USOO7784704B2
`
`(12) United States Patent
`Harter
`
`(10) Patent No.:
`(45) Date of Patent:
`
`US 7,784,704 B2
`Aug. 31, 2010
`
`(54) SELF-PROGRAMMABLE THERMOSTAT
`
`(76) Inventor: Robert J. Harter, 4233 Cliffside Dr. La
`Crosse, WI (US) 54.601
`
`(*) Notice:
`
`Subject to any disclaimer, the term of this
`patent is extended or adjusted under 35
`U.S.C. 154(b) by 871 days.
`(21) Appl. No.: 11/704,414
`ppl. No.:
`9
`(22) Filed:
`Feb. 9, 2007
`
`(65)
`
`Prior Publication Data
`US 2008/O191045A1
`Aug. 14, 2008
`
`5,165,465 A * 1 1/1992 Kenet ........................ 165,111
`5,170,935 A * 12/1992 Federspiel et al. ........ 236/44 C
`5,902,183 A
`5, 1999 D’Souza
`5,943,917 A
`8/1999 Truong et al.
`6,375,087 B1
`4/2002 Day et al.
`6.741,158 B2
`5, 2004 Engler et al.
`I k
`9.
`ES s
`3. s et al.
`acklund et al.
`2003/0040842 A1
`2/2003 Poth
`2010/0025483 A1* 2/2010 Hoeyncket al. ............ 236/1 C
`* cited b
`c1ted by examiner
`Primary Examiner Frantz F. Jules
`Assistant Examiner—Alexis K Cox
`Ettorney Agent, or Firm www.bobharter; Robert J.
`
`(51) Int. Cl.
`(2006.01)
`G05D 23/32
`(57)
`ABSTRACT
`(2006.01)
`GO5D 23/OO
`(52) U.S. Cl. ..................................... 236/1 c. 236/44 C A hybrid manual programmable thermostat for a furnace or
`(58) Field of Classification Search ............... 264. R.
`air conditioner offers the simplicity of a manual thermostat
`while providing the convenience and versatility of a program
`236/46 C, 1 C, 44 C
`mable one. Initially, the hybrid thermostat appears to function
`See application file for complete search history.
`as an ordinary manual thermostat; however, it privately
`References Cited
`observes and learns a user's manual temperature setting hab
`its and eventually programs itself accordingly. If users begin
`U.S. PATENT DOCUMENTS
`changing their preferred temperature settings due to seasonal
`4.335,847 A
`6, 1982 Levine ..................... 236/46 R
`changes or other reasons, the thermostat continues learning
`4,350,966 A
`9, 1982 Nelson
`and will adapt to those changes as well. For ease of use, the
`4,467, 178 A
`8/1984 Swindle ...................... 392/449
`thermostat does not require an onscreen menu as a user inter
`4,469.274 A
`9, 1984 Levine ...
`236/46R
`face. In some embodiments, the thermostat can effectively
`4,531,064 A
`7, 1985 Levine ...
`307/66
`program itself for temperature settings that are set to occurat
`4,669,654 A
`6, 1987 Levine et al.
`particular times daily or just on weekends, yet the user is not
`4,751,961 A
`6, 1988 Levine et al.
`required to enter the time of dav or the dav of the week
`5,056,712 A * 10, 1991 Enck ........................ 236/20R
`C
`y
`y
`5,088,645 A
`2, 1992 Bell ......................... 236,46R
`5,115,967 A *
`5/1992 Wedekind ................. 236,46R
`
`(56)
`
`
`
`5 Claims, 3 Drawing Sheets
`
`se-
`24h Ray
`
`48
`
`-44
`REEME" -
`
`14
`sists
`
`407
`
`MANNALY ENTEREO STPFTEMP
`Associate WTH
`2khr and,of 2.O.YTMESTAM
`
`- 46
`
`50
`\U
`
`--
`COMPARE
`"YES
`
`CONTRTMP can
`UNTACORONY
`
`20
`
`? 52
`ge.
