HVAC CONTROL SYSTEM AND METHOD

Abstract

An HVAC control system includes a user interface adapted to receive comfort feedback from a user. The processor is adapted to receive the comfort feedback and input data corresponding to one or more comfort factors and respective values ​​of the comfort factors. The processor is adapted to build a comfort model for the user based on both the comfort feedback and the input data, the comfort model correlating a comfort score to one or more values ​​of one or more of the comfort factors. The processor adjusts one or more functions of the HVAC unit to vary the one or more comfort factors so that a predetermined comfort score is achieved. Related methods and systems are also provided.

Description

[Technical field]

[0001] The present invention relates to an HVAC control system and a method for controlling an HVAC unit. [Background technology]

[0002] Power utilities and power grid managers face a common problem: the total electrical load varies significantly over time. Since the power grid has a fixed maximum capacity, unforeseen events can threaten the supply-demand balance, and although surplus power can be stored, peak demand loads can lead to blackouts. Power utilities are therefore challenged to implement peak load shedding.

[0003] Generally, residential power consumption is a small portion of total power consumption (15-30%), but during system peaks, residential power consumption can exceed total power consumption by as much as 50%. Much of this residential power consumption is due to air conditioning, which during peak times can exceed the household's power demand by as much as 50%.

[0004] Traditionally, load management has been done using variable pricing, for example higher energy prices during certain times of the day when peak loads are more likely to occur. The problem is that this is not as predictable as utilities would like. These price incentives do not provide guaranteed reductions or direct control over the load.

[0005] Recently, the advent of smart grids has enabled direct load control through automatic demand response (ADR or DR) programs. In particular, ADR programs can be used to reduce consumer demand during peak periods by directly communicating with consumer appliances that are part of the smart grid and controlling their energy consumption.

[0006] Because heating, ventilation, and air conditioning (HVAC) systems account for a large portion of the grid load, many ADR programs are targeted at the direct control of air conditioners and their energy consumption.

[0007] Basic ADR programs operate by cycling appliances on and off, while more advanced programs operate by adjusting temperature settings (often a fixed set point, or a fixed offset of the current set point). Cycling can be accomplished with load control switches, but adjusting the HVAC set point requires that the HVAC is programmable and connected to the smart grid.

[0008] While ADR programs allow utilities to achieve better load control, consumers are free to opt out and withdraw from the programs at any time. One reason for this is the crude adjustments mandated by these programs. For example, mandating a total offset of the current setpoint or a specific on-off cycle for all users will inevitably lead to a large number of users opting out. This limits the success and effectiveness of these ADR programs.

[0009] Residential ADR programs targeting air conditioning appliances have been piloted in several countries, with the most widespread programs in the U.S. These ADR trials have been widely studied, and a sampling of such studies is presented below in Table 1.

[0010] [Table 1]

[0011] As can be seen from the above sample, a key issue in these ADR trial studies is the relatively high opt-out rate (typically 20-30%). Opt-out rates are generally: i. Outdoor weather conditions (estimated to increase by 3.6% per 1°F); ii. Day of the week (Thursday vs. Wednesday: 3.6% increase); iii. DR event duration (increases by 3.3% per hour); iv. Incentive Structure This is influenced by several factors, including:

[0012] According to a 2010 KEMA analysis, the main reason for opting out is related to users' thermal discomfort. Based on a survey of DR program participants who opted out, 68% of users indicated that thermal discomfort issues were the reason for opting out.

[0013] Another related problem is the difficulty in determining how to control and regulate an HVAC system to satisfy a user, especially multiple users, and in particular, determining the set points required for the HVAC unit to achieve the performance required to satisfy the users.

[0014] It is an object of the present invention to overcome or ameliorate at least one of the disadvantages of the prior art, or to provide a useful alternative. Summary of the Invention

[0015] In a first aspect, the present invention provides a method for producing a composition comprising the steps of: a user interface adapted to receive comfort feedback from a user; a processor adapted to receive comfort feedback and input data corresponding to one or more comfort factors and respective values ​​of the comfort factors, and adapted to build a comfort model for the user based on the comfort feedback and the input data, the comfort model correlating a comfort score to one or more values ​​of one or more of the comfort factors, and the processor adjusting one or more functions of the HVAC unit to vary the one or more comfort factors such that a predetermined comfort score is achieved; The present invention provides an HVAC control system comprising:

[0016] An embodiment of the present invention provides in a second aspect a method of controlling an HVAC unit, the method comprising: receiving comfort feedback from a user; Receiving input data corresponding to one or more comfort factors and respective values ​​of the comfort factors; building, with a processor, a comfort model for the user based on the comfort feedback and the input data, the comfort model correlating a comfort score to one or more values ​​of one or more of the comfort factors; using a processor to adjust one or more functions of the HVAC unit to vary one or more comfort factors such that a predetermined comfort score is achieved; Equipped with.

