval and requiring temperature at start of said second interval as a temperature within said second comfort range of temperature and with said transit comfort range of temperature as said second comfort range of temperature; or
a sustain processing to sustain said first temperature within said first comfort range of temperature; and
computer readable program code configured to determine a plurality of parameters for said target processing based on a plurality of adaptation items, said plurality of parameters comprising a time, a duration and at least one parameter of said HVAC system for performing said target processing, said plurality of adaptation items comprising a representation of said thermal profile, a representation of said schedule for said target processing, said plurality of zone conditions and a plurality of target processing results, wherein selection of said plurality of parameters for said target processing regulates said first temperature based on said first comfort range of temperature while minimizing energy use and maximizing user comfort;
computer readable program code configured to operate said HVAC system to perform said target processing at said time for said duration, whereby operating said HVAC system to perform said target processing regulates said first temperature of said zone as specified by said schedule; and
computer readable program code configured to add a target processing result to said plurality of target processing results, wherein said target processing result comprises said plurality of parameters for said target processing, a representation of cost of said target processing and a change in a zone condition due to said target processing.