`
`TEMPERATRUR
`
`-
`-
`REC0SNEPATTERN WITH THE
`MANUALLY ENTRESETPTTERPSAND --54
`THER correspoils IMESTAPS
`
`ASE NPATTERNANDRCNER
`MORE OF RULES 1.io, SEABLISH
`LEARNE SETFINTTEMPANA
`LEARNETIME
`
`URINGAFIRSTPERISTARTING
`AT THE LEARNEd TIME, COMPARE
`THE LEARNESTRINTTEMPT
`THEATUAL TEMPAN CONTROL
`THE TEMPMUNTAccordNY
`
`-58
`
`PONENCOUNTERNANTHER RESNZEPATTERNR
`upon RECIEWING ANOTHERMANJAL set PoinTANDor
`APPLYNG ONE ORMORE OF rules 1-1, NITIANSA
`ScopRio ENWHCH THE TEMP COND UNIT
`SNTROLLEOACCORDINLY
`
`PETITIONER GOOGLE EX. 1019
`
`

`

`U.S. Patent
`
`Aug. 31, 2010
`
`Sheet 1 of 3
`
`US 7,784,704 B2
`
`FIG. 1
`
`
`
`TEMPERATURE
`CONDITIONING
`UNIT
`
`PETITIONER GOOGLE EX. 1019
`
`

`

`U.S. Patent
`
`Aug. 31, 2010
`
`Sheet 2 of 3
`
`US 7,784,704 B2
`
`FIG 2
`
`40
`
`38
`
`14
`
`TIMER
`24-hr and/or 7-day
`
`RECEIVE MANUAL
`SETPOINTS
`
`MANUAL
`SETPOINTS
`
`
`
`
`
`MANUALLY ENTERED SETPT TEMP
`ASSOCIATED WITH
`24-hr and/or 7-DAY TEMESTAMP
`
`46
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`COMPARE
`MANUAL SETPOINT TEMP
`TO ACTUAL TEMP AND
`CONTROL TEMP COND
`UNITACCORDINGLY
`
`
`
`52
`
`ACTUAL
`COMFORT ZONE
`TEMPERATRURE
`
`RECOGNIZE PATTERN WITH THE
`MANUALLY ENTERED SETPT TEMPS AND
`THER CORRESPONDING TIMESTAMPS
`
`
`
`54
`
`BASED ON PATTERN AND/OR ONE OR
`MORE OF RULES 1-10, ESTABLISH
`LEARNED SETPOINT TEMPANDA
`LEARNEDTIME
`
`DURINGA FIRST PERIOD STARTING
`AT THE LEARNED TIME, COMPARE
`THE LEARNED SETPOINT TEMP TO
`THE ACTUAL TEMP AND CONTROL
`THE TEMP COND UNIT ACCORDENGLY
`
`56
`
`58
`
`UPON ENCOUNTERING ANOTHER RECOGNIZED PATTERNOR
`UPON RECIEVING ANOTHER MANUAL SETPOINT AND/OR
`APPLYING ONE OR MORE OF RULES 1-10, INITIATINGA
`SECOND PERIOD IN WHICH THE TEMP COND UNIT
`IS CONTROLLED ACCORDINGLY
`
`
`
`
`
`PETITIONER GOOGLE EX. 1019
`
`

`

`U.S. Patent
`
`Aug. 31, 2010
`
`Sheet 3 of 3
`
`US 7,784,704 B2
`
`FIG. 3
`
`62
`
`SENSING THE ACTUAL ZONE TEMP
`
`42
`
`
`
`64
`
`
`
`66
`
`
`
`68
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`RECEIVE FIRST MANUALLY ENTERED SETPTTEMP
`
`CONTROL TEMPCOND UNIT IN RESPONSE TO
`FIRST MANUALLY ENTERED SETPT TEMP
`
`RECEIVE SECOND MANUALLY ENTERED SETPT TEMP
`
`CONTROL TEMPCOND UNIT IN RESPONSE TO
`SECOND MANUALLY ENTERED SETPT TEMP
`
`
`
`
`
`70
`
`72
`
`RECEIVE THIRD MANUALLY ENTERED SETPT TEMP
`
`CONTROL TEMPCOND UNIT IN RESPONSE TO
`HRD MANUALLY ENTERED SETPT TEMP
`
`74
`
`
`
`ASSIGNING TIMESTAMPS TO THE FIRST,
`SECOND AND THIRD MANUALLY ENTERED TEMPS
`
`76
`
`IDENTIFY ALEARNED SETPT TEMP BASED ON FiRST,
`SECOND AND THRD MANUALLY ENTERED TEMPS
`
`78
`
`CONTROLLING TEMP COND UNIT IN RESPONSE TO
`THE LEARNED SETPT TEMP
`
`80
`
`RECEIVE FOURTH MANUALLY ENTERED SETPT TEMP
`
`82
`
`
`
`CONTROL TEMPCOND UNIT IN RESPONSE TO
`FOURTH MANUALLY ENTERED SETPT TEMP
`
`-84
`
`SUBSEQUENTLY ONCE AGAIN CONTROLLING TEMP COND UNT
`IN RESPONSE TO THE LEARNED SETPT TEMP
`
`86
`
`PETITIONER GOOGLE EX. 1019
`
`

`

`US 7,784,704 B2
`
`1.