[0017] Other features and embodiments of the invention can be found in the appended claims.

[0018] Throughout this specification, including the claims, the words "comprises," "comprising," and other similar terms are to be construed in their inclusive sense, i.e., "including but not limited to," and not in their exclusive or exhaustive sense, unless otherwise expressly stated or unless the context clearly requires. [Brief description of the drawings]

[0019] Preferred embodiments in accordance with the best mode of the present invention will now be described, by way of example only, with reference to the accompanying drawings in which like reference numerals refer to similar parts throughout, unless otherwise specified, and in which:

[0020] [Figure 1] FIG. 1 is a schematic diagram of an HVAC control system according to an embodiment of the present invention. [Diagram 2] FIG. 2 is a schematic diagram of an HVAC control system according to another embodiment of the present invention. [Diagram 3] 1 is a graph of comfort levels experienced by various users as temperature and humidity vary over time. [Figure 4] 1 is a graph of actual current comfort level and user desired comfort level compared to the set point of an AC unit during a DR event. [Diagram 5] 13 is a graph of actual current comfort level and user desired comfort level compared to the set point and comfort offset of an AC unit during a DR event with a comfort offset according to an embodiment of the present invention. [Figure 6] 1 is an infographic for the Fixed Temperature Offset DR program showing users' relative satisfaction with Fixed Temperature Offset. [Figure 7] 1 is an infographic of a fixed temperature offset DR program showing relative user satisfaction with a fixed temperature offset compared to using a comfort offset according to an embodiment of the present invention. [Figure 8] 1 is an infographic of a fixed temperature offset DR program showing relative user satisfaction with a fixed temperature offset compared to relative user satisfaction with a comfort offset according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0021] With reference to the figure, an HVAC control system 1 is provided that comprises a user interface 2 adapted to receive comfort feedback from a user. A processor 3 is adapted to receive the comfort feedback and input data corresponding to one or more comfort factors and respective values ​​of the comfort factors. The processor 3 is adapted to build a comfort model 4 for the user based on the comfort feedback and the input data, the comfort model 4 correlating a comfort score to one or more values ​​of the comfort factors. The processor 3 adjusts one or more functions of an HVAC unit 5 to change one or more comfort factors such that a predetermined comfort score is achieved. The HVAC unit 5 is a heating, ventilation, air conditioning unit or any other electrical appliance that changes or brings about a change in a comfort factor or any other environmental condition that affects the comfort factor.

[0022] In the illustrated embodiment, the processor 3 may form part of a controller 6 for regulating one or more functions of the HVAC unit 5. For example, the controller 6 may be in the form of a stand-alone controller unit that regulates the functions of the HVAC unit 5 by sending infrared signals to an infrared receiver of the HVAC unit 5. The infrared receiver may be the same infrared receiver that typically receives infrared signals from a hand-held remote control.

[0023] In other embodiments, the processor 3 may form part of a server located remotely from the controller 6. The server and therefore the processor 3 are connected to the controller 6 via a communication network, which may be a wireless communication network. For example, typical communication protocols are cellular, Wi-Fi, and Bluetooth.

[0024] In other embodiments, the processor 3 may be located in one or more of a local server, a remote server (as described above), a cloud server, a smart phone, a laptop computer, and a desktop computer. The processor 3 may be multiple physical processors and therefore may be located in multiple locations. This provides redundancy or more processing power by allowing alternative physical processors to be used if one or more physical processors are unavailable.

[0025] In another embodiment, the one or more user interfaces 2 are adapted to receive comfort feedback from a plurality of users. The processor 3 is adapted to build a respective comfort model 4 for each user based on both the comfort feedback and the input data for each user. The respective comfort model 4 correlates a respective comfort score for each user to one or more values ​​of one or more comfort factors affecting the respective user. The processor 3 adjusts one or more functions of the one or more HVAC units 5 to vary the one or more comfort factors such that a respective predetermined comfort score for each user is achieved.

[0026] When one or more of the HVAC units 5 correspond to a respective one of the plurality of users, the processor 3 adjusts one or more functions of each HVAC unit 5 to vary one or more comfort factors so as to achieve a predetermined comfort score for each of the plurality of users.

