CONTROL DEVICE AND METHOD FOR CONTROLLING PERSONAL ENVIRONMENTAL COMFORT

Abstract

A control device for recurrently controlling the personal environmental comfort in a building with one or more rooms equipped with a comfort system, includes: interfaces for obtaining sensor data, operational data and external data; a database for storing these data; a first machine learning module trained using the stored data in order to generate personal preferred settings per person; a second machine learning module trained using the stored data in order to generate predictive models per room and/or per room type; and a control unit that, on the basis of the preferred settings for one or more persons and/or the predictive models, adjusts settings of one or more apparatuses in the comfort system to improve the personal environmental comfort for users of the building.

Claims

1.-15. (canceled)

16. A control device for recurrently controlling the personal environmental comfort in a building with one or more rooms equipped with a comfort system, said control device comprising: an interface for obtaining sensor data from one or more sensors in said comfort system; an interface for obtaining operational data from one or more apparatuses in said comfort system; an interface for obtaining external data from one or more sources external to said comfort system; wherein said control device also comprises the following: a database for storing said sensor data, said operational data and said external data, together referred to as stored data; a first machine learning module trained using said stored data in order to generate personal preferred settings per person; a second machine learning module trained using said stored data in order to generate predictive models per room and/or per room type, wherein a predictive model models the trend of a sensor datum, an operational datum, an external datum or a preferred setting in a formula that allows a future value thereof to be predicted; and a control unit that, on the basis of said preferred settings for one or more persons and/or said predictive models, adjusts one or more settings of one or more apparatuses in said comfort system in order to improve the personal environmental comfort for said one or more persons when present in said building.

17. The control device according to claim 16, further comprising: a short-term recommendation module configured to generate, on the basis of said preferred settings and said predictive models, a short-term recommendation for an apparatus from said comfort system, for a room from said building, or for said building, wherein said short-term recommendation comprises one or more instructions to adjust said one or more settings of one or more apparatuses within an interval of 24 hours.

18. The control device according to claim 16, further comprising: a long-term recommendation module configured to generate, on the basis of said preferred settings and said predictive models, a long-term recommendation for an apparatus from said comfort system, for a room from said building, or for said building.

19. The control device according to claim 18, further comprising: a reminder module, configured to check, via messages, whether said long-term recommendation is being followed.

20. The control device according to claim 16, further comprising: a restriction module configured to impose one or more rule-based restrictions on said control unit, wherein a rule-based restriction restricts possible adjustment of said one or more settings of one or more apparatuses in said comfort system.

21. The control device according to claim 16, further comprising: an interface for obtaining sensor data indicative of the presence of a given person in a given room or in said building.

22. The control device according to claim 16, wherein said control unit is configured to adjust, on the basis of the average of preferred settings for multiple persons, one or more settings of one or more apparatuses in said comfort system in order to improve the personal environmental comfort for said multiple persons when present in said building.

23. The control device according to claim 16, wherein said control unit is configured to compare an adjustment of a setting obtained on the basis of said predictive models with a preferred setting and to implement said adjustment only when the difference with respect to said preferred setting exceeds a predefined threshold.

24. The control device according to claim 16, wherein said control unit is configured to implement an adjustment of a setting with a delay.

25. The control device according to claim 24, wherein said delay is a personal preferred setting.

26. The control device according to claim 16, wherein said comfort system comprises a ventilation system.

27. The control device according to claim 16, wherein said comfort system comprises a sunblind system.

28. A computer-implemented method for recurrently controlling the personal environmental comfort in a building with one or more rooms equipped with a comfort system, said method comprising: obtaining sensor data from one or more sensors in said comfort system; obtaining operational data from one or more apparatuses in said comfort system; obtaining external data from one or more sources external to said comfort system; wherein said method also comprises the following: storing said sensor data, said operational data and said external data in a database, said data together referred to as stored data; training a first machine learning module using said stored data in order to generate personal preferred settings per person; training a second machine learning module using said stored data in order to generate predictive models per room and/or per room type, wherein a predictive model models the trend of a sensor datum, an operational datum, an external datum or a preferred setting in a formula that allows a future value thereof to be predicted; and adjusting one or more settings of one or more apparatuses in said comfort system on the basis of said preferred settings for one or more persons and/or said predictive models in order to improve the personal environmental comfort for said one or more persons when present in said building.