`SELF-PROGRAMMABLE THERMOSTAT
`
`FIELD OF THE INVENTION
`
`The Subject invention generally pertains to a room or build
`ing thermostat and more specifically to a method of program
`ming Such a thermostat, wherein the thermostat can in effect
`program itself for various daily and/or weekly temperature
`setpoints upon learning temperature setting habits of a user
`and can do Such self-programming without ever knowing the
`actual time of day or day of the week.
`
`10
`
`BACKGROUND OF RELATED ART
`
`15
`
`25
`
`30
`
`Furnaces, air conditioners and other types of temperature
`conditioning units typically respond to a thermostat in con
`trolling the air temperature of a room or other area of a
`building. Currently, thermostats can be classified as manual
`or programmable.
`With manual thermostats, a user manually enters into the
`thermostat a desired temperature setpoint, and then thermo
`stat controls the temperature conditioning unit to bring the
`actual room temperature to that setpoint. At various times
`throughout the day, the user might adjust the setpoint for
`comfort or to save energy. When operating in a heating mode,
`for instance, a user might lower the setpoint temperature at
`night and raise it again in the morning. Although manual
`thermostats are easy to understand and use, having to repeat
`edly adjust the setpoint manually can be a nuisance.
`Programmable thermostats, on the other hand, can be pro
`grammed to automatically adjust the setpoint to predeter
`mined temperatures at specified times. The specified times
`can initiate automatic setpoint adjustments that occur daily
`Such as on Monday-Friday, or the adjustments might occur
`weekly on days such as every Saturday or Sunday. For a given
`day, programmable thermostats can also be programmed to
`make multiple setpoint adjustments throughout the day. Such
`as at 8:00 AM and 11:00 PM on Saturday or at 6:00 AM and
`10 PM on Monday through Friday. Such programming, how
`ever, can be confusing as it can involve several steps includ
`40
`ing: 1) synchronizing the thermostat's clock with the current
`time of day; 2) entering into the thermostat the current date or
`day of the week; and 3) entering various chosen days, times
`and setpoint temperatures. One or more of these steps may
`need to be repeated in the event of daylight savings time,
`45
`electrical power interruption, change in userpreferences, and
`various other reasons.
`Consequently, there is a need for a thermostat that offers
`the simplicity of a manual thermostat while providing the
`convenience and Versatility of a programmed thermostat.
`
`35
`
`50
`
`SUMMARY OF THE INVENTION
`
`An object of the invention is to provide an essentially
`self-programmable thermostat for people that do not enjoy
`programming conventional programmable thermostats.
`An object of some embodiments of the invention is to
`provide a programmable thermostat that does not rely on
`having to know the time of day, thus a user does not have to
`enter that.
`Another object of some embodiments is to provide a pro
`grammable thermostat with both daily and weekly occurring
`settings, yet the thermostat does not rely on having to know
`the day of the week, thus a user does not have to enter that.