[0027] For example, in a multi-room building where each room is served by one HVAC unit 5, if there is only one user per room, the processor 3 adjusts one or more functions of the HVAC unit 5 servicing each room to vary one or more comfort factors such that a predetermined comfort score for the user is achieved. If there are multiple users in a room, the processor 3 adjusts one or more functions of the HVAC unit 5 servicing that room to vary one or more comfort factors such that an average predetermined comfort score for the users in that room is achieved.

[0028] The comfort model 4 is built for each user and is personalized for that user based on comfort feedback from that user and on input data about that user corresponding to one or more comfort factors affecting the user and the respective values ​​of these comfort factors. The comfort model 4 is a machine learning regression model or any other model based on artificial intelligence or machine learning algorithms.

[0029] The comfort model 4 may be implemented by the processor 3 with or without the aid of a memory unit associated with the processor. The memory unit may be locally connected to the processor 3. For example, if the processor 3 forms part of the controller 6, the memory unit may also be located in the controller 6. Alternatively, the memory unit may be located remotely, for example in a remote server. The memory unit may be multiple physical memory units and thus located in multiple locations. This allows for redundancy or more memory space, where alternative physical memory units can be used if one or more physical memory units are unavailable.

[0030] The memory unit also records input data corresponding to one or more comfort factors and respective values ​​of the comfort factors. The input data may be either or both of currently received input data or historical input data received in the past. The comfort factors may be one or more of temperature, humidity, light intensity, time of day, occupancy, user identifier, and any other factor that affects the comfort of a user. The user identifier may be used to identify a particular user such that a comfort model 4 for the particular user may be identified and used by the HVAC control system 1. The input data may be provided via one or more sensors. For example, the sensors may include one or more of a temperature sensor, a humidity sensor, a light intensity sensor, a motion sensor, and an access control sensor.

[0031] The comfort feedback may be in the form of a descriptive expression. Advantageously, this means that the HVAC control system 1 is easy to use for all types of users. The descriptive expression may be in the form of a comfort opinion, advantageously not necessarily in metric or numerical form, such as a temperature setting. For example, the descriptive expression may be "hot", "cold", "warm", "cool", "comfortable", "humid" or "dry". Any natural language descriptor of a comfort factor or environmental condition may be used.

[0032] The comfort score is a discrete or continuous numeric range. For example, the comfort score may be a discrete numeric range, such as integers from -3 to 3. Alternatively, the comfort score may be a continuous numeric range, such as real numbers from -3 to 3. A continuous numeric range has the advantage that the comfort score may be of any precision. For example, one discrete numeric range is:

[0033] +3...Hot +2...too warm +1... a little warm 0. Comfortable -1... a little cold -2: Too cold -3 Freezing

[0034] A continuous numerical range is the numbers listed in the discrete numerical ranges above as well as all real numbers in between. This means, for example, that a comfort score of 0.5 may be assigned halfway between "comfortable" and "slightly warm". As another example, a comfort score of -2.5 means halfway between "too cold" and "freezing". The comfort score is calculated by the comfort model 4. The use of numerical ranges provides an easy and consistent indication of the comfort level for a particular user, regardless of how the user expresses the comfort feedback. As mentioned above, the comfort feedback may be expressed as a descriptive expression or natural language descriptor, which is much easier and more natural for the user.

[0035] Each predefined comfort score may be a comfort score within a predefined comfort offset range. That is, rather than a single numerical value, the predefined comfort score may be a range of comfort scores defined by an offset range. This provides the HVAC control system 1 with some flexibility to achieve the predefined comfort score, especially in the case of system limitations, such as, for example, power derating commands as described below. In the above example, the comfort scores are numbers within a continuous numerical range from -3 to 3. The predefined comfort score may be 0. However, comfort offset ranges may be used. For example, the predefined comfort offset range may be +0.5, meaning that the predefined comfort score is any number between 0 and +0.5. The predefined comfort score is simply a comfort score preselected, for example as a target or threshold. The predefined comfort offset range is simply a comfort offset range preselected, for example as a target or threshold. The predefined comfort offset range and the comfort offset range may also be referred to as predefined comfort offset and comfort offset, respectively, for brevity.

[0036] In this embodiment, the processor 3 is adapted to receive a power derating command from a power utility company and adjust one or more functions of one or more HVAC units 5 to vary one or more comfort factors such that one or more of the predetermined comfort scores are achieved while complying with the power derating command.

