Whole building air quality control system
A whole building air quality control system includes an indoor air quality (IAQ) component having at least one control state, a plurality of sensors configured to measure a plurality of building conditions of a building space, and a controller communicably coupled to the IAQ component and the plurality of sensors. The controller includes memory storing a desired air quality index (AQI). The AQI includes a categorical variable. The controller is configured to iteratively modify a control state of the IAQ component using a machine learning algorithm until the plurality of building conditions of the building space satisfy the desired AQI.
1. A whole building air quality control system, comprising:
an indoor air quality (IAQ) component having at least one control state;
a plurality of sensors configured to measure a plurality of building conditions of a building space; and
a controller communicably coupled to the IAQ component and the plurality of sensors, the controller comprising memory storing a desired air quality index (AQI) and a plurality of IAQ parameter ranges for each of the plurality of building conditions, the desired AQI comprising a categorical variable having a number of values, each value of the categorical variable corresponding to a subset of the plurality of IAQ parameter ranges for each of the plurality of building conditions, the controller configured to iteratively modify a control state of the IAQ component using a machine learning algorithm until the plurality of building conditions of the building space satisfy the desired AQI.
2. The whole building air quality control system of claim 1 , further comprising a user interface that is communicably coupled to the controller, wherein the controller is further configured to:
receive a first desired AQI;
obtain a first subset of the plurality of AQI ranges based on the desired AQI, wherein the first subset includes a first range of a first IAQ parameter and a first range of a second IAQ parameter;
iteratively modifying the control state of the IAQ component using the machine learning algorithm until the plurality of building conditions of the building space satisfies the first range of the first IAQ parameter and the second range of the second IAQ parameter;
receive a second desired AQI;
obtain a second subset of the plurality of AQI ranges based on the desired AQI, wherein the first subset includes a second range of the first IAQ parameter that is different from the first range of the first IAQ parameter and a second range of the second IAQ parameter that is different from the first range of the second IAQ parameter; and
iteratively modifying the control state of the IAQ component using the machine learning algorithm until the plurality of building conditions of the building space satisfies the second range of the first IAQ parameter and the second range of the second IAQ parameter.
3. The whole building air quality control system of claim 2 , wherein the first IAQ parameter is indicative of an amount of a pollutant in the building space and the second IAQ parameter is indicative of a level of comfort of an occupant within the building space.
4. The whole building air quality control system of claim 2 , wherein the first IAQ parameter is indicative of an amount of a first pollutant in the building space and the second IAQ parameter is indicative of a temperature of the building space.
5. The whole building air quality control system of claim 1 , wherein iteratively modifying the control state comprises:
determining a predicted control state based on the desired AQI via the machine learning algorithm;
transmitting a command to the IAQ component based on the predicted control state; and
updating the control state based on a deviation between the plurality of building conditions and the desired AQI.
6. The whole building air quality control system of claim 5 , wherein determining the predicted control state comprises determining a relationship between the categorical variable and the control state of the IAQ component using an artificial neural network.
7. The whole building air quality control system of claim 5 , wherein the categorical variable is indicative of a category of air quality, and wherein modifying the predicted control state comprises:
receiving the plurality of building conditions; and
modifying the predicted control state if at least one building condition of the plurality of building conditions does not satisfy a respective one of the IAQ parameter ranges of the subset of the plurality of IAQ parameter ranges.
8. The whole building air quality control system of claim 7 , wherein the controller is configured to modify the predicted control state based on a deviation between the at least one building condition and the respective one of the IAQ parameter ranges.
9. The whole building air quality control system of claim 1 , further comprising a user interface configured to receive user input comprising a qualitative parameter, wherein, in response to a determination that the plurality of building conditions satisfies the desired AQI, the controller is further configured to:
evaluate an objective function based on the qualitative parameter; and
iteratively modify the control state until the plurality of building conditions satisfy both the desired AQI and the objective function.
10. The whole building air quality control system of claim 9 , wherein the qualitative parameter comprises at least one of an efficiency metric that is indicative of an overall efficiency of the IAQ component and a comfort index that is indicative of a level of comfort of an occupant within the building space.
11. The whole building air quality control system of claim 9 , wherein iteratively modifying the control state until the plurality of building conditions satisfies the objective function comprises determining one of a minimum value or maximum value of the objective function using a multi-variable optimization algorithm.
12. A non-transitory computer-readable medium having instructions stored thereon that, upon execution by a computing device, cause the computing device to perform operations comprising:
receiving from a plurality of sensors, a plurality of building conditions of a building space;
receiving a desired AQI, the desired AQI comprising a categorical variable having a number of values, each value of the categorical variable corresponding to a subset of a plurality of IAQ parameter ranges for each of the plurality of building conditions;
determining a predicted control state of an IAQ component based on the desired AQI using a machine learning algorithm;
transmitting a command to the IAQ component based on the predicted control state;
iteratively modifying the predicted control state using the machine learning algorithm until the plurality of building conditions of the building space satisfy the desired AQI.
13. The non-transitory computer-readable medium of claim 12 , wherein determining the predicted control state comprises determining a relationship between the categorical variable and a control state of the IAQ component using an artificial neural network.
14. The non-transitory computer-readable medium of claim 12 , wherein the categorical variable is indicative of a category of air quality, and wherein the instructions are further configured cause the computing device to modify the predicted control state in response to a determination that at least one building condition of the plurality of building conditions does not satisfy a respective one of the IAQ parameter ranges of the subset of the plurality of IAQ parameter ranges.
15. The non-transitory computer-readable medium of claim 14 , wherein the instructions are further configured to cause the computing device to modify the predicted control state based on a deviation between the at least one building condition and the respective one of the IAQ parameter ranges.
16. The non-transitory computer-readable medium of claim 12 , wherein the instructions are further configured to cause the computing device to:
receiving a qualitative parameter;
evaluating an objective function based on the qualitative parameter; and
iteratively modifying the predicted control state until the plurality of building conditions satisfy both the desired AQI and the objective function.
17. A control device, comprising:
a communications interface configured to communicably couple the control device to an IAQ component and a plurality of sensors configured to measure a plurality of building conditions of a building space;
a user interface configured to receive user input comprising a qualitative parameter;
a memory storing a desired AQI and a plurality of IAQ parameter ranges for each of the plurality of building conditions, the desired AQI comprising a categorical variable having a number of values, each value of the categorical variable corresponding to a subset of the plurality of IAQ parameter ranges for each of the plurality of building conditions;
a processing circuit communicably coupled to the communications interface, the user interface, and the memory, the processing circuit configured to:
determine a predicted control state based on both the qualitative parameter and the desired AQI using a machine learning algorithm; and
transmit a control signal to the IAQ component based on the predicted control state.
18. The control device of claim 17 , wherein determining the predicted control state comprises determining a relationship between the categorical variable and a control state of the IAQ component using an artificial neural network.
19. The control device of claim 17 , wherein the categorical variable is indicative of a category of air quality, further comprising:
receiving, from the plurality of sensors, the plurality of building conditions of the building space; and
modifying the predicted control state in response to a determination that at least one building condition of the plurality of building conditions does not satisfy a respective one of the IAQ parameter ranges of the subset of the plurality of IAQ parameter ranges.
20. The control device of claim 19 , wherein, in response to determining that the plurality of building conditions satisfies the desired AQI, the processing circuit is configured to:
evaluate an objective function based on the qualitative parameter; and
iteratively modify the predicted control state until the plurality of building conditions satisfy both the desired AQI and the objective function.