`Another object of some embodiments is to provide a pro
`grammable thermostat that does not rely on onscreen menus
`for programming.
`
`55
`
`60
`
`65
`
`2
`Another object of some embodiments is to provide a ther
`mostat that effectively programs itself as it is being used as a
`manual thermostat.
`Another object of some embodiments is to provide a ther
`mostat that automatically Switches from a manual mode to a
`programmed mode when it recognizes an opportunity to do
`SO.
`Another object of some embodiments is to provide a ther
`mostat that automatically Switches from a programmed mode
`to a manual mode simply by manually entering a new desired
`setpoint temperature.
`Another object of some embodiments is to observe and
`learn the temperature setting habits of a user and automati
`cally program a thermostat accordingly.
`Another object of some embodiments is to provide a self
`programming thermostat that not only learns a user's tem
`perature setting habits, but if those habits or temperature
`setting preferences change over time, the thermostat
`continues learning and will adapt to the new habits and set
`points as well.
`Another object of some embodiments is to minimize the
`number of inputs and actions from which a user can choose,
`thereby simplifying the use of a thermostat.
`Another object of some embodiments is to provide a ther
`mostat that can effectively self-program virtually an infinite
`number of setpoint temperatures and times, rather than be
`limited to a select few number of preprogrammed settings.
`Another object of some embodiments is to provide a simple
`way of clearing programmed settings of a thermostat.
`One or more of these and/or other objects of the invention
`are provided by a thermostat and method that learns the
`manual temperature setting habits of a user and programs
`itself accordingly.
`
`BRIEF DESCRIPTION OF THE DRAWINGS
`
`FIG. 1 is a schematic view of a thermostat controlling a
`temperature conditioning unit.
`FIG. 2 shows an example of algorithm for a thermostat
`method.
`FIG. 3 shows another example of algorithm for a thermo
`stat method.
`
`DESCRIPTION OF THE PREFERRED
`EMBODIMENT
`
`FIGS. 1-3 show athermostat 10 and a method for automati
`cally programming it. Initially, thermostat 10 might first
`appear and function as an ordinary manual thermostat. Ther
`mostat 10, for instance, includes a manual input 12 (e.g., dial,
`keyboard, pointer, slider, potentiometer, pushbutton, etc.)
`that enables a user to manually entera manual setpoint 14 that
`defines a manually entered setpoint temperature 16. The
`manually entered setpoint temperature 16 is the user's desired
`target temperature for a comfort Zone 18. Upon comparing
`the manually entered setpoint temperature 16 to the comfort
`Zone's actual temperature 20 (provided by a temperature
`sensor 22), thermostat 10 provides an output signal 24 that
`controls a temperature conditioning unit 26 (e.g., furnace,
`heater, air conditioner, heat pump, etc.) to heat or cool air 28
`in comfort Zone 18, thereby urging the comfort Zone's actual
`temperature 20 toward the manually entered setpoint tem
`perature 16.
`A digital display 30 can be used for displaying the current
`setpoint temperature, and another display 32 can show the
`comfort Zone's actual temperature. Displays 30 and 32 could
`be combined into a single display unit, wherein the combined
`
`PETITIONER GOOGLE EX. 1019
`
`

`

`3
`display unit could show the current setpoint temperature and
`the Zone's actual temperature simultaneously or in an alter
`nating manner. Thermostat 10 might also include a selector
`switch 34 for manually switching between a cooling mode for
`cooling Zone 18 and a heating mode for heating Zone 18.
`Items such as display 30, selector switch34, manual input 12,
`and output 24 are well known to those of ordinary skill in the
`art. One or more of Such items, for example, can be found in
`a model CT8775C manual thermostat provided by Honey
`well Inc. of Golden Valley, Minn.