[0037] The power reduction command may be issued under an automatic demand response (ADR or DR) program. ADR programs may be used to reduce consumer demand during peak periods by communicating directly with consumer appliances that are part of the smart grid and controlling their energy consumption. Such communications include power reduction commands, also referred to as DR events or DR triggers. Basic ADR programs operate by issuing power reduction commands to HVAC units or systems enrolled in the ADR program that instruct the HVAC units to repeatedly turn on and off, while more advanced programs operate by issuing power reduction commands that adjust temperature settings. This temperature setting adjustment is often in the form of a fixed set point or a fixed offset of the current set point.

[0038] However, since the embodiment of the present invention is based on achieving a comfort score rather than a fixed temperature set point or a fixed offset of the current set point, a power derating command based on a fixed temperature set point or a fixed offset of the current set point may be adhered to even while achieving a comfort score. This is because the comfort score is correlated to one or more values ​​of one or more comfort factors. This means that a certain range of temperatures may correlate to the same comfort score. This is believed to be because a user is not very sensitive to a certain temperature range. Alternatively, a combination of a temperature range and other values ​​of a comfort factor may correlate to the same comfort score. For example, a combination of a temperature change and a change in light intensity or humidity may not result in a discernible change in comfort by the user, and therefore correlates to the same comfort score. As a result, a predetermined comfort score for a user may be achieved while adhering to a power derating command based on a fixed temperature set point or a fixed offset of the current set point.

[0039] If the predetermined comfort score is a comfort score within a predetermined comfort offset range, then of course, this allows for greater flexibility in achieving the predetermined comfort score (i.e., a comfort score within a predetermined comfort offset range) while still adhering to power derating commands, such as those based on a fixed temperature set point or a fixed offset of the current set point.

[0040] In some embodiments, the present invention contemplates that a power derate command corresponds to a power derate value, and that a power derate command is complied with if the power derate value is achieved for a predetermined period of time.

[0041] To this end, the processor 3 calculates a mapping between, for example, a fixed set point offset (provided by the power derating command) and a corresponding comfort score (either a single comfort score or a range of comfort offsets) that corresponds to the fixed set point offset when averaged over a predefined period of time. One particular way to do this is to use the processor 3 to simulate how the HVAC unit 5 would be deployed with varying assumed comfort scores. Then, at any given time, the set point to be deployed for a given comfort score can be calculated. By considering the difference between this assumed set point and the actual set point to be deployed, the set point offset corresponding to the given comfort score can be inferred. By averaging the set point offsets over a predefined period of time and performing a grid search for various comfort scores, the above-mentioned mapping can then be obtained.

[0042] In other embodiments, the present invention contemplates that the power derating commands correspond to a total power derating value across multiple HVAC units, and that the power derating commands for each user are complied with when the total power derating value is achieved.

[0043] To this end, the processor 3 calculates a mapping between, for example, a fixed set point offset (provided by the power derating command) and a corresponding comfort score (either a single comfort score or a range of comfort offsets) that corresponds to the fixed set point offset when averaged across multiple users. One particular way to do this is to use the processor 3 to simulate how the HVAC unit 5 would be deployed with varying assumed comfort scores. Then, at any given time for a given number of users, the set point that would be deployed for a given comfort score can be calculated. By considering the difference between this assumed set point and the actual set point that is deployed, the set point offset that corresponds to the given comfort score can be inferred. By averaging the set point offsets over a number of users in a certain period of time and performing a grid search for various comfort scores, the above-mentioned mapping can then be obtained.

[0044] The above example is based on the power derate commands being temperature setpoint offsets for each HVAC unit, although the same principles and similar calculations can be made for other types of power derate commands.

[0045] The one or more functions of the HVAC unit 5 that are adjusted by the processor 3 may be any function that changes or brings about a change in one or more of the comfort factors. Typically, the functions of the HVAC unit 5 that are adjusted are one or more of the following: temperature setting, mode, fan speed, humidity setting, and on / off setting. In one embodiment, the functions of the HVAC unit 5 that are adjusted are those that are normally adjusted manually by a user or by a remote control. For example, the functions may be those that are normally adjusted using infrared signals from a remote control operated by a user. In this case, the HVAC control system is used instead of a remote control by including a controller 6 that is adapted to send infrared signals to the HVAC unit 5. Advantageously, this allows the HVAC control system 1 to be retrofitted to existing HVAC units 5. If there are multiple HVAC units 5, there may be multiple processors 3, one for each HVAC unit 5. In this case, there may be multiple processors 3, one for each HVAC unit 5, or there may be one processor 3 that controls all HVAC units 5 individually. In other embodiments, the functions of the HVAC unit 5 that are adjusted include those that are not typically adjustable by a user. For example, the processor 3 may control the power supply of the HVAC unit 5. In this case, the HVAC unit 5 may include an integrated control unit connected to or in communication with the processor 3.