29. A computer program product comprising instructions that can be executed on a computer in order to carry out the following steps, if said program is executed on a computer, for recurrently controlling the personal environmental comfort in a building with one or more rooms equipped with a comfort system: obtaining sensor data from one or more sensors in said comfort system; obtaining operational data from one or more apparatuses in said comfort system; obtaining external data from one or more sources external to said comfort system; wherein said computer program product also comprises instructions that can be executed on a computer in order to carry out the following steps: storing said sensor data, said operational data and said external data in a database, said data together referred to as stored data; training a first machine learning module using said stored data in order to generate personal preferred settings per person; training a second machine learning module using said stored data in order to generate predictive models per room and/or per room type, wherein a predictive model models the trend of a sensor datum, an operational datum, an external datum or a preferred setting in a formula that allows a future value thereof to be predicted; and adjusting one or more settings of one or more apparatuses in said comfort system on the basis of said preferred settings for one or more persons and/or said predictive models in order to improve the personal environmental comfort for said one or more persons when present in said building.

30. A computer-readable storage medium comprising the computer program product according to claim 29.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0075] FIG. 1 is a functional block diagram of one embodiment of the control device 100 according to this invention;

[0076] FIG. 2 illustrates one embodiment of the method for recurrently controlling the personal environmental comfort in a building according to this invention; and

[0077] FIG. 3 shows a computer system suitable for carrying out one or more steps in embodiments of the method according to this invention.

DESCRIPTION OF EMBODIMENTS

[0078] FIG. 1 shows a comfort system 120 with sensors 121 and apparatuses 122. The comfort system 120 is assumed to have been fitted in a building with multiple rooms and was configured or adjusted by a fitter. The apparatuses 122 are controllable apparatuses or components that form part of the comfort system 120 such as one or more ventilation motors, valves, flow regulators, air vents, a sunshade, etc. The sensors 121 also form part of the comfort system and comprise, for example, one or more temperature sensors in the various rooms, one or more humidity sensors in the various rooms, one or more CO.sub.2 sensors in the various rooms, one or more light sensors, etc. FIG. 1 also shows external sources 123 and presence sensors 124 external to the comfort system. The external sources are sensors or other information sources that do not form part of the comfort system, such as, for example, web applications, sensors or other apparatuses that are connected via the Internet and provide weather forecasts or current temperatures. The presence sensors 124 are present in the building in which the comfort system 120 is installed and comprise, for example, infrared (IR) detectors, heat detectors, time-of-flight (TOF) detectors, 2D cameras, 3D cameras, etc., but can also comprise portable apparatuses such as a user's smartphone on which a mobile application is run which tracks the user's location and on the basis thereof reports whether the user is present in the building or is present in a certain room of the building.

[0079] FIG. 1 further shows one embodiment of the control device 100 according to the invention. This embodiment 100 comprises a first interface, 101 or IF1, configured to receive sensor data from the sensors 121 of the comfort system 120, a second interface, 102 or IF2, configured to receive operational data from the controllable apparatuses 122 of the comfort system 120, a third interface, 103 or IF3, configured to receive external data from the external sources 123, and a fourth interface, 104 or IF4, configured to receive sensor data indicative of the presence of one or more persons in one or more rooms of the building from the presence sensors 124. The interfaces 101-104 can be wireless or wired.

[0080] The data that are received via the interfaces 101-104 over the long term, meaning at least 24 hours but preferably multiple days, weeks, months or years, are stored in a database 105 that forms part of the control device 100. The database 105 will thus store historical data. The database can be installed on servers of the producer of the comfort system or in a cloud storage system that is managed by a third party, typically a cloud system operator.

[0081] A first machine learning module, 106 or ML1, is trained to generate personal preferred settings 108 per person and per room type using the historical data stored in database 105. A preferred setting 108 is a desired value for a parameter (sensor, apparatus, system or environmental parameter) in order to achieve a personal comfort preference in a given situation. The preferred setting 108 is typically a multidimensional reference point indicative of what a given person finds comfortable under certain surrounding circumstances such as the weather, the season, the day of the week, the time interval, etc. The preferred settings 108 are learnt by the machine learning module 106 per person and per room type. Thus, the first machine learning module 106 learns from the historical data stored in database 105, for example, that person 1, when they are present alone in the living room between 15:00 and 16:00 on a weekend day in October on which it will rain all day long, wants a temperature of 22.5° C. in the living room and wants the bathroom to be warmed up because the person will probably take a shower there within the next half hour. The preferred settings 108 are preferably stored per person and per room type. Thus, while a person might have a preferred temperature of 23 degrees for the living room, the same person might have a preferred temperature of 18 degrees for the bedroom.