`Although thermostat 10 can operate as a regular manual
`thermostat by controlling unit 26 as a function of a differential
`between the actual Zone temperature and the most recently
`entered manual setpoint temperature, thermostat 10 includes
`a microprocessor 36 (e.g., computer, CPU, firmware pro
`grammed chip, etc.) that enables thermostat 10 to observe the
`temperature setting habits of the user (e.g., person that manu
`ally enters setpoint temperatures into the thermostat). After
`several manual settings, microprocessor 36 may learn the
`user's preferred setpoint temperatures and timestamps them
`with the aide of a timer 38. With one or more learned setpoint
`temperatures and timestamps 48, microprocessor 36 can
`begin anticipating the user's desires and automatically adjust
`the thermostat's setpoint temperatures accordingly. Thus,
`thermostat 10 can begin operating as a programmed thermo
`stat, rather than just a manual one.
`Since a user's desired temperature setpoints and time pref
`erences might change for various reasons, any manually
`entered setpoint temperature 16 overrides the currently active
`setpoint temperature regardless of whether the current set
`point temperature was manually entered or was automatically
`activated as a learned setpoint temperature. Once overridden,
`another learned setpoint temperature might later be activated
`at a learned time to return thermostat 10 back to its pro
`grammed mode. Thus, thermostat 10 is somewhat of a hybrid
`manual/programmable thermostat in that it can shift auto
`matically between manual and programmed operation.
`To assign timestamps 48 to manually entered setpoint tem
`40
`peratures, timer 38 can actually comprise one or more timers
`and/or counters. In some embodiments, for example, timer 38
`includes a continuously running daily or 24-hour timer that
`resets itself every 24 hours. The time increments can be in
`minutes, seconds, or any preferred unit. In some cases, timer
`38 is a continuously operating weekly or 168-hour timer that
`resets itself every seven days. The increments can be in days,
`hours, minutes, seconds, or any preferred unit. The weekly
`timer could also be a seven-increment counter that indexes
`one increment every 24hours in response to a daily or 24-hour
`timer. Timer 38, however, is not necessarily synchronized
`with the actual time of day or day of the week. Such synchro
`nization preferably is not required; otherwise the user might
`have to manually enter or set the correct time and day of the
`week.
`In the case where timer 36 comprises a weekly timer in the
`form of a 7-increment counter triggered by each 24-hour
`cycle of a daily timer, timestamp 48 might a be a two-part
`number such as (X and Y) wherein X cycles from 1 to 7 as a
`weekly timer, and Y cycles from 0 to 1,439 (1,440 minutes per
`day) as a daily timer. In this case, a timestamp 48 might be (3
`and 700) to indicate 700 minutes elapsed during day-3.
`Whether day-3 represents Monday, Tuesday or some other
`day is immaterial, and whether the 700-minute represents
`2:00 AM, 7:30 PM or some other time of day is also imma
`terial. As one way to provide a programmable thermostat that
`can operate independently of an actual time of day clock and
`
`30
`
`25
`
`35
`
`45
`
`50
`
`55
`
`60
`
`65
`
`US 7,784,704 B2
`
`10
`
`15
`
`4
`to provide thermostat 10 with other functionality, micropro
`cessor 36 can be firmware programmed to execute one or
`more of the following rules:
`Rule-1. Upon receiving a manually entered setpoint tem
`perature, microprocessor assigns an (X and Y) timestamp 48
`to the manually entered setpoint temperature, wherein the
`timestamp indicates when the setpoint temperature was
`entered relative to other timestamps. The manually entered
`setpoint temperature and its timestamp 48 are stored in
`memory for later reference.
`Rule-2 Microprocessor 36 looks for patterns of manual
`setpoints, wherein each manual setpoint has a manually
`entered setpoint temperature and a timestamp 48.
`A daily pattern, for example, can be defined as three con
`secutive days in which a series of three similar manually
`entered setpoint temperatures (e.g., within a predetermined
`deviation of perhaps 2°F. or 5°F. of each other) have similar
`daily timestamps 48 (e.g., each Y-value being within a prede
`termined deviation of perhaps 90 minutes of each other).