[0046] Typically, the user interface 2 is in the form of a software application. The user interface 2 may be implemented in one or more of a personal device, a smartphone, a laptop computer, a desktop computer, and a terminal. For example, the user interface 2 may be an app that can be downloaded to a smartphone. In this case, the processor 3 may be the smartphone under the direction of the app. Alternatively, the processor 3 may be in another device, for example the controller 6 or a remote server, and the user interface 2 in the form of an app communicates with the processor 3. The communication may be via a cellular network, Wi-Fi, Bluetooth, or any other communication protocol.

[0047] In another aspect, the present invention provides an embodiment of a method for controlling an HVAC unit. In one embodiment, the method comprises receiving comfort feedback from a user, receiving input data corresponding to one or more comfort factors and respective values ​​of the comfort factors, constructing, with a processor 3, a comfort model 4 for the user based on the comfort feedback and the input data, the comfort model correlating a comfort score to one or more values ​​of one or more of the comfort factors, and adjusting, with the processor 3, one or more functions of the HVAC unit 5 to vary the one or more comfort factors such that a predetermined comfort score is achieved.

[0048] In another embodiment of the method, comfort feedback is received from a plurality of users. Based on both the comfort feedback and the input data for the respective users, a respective comfort model 4 is constructed for the respective user. The respective comfort model 4 correlates a respective comfort score for the respective user to one or more values ​​of one or more comfort factors affecting the respective user. One or more functions of the one or more HVAC units 5 are adjusted to change the one or more comfort factors such that a respective predetermined comfort score for the respective user is achieved.

[0049] In an additional embodiment, the method includes receiving a power derating command from a power utility and, using a processor 3, adjusting one or more functions of one or more HVAC units 5 to vary one or more comfort factors such that one or more predetermined comfort scores are achieved while complying with the power derating command.

[0050] Further embodiments and features of the method are apparent from the above description, particularly with regard to the HVAC control system 1.

[0051] Embodiments of the present invention are particularly advantageous when used in conjunction with ADR programs that issue power derating commands, also known as DR events or DR triggers, to reduce power consumption. Power derating commands are issued by utility companies to directly control or mandate operational limitations of electrical appliances, such as, for example, HVAC units 5. Such power derating commands include fixed temperature setpoints or fixed temperature setpoint offsets (i.e., fixed offsets of the current setpoints) for each HVAC unit 5.

[0052] The inventors performed an analysis on the impact of a 2°C temperature offset on users' comfort levels (thermal comfort). As best shown in FIG. 6, the analysis revealed that approximately 26% of users perceive a +2°C temperature offset as "too warm," which closely matches the empirically observed opt-out rate from past studies. Based on their comfort profile, these users are likely to be sensitive to thermal discomfort, may also be sensitive to thermal comfort factors other than temperature, such as humidity, or may live in homes with poor insulation or high exposure to thermal radiation.

[0053] Of the remaining 74% of users, our analysis and comfort model predicts that 39% will feel "a little warm," which is considered an appropriate level of reduction. However, for an additional 34% of users, our analysis and comfort model predicts that the users will only feel the conditions as slightly warm (and may barely notice the difference). This means that for these users, the reduction is not enough, and there is an opportunity to increase the temperature offset even further without causing them undue discomfort.

[0054] Rather than using a typical fixed "one size fits all" temperature offset, embodiments of the present invention create custom reduction levels for each user based on knowing the user's personal comfort preferences. As described above, this is done by building a comfort model 4 for each user based on both comfort feedback and input data for that user, which correlates a comfort score for that user to one or more values ​​of one or more comfort factors that affect that user.

[0055] Under such a regime, the temperature offset is reduced for heat-sensitive segments (i.e., people predicted to be "too warm" in the analysis above). A +2°C offset is maintained for the 39% of users who are appropriately curtailed, and the temperature offset is increased for users who are likely to be able to maintain comfort at higher temperatures. Figures 7 and 8 show this temperature offset distribution using comfort model 4 according to an embodiment of the present invention. By implementing DR events based on comfort model 4 rather than fixed cycle rates and fixed offsets or set points, opt-out rates can be better controlled and power reduction or mitigation can be maximized or better managed. The HVAC control system 1 according to an embodiment of the present invention directly controls HVAC energy consumption by dynamically adjusting the set points of HVAC units and other consumer appliances based on comfort levels and comfort model 4.