[0082] A second machine learning module, 107 or ML2, is trained to generate predictive models 109 using the historical data stored in database 105. A predictive model 109 is a formula that models the trend of a parameter (sensor, apparatus, system or environmental parameter) or models the trend of a personal preferred setting 108 so that future values of this parameter or preferred setting 108 can be predicted using this formula. The second machine learning module 107 will thus recognize patterns in the historical data and try to predict under which surrounding circumstances a user takes a certain action.

[0083] The control device 100 from FIG. 1 further includes a control unit 110 or controller which, on the basis of the preferred settings 108, the predictive models 109 and the current surrounding circumstances 115 pulled from the current sensor data obtained from the sensors 121, and/or the current operational settings obtained from the apparatuses 122, and/or the current external data obtained from the external sources 123, and/or the current presence data obtained from the presence sensors 124, adjusts the settings of one or more controllable apparatuses 122 in order to improve personal environmental comfort for users of the dwelling. The control device 100 will thus recurrently (continuously or regularly at fixed intervals or otherwise) control the settings of the controllable apparatuses 122 in the comfort system according to the preferred settings of the users of the building so that the users experience the desired personal environmental comfort as often as possible, considering the surrounding circumstances, without having to manually adjust the settings themselves. By virtue of the predictive models, the control unit 110 will have a predictive character and will also be able to adjust the settings of controllable apparatuses 122 in anticipation of what is expected to happen in the building, for example the arrival or departure of certain users, the use of certain rooms such as the bathroom or kitchen by certain users, etc.

[0084] For non-controllable apparatuses of the comfort system 120 such as, for example, an air vent above a window, or for apparatuses that do not form part of the comfort system 120 but do have an effect on environmental comfort, such as, for example, a non-actuable sunblind, the short-term recommendation module 111 will generate instructions. In embodiments of the control device according to the invention, the short-term recommendation module 111 can also be configured to generate instructions for controllable or actuable apparatuses that form part of the comfort system 120, for example in a phase in which the control unit 110 cannot yet be relied upon because the machine learning modules 106, 107 are insufficiently trained, as a result of which the preferred settings 108 and predictive models 109 are still rapidly changing. These instructions can be delivered to a user or administrator of the building in the form of electronic messages so that actions can be taken which will further contribute toward improving the personal environmental comfort of the users of the building.

[0085] The long-term recommendation module 112 will generate long-term recommendations for the administrator of the building to make certain investments which will further improve the personal environmental comfort of the users of the building, or which make it possible to reduce the energy consumption of the building while maintaining personal environmental comfort for the users of the building so that there is a return on the investments. The reminder module 113 will monitor whether the long-term recommendation is being followed by asking the administrator of the building.

[0086] The control device 100 from FIG. 1 further includes a restriction module 114 that imposes rule-based restrictions on the control unit 110 so as to avoid that the control unit 110 contravenes local regulations or the system can no longer perform measurements. The ventilation of buildings or rooms has to comply with regulations which are often country specific. The locally applicable regulations are configured as rules by the producer or fitter of the comfort system which are used by the restriction module 114 to block adjustments to the settings of apparatuses 122 by the control unit 110 that go against the local regulations. The administrator or users of the building can also input rules which are imposed via the restriction module 114. Thus, for example, a user can indicate that a certain apparatus should not be operated, temporarily or otherwise, for example because it is being cleaned.

[0087] FIG. 2 is a pseudo flow diagram of the steps in one embodiment of the method for recurrently controlling the personal environmental comfort in a building. In step 201, sensor data are obtained from sensors in the comfort system of the building. In step 202, operational data or current settings are obtained from controllable apparatuses in the comfort system of the building. In step 203, external data are obtained from sources external to the comfort system of the building. The collected sensor data, operational data and external data are stored in a long-term database in step 204. A first machine learning module will use the stored data to generate profiles with preferred settings per user of the building and per room type, as indicated by step 205. A second machine learning module will use the stored data to generate predictive models, as indicated in step 206. The preferred settings and predictive models are then used in step 207 to recurrently control the settings of apparatuses in the comfort system for the benefit of the personal environmental comfort of the users of the building. For non-controllable apparatuses in the building which can contribute toward environmental comfort, short-term recommendations are generated in step 208. Long-term recommendations are also generated in step 209 in order to advise the administrator or a user of the building with respect to measures or investments that can be made for the benefit of personal environmental comfort and/or the energy consumption of the building. Which actions the administrator or user takes in response to the long-term recommendation is monitored in step 210. Although FIG. 2 shows steps 201-210 in sequence, it will be clear to a person skilled in the art that the steps of the method illustrated by FIG. 2 can be performed in parallel by different components of a control device for a comfort system.