`Such a daily pattern can then be assigned a learned daily
`setpoint temperature and a learned daily time. The learned
`daily setpoint temperature could be, for example, an average
`of the three similar manually entered setpoints temperatures
`or the most recent of the three. The learned daily time could
`be, for example, 20 minutes before the three similar times
`tamps. For future automatic settings, the 20 minutes might
`allow microprocessor 36 to activate the learned daily setpoint
`temperature before the user would normally want to adjust the
`setpoint.
`A weekly pattern, for example, can be defined as three
`manual setpoints spaced 7 days apart (e.g., same X-value
`after one complete 7-day cycle) in which three similar manu
`ally entered setpoint temperatures (e.g., within 2°F. or 5°F. of
`each other) have similar timestamps 48 (e.g., each Y-value
`being within 90 minutes of each other). Such a weekly pattern
`can then be assigned a learned weekly setpoint temperature
`and a learned weekly time. The learned weekly setpoint tem
`perature could be, for example, an average of the three similar
`manually entered setpoints temperatures spaced 7 days apart
`or the most recent of the three. The learned time could be, for
`example, 20 minutes before the three similar timestamps.
`Rule-3-Automatically activate a learned daily setpoint
`temperature at its learned daily time (at its assigned Y-value),
`whereby thermostat 10 controls unit 26 based on the learned
`daily setpoint temperature and continues to do so until inter
`rupted by one of the following: a) the user enters a manually
`entered setpoint temperature (adjusts the temp), b) another
`learned daily setpoint temperature becomes activated at its
`learned daily time, or c) a learned weekly setpoint tempera
`ture becomes activated at its learned weekly time.
`Rule-4 Automatically activate a learned weekly setpoint
`temperature at its learned weekly time (at its assigned X and
`Y values), whereby thermostat 10 controls unit 26 based on
`the learned weekly setpoint temperature and continues to do
`so until interrupted by one of the following: a) the user enters
`a manually entered setpoint temperature (adjusts the temp), b)
`a learned daily setpoint temperature becomes activated at its
`learned daily time (but see Rule-5), or c) another learned
`weekly setpoint temperature becomes activated at its learned
`weekly time.
`Rule-5-A weekly pattern overrides or Supersedes a daily
`pattern if their assigned timestamps 48 are within a predeter
`mined period of each other such as, for example, within three
`hours of each other based on theY-values of their timestamps.
`Rule-6 If a user enters a manually entered setpoint tem
`perature, thermostat 10 controls unit 26 in response to the
`manually entered setpoint temperature and continues to do so
`
`PETITIONER GOOGLE EX. 1019
`
`

`

`US 7,784,704 B2
`
`10
`
`15
`
`25
`
`30
`
`40
`
`45
`
`50
`
`55
`
`60
`
`65
`
`5
`until interrupted by one of the following: a) the user enters
`another manually entered setpoint temperature (adjusts the
`temp), b) a learned daily setpoint temperature becomes acti
`vated at its learned daily time, or c) a learned weekly setpoint
`temperature becomes activated at its learned weekly time.
`Rule-7. If a user enters two manually entered setpoint
`temperatures within a predetermined short period of each
`other, e.g., within 90 minutes of each other, the first of the two
`manual entries is disregarded as being erroneous and is not to
`be considered as part of any learned pattern.
`Rule-8 Ifa learned daily setpoint temperature is activated
`at a learned time and is soon interrupted by the user entering
`a manually entered setpoint temperature within a predeter
`mined short period (e.g., within 3 hours), and this occurs a
`predetermined number of days in a row (e.g., 3 days in a row
`as indicated by the X-value of timer 38), then the daily pattern
`associated with the learned daily setpoint temperature is
`erased from the memory.
`Rule-9 If a learned weekly setpoint temperature is acti
`vated at a learned time and is soon interrupted by the user
`entering a manually entered setpoint temperature within a
`predetermined short period (e.g., within 3 hours), and this
`occurs a predetermined number of weeks in a row (e.g., 2
`weeks in a row as indicated by an additional counter that
`counts the cycles of the X-value of timer 38), then the weekly
`pattern associated with the learned weekly setpoint tempera
`ture is erased from the memory.
`Rule-10 Actuating Switch 34 between cool and heat or
`actuating some other manual input can be used for erasing the
`entire collection of learned data.