[0056] The comfort model 4 can be trained in a supervised manner, learning and generalising from comfort feedback provided by the user via the user interface 2 (e.g. the 7-point scale from "freezing" to "hot" mentioned above).

[0057] After training, the comfort model 4 is able to predict how comfortable a user will feel given some particular comfort factors. The comfort model 4 outputs a continuous comfort level ranging from -3 ("freezing") to 3 ("hot"), where 0 represents "comfortable", which is the continuous counterpart of the discrete 7-point comfort feedback used by users.

[0058] The same indoor conditions may be felt differently by two different users, which is reflected by the comfort prediction provided by the comfort model 4. The user identifier may be part of the comfort factors and thereby can be associated with the user's individual comfort model 4, which provides a personalized comfort output.

[0059] For example, if user A is more sensitive to humidity than user B under the same indoor conditions, comfort model 4 will provide a personalized comfort prediction that reflects the users' thermal preferences. This is best shown in Figure 3.

[0060] In one embodiment, to determine which state the HVAC unit 5 will evolve into, the processor 3 or controller 6 operates as a closed feedback loop controller. The controller constantly adjusts the HVAC settings (power, mode, set temperature) to minimize tracking error.

[0061] In the case of a thermostat, a typical controller attempts to minimize the difference between the desired room temperature and the current room temperature.

[0062] In an embodiment of the invention, the goal is not to maintain room temperature, but to adjust the HVAC settings (or other functions) so that the room conditions are comfortable for the user (according to comfort model 4). To do so, the processor 3 or controller 6 minimizes a tracking error based on the user's comfort, i.e., the difference between a given user's desired comfort level (obtained from comfort model 4) and the current comfort level (0 for "comfortable"). This is best shown in FIG. 1.

[0063] Upon a power derate command (ie, a DR event or a DR trigger), the goal is to utilize the comfort model and the processor 3 to control the HVAC unit 5 to reach a particular comfort level.

[0064] The comfort score (including a single comfort score or a comfort offset range) associated with the current DR event is then used to calculate a desired comfort score that is provided to processor 3. This enables processor 3 to deploy a higher than normal set point that results in reduced energy consumption while still maintaining a comfort score for the user or a comfort score within the comfort offset range (e.g., a comfort offset of +0.5).

[0065] Compared to traditional DR event implementations with fixed offsets, comfort variance comes at the expense of setpoint variance, i.e., temperature offsets are deployed to users, and the offsets are fixed in space (same for all users) and fixed in time (do not change during the course of the event), resulting in greater variance in user comfort and leading to opt-outs.

[0066] In contrast, with an embodiment of the present invention based on Comfort Model 4, a comfort score or comfort offset range is applied to the user, and the score or offset is fixed in space and time, resulting in a stable and similar level of comfort for the user throughout the DR event. The resulting deployed set point offset varies throughout space and time, providing personalized comfort and compensating for constant variations in comfort factors. This is best shown in Figures 4 and 5.

[0067] While traditional DR event implementations feature fixed set temperature offsets, embodiments of the present invention implement DR events based on comfort scores (including a single comfort score or comfort scores within a predefined comfort offset range) based on a comfort model 4. In one embodiment, the HVAC control system 1 calculates a mapping between fixed set point offsets and corresponding comfort scores or offsets that correspond, on average (over space and time), to similar or the same set point offsets.

[0068] To that end, the processor 3 is used to simulate in which state the HVAC unit would be deployed if the assumed comfort score or offset were changed. Thus, at any given time for a given number of users, the set point to be deployed for a given comfort score or offset can be calculated. By considering the difference between this assumed set point and the actual set point to be deployed, the set point offset corresponding to the given comfort score or offset can be inferred. See FIG. 2.

[0069] The above mapping can be obtained by averaging the setpoint offsets over a period of time and over a number of users and performing a grid search for various comfort scores or offset values.