[0088] FIG. 3 shows a suitable computer system 300 for carrying out one or more steps in the method of the above one or more embodiments. Computer system 300 can generally take the form of a computer suitable for general purposes with a bus 310, a processor 302, a local memory 304, one or more optional input interfaces 314, one or more output interfaces 316, a communication interface 312, a storage element interface 306 and one or more storage elements 308. Bus 310 can comprise one or more conductors, which enable communication between the components of the computer system 300. Processor 302 can comprise any type of conventional processor or microprocessor which interprets and executes program instructions. Local memory 304 can comprise a random-access memory (RAM) or another type of dynamic storage device which stores information and instructions for execution by processor 302, and/or a read-only memory (ROM) or another type of static storage device which stores static information and instructions for use by processor 302. Input interface 314 can comprise one or more conventional mechanisms which allow an operator to input information into the computer device 300, such as a keyboard 320, a mouse 330, a stylus, voice recognition and/or biometric mechanisms, a touchscreen, etc. Output interface 316 can comprise one or more conventional mechanisms which provide information to the operator, such as a display 340, a printer 350, a speaker, etc. Communication interface 312 can comprise a transceiver-type mechanism, such as for example one or more Ethernet interfaces, which allow the computer system 300 to communicate with other devices and/or systems 381, 382, 383, for example a transceiver for communicating with the comfort system 120 in FIG. 1. The communication interface 312 of computer system 300 can be connected to another computer system by means of a local area network (LAN) or a wide area network (WAN), such as for example the Internet. Storage element interface 306 can comprise a storage interface, such as for example a Serial Advanced Technology Attachment (SATA) interface or a Small Computer System Interface (SCSI), for connecting bus 310 with one or more storage elements 308, such as one or more local disks, for example SATA disk drives, and for reading and writing data to and/or from these storage elements 308. Although the storage elements 308 above are described as a local disk, in general any other suitable medium that can be read by computer, such as a removable magnetic disk, optical storage media, such as a CD or DVD, CD-ROM, solid-state drives, or flash memory cards, can be used. The system 300 described hereinabove can also operate as a virtual machine on top of the physical hardware.

[0089] The steps described in the above one or more embodiments can be implemented as program instructions, stored in the local memory 304 of the computer system 300, for execution by the processor 302 thereof. Alternatively, the instructions can be stored in the storage element 308 or be accessible from another computer system via the communication interface 312.

[0090] Although the present invention has been illustrated by means of specific embodiments, it will be clear to a person skilled in the art that the invention is not limited to the details of the above illustrative embodiments, and that the present invention can be carried out with various changes and modifications without thereby departing from the area of application of the invention. Therefore, the present embodiments have to be seen in all areas as being illustrative and non-restrictive, and the area of application of the invention is described by the attached claims and not by the above description, and any changes which fall within the meaning and scope of the claims are therefore incorporated herein. In other words, it is assumed that this covers all changes, variations or equivalents which fall within the area of application of the underlying basic principles and the essential attributes of which are claimed in this patent application. In addition, the reader of this patent application will understand that the terms “comprising” or “comprise” do not exclude other elements or steps, that the term “a(n)/one” does not exclude the plural and that a single element, such as a computer system, a processor or another integrated unit, can perform the functions of various auxiliary means which are mentioned in the claims. Any references in the claims cannot be interpreted as a limitation of the respective claims. The terms “first”, “second”, “third”, “a”, “b”, “c” and the like, when used in the description or in the claims, are used to distinguish between similar elements or steps and do not necessarily indicate a sequential or chronological order. In the same way, the terms “top side”, “bottom side”, “above”, “below” and the like are used for the sake of the description and do not necessarily refer to relative positions. It should be understood that these terms are interchangeable under the appropriate circumstances and that embodiments of the invention can function according to the present invention in different sequences or orientations than those described or illustrated above.