`Rules 1-10 might be summarized more concisely but per
`haps less accurately as follows:
`1) Assign timestamps 48 to every manually entered set
`point temperature.
`35
`2) Identify daily patterns (similar manually entered tem
`peratures and times 3 days in a row), and identify weekly
`patterns (3 similar manually entered temperatures and times
`each spaced a week apart). Based on those patterns, establish
`learned setpoint temperatures and learned times.
`3) Activate learned daily setpoints at learned times, and
`keep them active until the activated setpoint is overridden by
`the next learned setpoint or interrupted by a manually entered
`setpoint.
`4) Activate learned weekly setpoints at learned times, and
`keep them active until the activated setpoint is overridden by
`the next learned setpoint or interrupted by a manually entered
`setpoint.
`5) If a learned weekly setpoint and a learned daily setpoint
`are set to occur near the same time on given day, the learned
`daily setpoint is ignored on that day because the day is prob
`ably a Saturday or Sunday.
`6) Whenever the user manually adjusts the temperature, the
`manually entered setpoint temperature always overrides the
`currently active setting. The manually entered setpoint
`remains active until it is interrupted by a Subsequent manual
`or learned setting.
`7) If a user repeatedly tweaks or adjusts the temperature
`within a short period, only the last manually entered setpoint
`temperature is used for learning purposes, as the other set
`tings are assumed to be trial-and-error mistakes by the user.
`8) If a user has to repeatedly correct a learned daily setpoint
`(e.g., correct it 3 days in a row), that learned setpoint is deleted
`and no longer used. Using 3 days as the cutoff avoids deleting
`a good daily pattern due to 2 days of corrections over a
`weekend.
`
`6
`9) If a user has to repeatedly correct a learned weekly
`setpoint (e.g., correct it 2 weeks in a row), that learned set
`point is deleted and no longer used.
`10) Switching between heating and cooling, for at least 5
`seconds or so, deletes the entire collection of learned data.
`To execute one or more of the aforementioned rules, micro
`processor 36 could operate under the control of various algo
`rithms, such as, for example, an algorithm 40 of FIG. 2, an
`algorithm 42 of FIG. 3, a combination of algorithms 40 and
`42, or another algorithm altogether.
`Referring to the example of FIG. 2, a block 44 represents
`receiving a plurality of manual setpoints 14 that are manually
`entered at various points in time over a period, each of the
`manual setpoints 14 provides a manually entered setpoint
`temperature 16 that in block 46 becomes associated with a
`timestamp 48 via timer 38.Timer 38 can run independently or
`irrespective of the actual time of day and irrespective of the
`actual day of the week. In blocks 50 and 52, thermostat 10
`controls unit 26 as a function of a differential between the
`actual Zone temperature 20 and a currently active manually
`entered setpoint. In block 54, microprocessor 36 recognizes
`patterns with the manually entered setpoints. Based on the
`patterns, in block 56 microprocessor 10 establishes learned
`setpoint temperatures and corresponding learned times. In
`block 58, some time after controlling unit 26 in response to
`the manually entered setpoint temperatures (block 50), auto
`matically Switching at the learned time to controlling the
`temperature conditioning unit in response to the learned set
`point temperature. This might continue until interrupted by
`block 60, wherein microprocessor 36 encounters another rec
`ognized pattern or upon receiving another manual setpoint, at
`which point unit 26 is controlled in response thereto.
`Referring to the example of FIG. 3, a block 62 represents
`microprocessor 36 receiving temperature feedback signal 20
`from temperature sensor 22. Sensor 22 could be incorporated
`within thermostat 10, as shown in FIG. 1, or sensor 22 could
`be installed at Some other location to sense the room tempera
`ture such as the temperature of air 28 entering unit 26. Blocks
`64, 66 and 68 represent microprocessor 36 sequentially
`receiving first, second and third manually entered setpoint
`temperatures. Blocks 70, 72 and 74 represent thermostat 10
`controlling unit 26 at sequential periods in response to a
`differential between the comfort Zone temperature and the
`various manually entered setpoint temperatures. Block 76
`represents assigning timestamps 48 to the various manually
`entered setpoint temperatures. A block 78 represents micro
`processor 36 identifying a learned setpoint temperature based
`on the first, second and third manually entered setpoint tem
`peratures. In block 80, thermostat 10 controls unit 26 in
`response to a differential between the learned setpoint tem
`perature and the actual Zone temperature. Block 82 represents
`Subsequently receiving a fourth manually entered setpoint
`temperature. Block 84 represents controlling unit 26 in
`response to the fourth manually entered setpoint temperature.