[0070] The following example shows the mapping between comfort offset ranges and setpoint offsets. [Table 2]

[0071] It is also understood that the above-described embodiments are merely exemplary embodiments used to explain the principles of the present invention, and the present invention is not limited thereto. Those skilled in the art can make various modifications and alterations without departing from the spirit and essence of the present invention, and these modifications and alterations are also included within the scope of the present invention. Thus, although the present invention has been described with reference to specific examples, those skilled in the art will understand that the present invention can be embodied in many other forms. It is also understood that the features of the various examples described can be combined in other combinations. The invention as originally claimed in the present application is set forth below. [1] a user interface adapted to receive comfort feedback from a user; a processor adapted to receive the comfort feedback and input data corresponding to one or more comfort factors and respective values ​​of the comfort factors, and adapted to construct a comfort model for the user based on the comfort feedback and the input data; the comfort model correlating a comfort score to one or more values ​​of one or more of the comfort factors; the processor adjusts one or more functions of the HVAC unit to vary one or more comfort factors such that a predetermined comfort score is achieved; HVAC control systems. [2] the one or more user interfaces are adapted to receive comfort feedback from a plurality of users; the processor is adapted to construct a respective comfort model for the respective user based on both the comfort feedback and the input data relating to the respective user; the respective comfort model correlating a respective comfort score for the respective user to one or more values ​​of one or more of the comfort factors affecting the respective user; the processor adjusts one or more functions of one or more of the HVAC units to vary one or more comfort factors such that a respective predetermined comfort score for each user is achieved; [1] 2. The HVAC control system according to claim 1 . [3] one or more of the HVAC units corresponding to a respective one of a plurality of users; the processor adjusts one or more functions of each HVAC unit to vary one or more comfort factors such that a respective predetermined average comfort score for each of the plurality of users is achieved; [2] The HVAC control system according to the present invention. [4] Each predetermined comfort score is a comfort score within a predetermined comfort offset range. 1. An HVAC control system according to any one of claims 1 to 3. [5] The processor is adapted to receive a power reduction command from a power utility; the processor adjusts one or more functions of one or more of the HVAC units to vary one or more comfort factors such that one or more of the predetermined comfort scores are achieved while complying with the power derating commands. 1. An HVAC control system according to any one of claims 1 to 4. [6] the power derate command corresponds to a power derate value; the power derating command is complied with if the power derating value is achieved for a predetermined period of time; [5] ,The HVAC control system described in. [7] the power derating command corresponds to a total power derating value across the plurality of HVAC units; the power reduction command for each user is complied with if the total power reduction value is achieved; 1. The HVAC control system according to claim 5 or 6. [8] the power derate command is a temperature setpoint offset for each HVAC unit; 1. An HVAC control system according to any one of claims 5 to 7. [9] The comfort factor is one or more of temperature, humidity, light intensity, time of day, occupancy rate, and user identifier; 1. An HVAC control system according to any one of claims 1 to 8.

[10] the function of the HVAC unit being adjusted is one or more of a temperature setting, a mode, a fan speed, a humidity setting, and an on / off setting; 1. An HVAC control system according to any one of claims 1 to 9.

[11] The comfort feedback is in the form of a descriptive expression. 1. An HVAC control system according to any one of claims 1 to 10.

[12] The comfort score is a discrete or continuous numerical range. 1. An HVAC control system according to any one of claims 1 to 11.

[13] The comfort model is a machine learning regression model. 1. An HVAC control system according to any one of claims 1 to 12.

[14] the user interface being in the form of a software application; 1. An HVAC control system according to any one of claims 1 to 13.

[15] The user interface is implemented in one or more of a personal device, a smartphone, a laptop computer, a desktop computer, and a terminal; 1. An HVAC control system according to any one of claims 1 to 14.

[16] the processor is located on one or more of a local server, a remote server, a cloud server, a smartphone, a laptop computer, and a desktop computer; 1. An HVAC control system according to any one of claims 1 to 15.

[17] the input data being provided via one or more sensors; 16. An HVAC control system according to any one of claims 1 to 16.

[18] the sensors include one or more of a temperature sensor, a humidity sensor, a light intensity sensor, a motion sensor, and an access control sensor;

[17] The HVAC control system described.

[19] 1. A method for controlling an HVAC unit, comprising: receiving comfort feedback from a user; receiving input data corresponding to one or more comfort factors and respective values ​​of said comfort factors; constructing, with a processor, a comfort model for the user based on the comfort feedback and the input data, the comfort model correlating a comfort score to one or more values ​​of one or more of the comfort factors; adjusting, with the processor, one or more functions of the HVAC unit to vary one or more comfort factors such that a predetermined comfort score is achieved; method.