`Sometime after that, thermostat 10 returns to controlling unit
`26 in response to the learned setpoint temperature, as indi
`cated by block 86.
`Although the invention is described with respect to a pre
`ferred embodiment, modifications thereto will be apparent to
`those of ordinary skill in the art. The scope of the invention,
`therefore, is to be determined by reference to the following
`claims:
`
`The invention claimed is:
`1. A thermostat method for a temperature conditioning
`unit, wherein the temperature conditioning unit helps control
`a temperature of a comfort Zone, the method comprising:
`
`PETITIONER GOOGLE EX. 1019
`
`

`

`10
`
`15
`
`7
`receiving a first manually entered setpoint temperature,
`which is assigned a first timestamp;
`controlling the temperature conditioning unit in response
`to the first manually entered setpoint temperature;
`receiving a second manually entered setpoint temperature,
`which is assigned a second timestamp;
`controlling the temperature conditioning unit in response
`to the second manually entered setpoint temperature;
`receiving a third manually entered setpoint temperature,
`which is assigned a third timestamp;
`controlling the temperature conditioning unit in response
`to the third manually entered setpoint temperature;
`identifying a learned setpoint temperature based on the first
`manually entered setpoint temperature, the second
`manually entered setpoint temperature, and third manu
`ally entered setpoint temperature; and
`controlling the temperature conditioning unit in response
`to the learned setpoint temperature; and wherein
`the first timestamp, the second timestamp, and the third
`timestamp are based on a 24-hour timer and all lie within
`a predetermined range of each other based on the
`24-hour timer.
`2. The thermostat method of claim 1, wherein first manu
`ally entered setpoint temperature, the second manually
`entered setpoint temperature, and third manually entered set
`point temperature all lie within 5 F. of each other.
`3. The thermostat method of claim 1, further comprising:
`after controlling the temperature conditioning unit in
`response to the learned setpoint temperature, receiving a
`fourth manually entered setpoint temperature; and
`30
`after receiving the fourth manually entered setpoint tem
`perature, controlling the temperature conditioning unit
`in response to the fourth manually entered setpoint tem
`perature.
`
`25
`
`US 7,784,704 B2
`
`8
`4. The thermostat method of claim 3, further comprising:
`after controlling the temperature conditioning unit in
`response to the fourth manually entered setpoint tem
`perature, returning to controlling the temperature con
`ditioning unit in response to the learned setpoint tem
`perature.
`5. A thermostat method for a temperature conditioning
`unit, wherein the temperature conditioning unit helps control
`a temperature of a comfort Zone, the method comprising:
`receiving a first manually entered setpoint temperature,
`which is assigned a first timestamp;
`controlling the temperature conditioning unit in response
`to the first manually entered setpoint temperature;
`receiving a second manually entered setpoint temperature,
`which is assigned a second timestamp;
`controlling the temperature conditioning unit in response
`to the second manually entered setpoint temperature;
`receiving a third manually entered setpoint temperature,
`which is assigned a third timestamp;
`controlling the temperature conditioning unit in response
`to the third manually entered setpoint temperature;
`identifying a learned setpoint temperature based on the first
`manually entered setpoint temperature, the second
`manually entered setpoint temperature, and third manu
`ally entered setpoint temperature; and
`controlling the temperature conditioning unit in response
`to the learned setpoint temperature; and wherein
`the first timestamp, the second timestamp, and the third
`timestamp are based on a 168-hour timer
`
`PETITIONER GOOGLE EX. 1019
`
`