[20] Comfort feedback is received from a plurality of users; a respective comfort model is constructed for each user based on both the comfort feedback and the input data for the respective user; the respective comfort model correlating a respective comfort score for the respective user to one or more values ​​of one or more of the comfort factors affecting the respective user; one or more functions of one or more of said HVAC units are adjusted to vary one or more comfort factors such that a respective predetermined comfort score for each user is achieved; The method described in

[19] .

[21] Received a power reduction order from the power company, using a processor to adjust one or more functions of one or more of the HVAC units to vary one or more comfort factors such that one or more of the predetermined comfort scores are achieved while complying with the power derating commands; The method according to any one of

[19] to

[21] .

Claims

1. One or more user interfaces adapted to receive comfort feedback from a plurality of users; a processor adapted to receive the comfort feedback and input data corresponding to one or more comfort factors and respective values ​​of the comfort factors, the processor is adapted to construct a respective comfort model for the respective user based on both the comfort feedback and the input data relating to the respective user; the respective comfort model correlating a respective comfort score for the respective user to one or more values ​​of one or more of the comfort factors affecting the respective user; the processor adjusts one or more functions of one or more HVAC units to vary one or more comfort factors such that a respective predetermined comfort score for each user is achieved; wherein the processor is adapted to receive a power derating command from a utility company, and the processor adjusts one or more functions of one or more of the HVAC units to vary one or more comfort factors such that one or more of the predetermined comfort scores are achieved while complying with the power derating command; wherein the power derating command corresponds to a total power derating value across a plurality of the HVAC units, and the power derating command for each user is complied with if the total power derating value is achieved; wherein the one or more comfort models are machine learning regression models. HVAC control system.

2. one or more of the HVAC units serving a respective one of a plurality of users; the processor adjusts one or more functions of each HVAC unit to vary one or more comfort factors such that a respective predetermined average comfort score for each of the plurality of users is achieved. The HVAC control system of claim 1 .

3. Each predetermined comfort score is a comfort score within a predetermined comfort offset range. The HVAC control system according to claim 1 or 2.

4. the power derate command corresponds to a power derate value; the power derating command is complied with if the power derating value is achieved for a predetermined period of time; The HVAC control system of claim 1 .

5. the power derate command is a temperature setpoint offset for each HVAC unit; The HVAC control system according to any one of claims 1 to 4.

6. The comfort factors are one or more of temperature, humidity, light intensity, time of day, occupancy rate, and user identifier.

6. The HVAC control system according to claim 1 .

7. the function of the HVAC unit being adjusted is one or more of a temperature setting, a mode, a fan speed, a humidity setting, and an on / off setting; 7. The HVAC control system according to claim 1.

8. The comfort feedback is in the form of a descriptive expression.

8. The HVAC control system according to claim 1 .

9. The comfort score is a discrete or continuous numerical range.

9. The HVAC control system according to any one of claims 1 to 8.

10. the user interface being in the form of a software application; 10. The HVAC control system of claim 1.

11. The user interface is implemented in one or more of a personal device, a smartphone, a laptop computer, a desktop computer, and a terminal; 11. The HVAC control system of claim 1.

12. the processor is located on one or more of a local server, a remote server, a cloud server, a smart phone, a laptop computer, and a desktop computer; 12. The HVAC control system of claim 1.

13. the input data is provided via one or more sensors; 13. The HVAC control system of any one of claims 1 to 12.

14. the sensors include one or more of a temperature sensor, a humidity sensor, a light intensity sensor, a motion sensor, and an access control sensor; 14. The HVAC control system of claim 13.

15. 1. A method for controlling an HVAC unit, comprising: receiving comfort feedback from a plurality of users; receiving input data corresponding to one or more comfort factors and respective values ​​of said comfort factors; using a processor to construct a respective comfort model for the respective user based on both the comfort feedback and the input data for the respective user, where the respective comfort model correlates a respective comfort score for the respective user to one or more values ​​of one or more of the comfort factors affecting the respective user; using a processor to adjust one or more functions of one or more of said HVAC units to vary one or more comfort factors such that a respective predetermined comfort score for each user is achieved; receiving a power derating command from a utility company and, with a processor, adjusting one or more functions of one or more of said HVAC units to vary one or more comfort factors such that one or more of said predetermined comfort scores are achieved while complying with said power derating command; wherein the power derating command corresponds to a total power derating value across a plurality of the HVAC units, and the power derating command for each user is complied with if the total power derating value is achieved; wherein the one or more comfort models are machine learning regression models. method.

Images