Generation of coordinated control commands for groupings of heating/ventilation/cooling thermostats
View Patent ↗A method of generating coordinated roof top unit (RTU) control commands includes selecting a building segment, accessing a model corresponding to the segment, training the model with historical data, segment layout data, and heating/ventilation/cooling (HVC) configuration, pushing the trained model to a wireless area controller in communication with thermostats in the segment, optimizing the trained first model using aggregated monitored data provided by sensors distributed in the segment, generating control commands to coordinate the operation of the thermostats, providing the control commands to the thermostats, receiving monitored data from the sensors, aggregating the received monitored data, and the wireless area controller optimizing and/or updating the model with the aggregated received data. A system to implement the method and a non-transitory computer-readable medium containing executable instructions are also disclosed.
1. A method of generating control commands to coordinate roof top unit (RTU) operation across a segment of a building, the method comprising:
a server ( 120 ) selecting a first segment ( 106 , 108 ) from a plurality of building segments, each building segment being a volumetric portion of the building, and the server located remote from the building;
accessing a first model algorithm ( 134 ) corresponding to the first segment;
training the first model by applying at least one of historical data ( 133 ), segment layout records ( 138 ), and heating/ventilation/cooling (HVC) configuration ( 136 ), the segment layout data, and the HVC configuration each containing data corresponding to the first segment;
pushing the trained first model to a wireless area controller edge device ( 114 ), the wireless area controller edge device being in communication with one or more thermostats positioned in the first segment;
optimizing the trained first model using aggregated monitored data, the monitored data provided by sensors ( 110 , 112 ) distributed in the first segment;
generating control commands to coordinate the operation of the one or more thermostats ( 110 ) in the first segment;
providing the generated control commands to the one or more thermostats in the first segment;
receiving monitored data ( 139 ) from first segment sensors;
aggregating the received monitored data; and
the wireless area controller edge device performing at least one of an optimization and an update to the first model with the aggregated received data.
2. The method of claim 1 , the historical data containing data corresponding to the first segment.
3. The method of claim 1 , including the first model predicting the RTU operational performance for the first segment.
4. The method of claim 1 , including providing monitored time series data from the sensors, and using the monitored time series data during at least one of the training and the optimizing of the first model.
5. The method of claim 1 , including applying a predetermined boundary range in generating the control commands.
6. The method of claim 1 , including performing the optimizing of the first model at predetermined time intervals, the interval based on a monitoring interval applied to acquiring monitored data at the sensors.
7. The method of claim 1 , including the wireless area controller edge device requesting from the server an updated model for a corresponding segment.
8. The method of claim 1 , including the wireless area controller edge device communicating with other segments in the building, and creating respective aggregations of data based on monitored data received from sensors in the other segments.
9. A non-transitory computer readable medium containing computer-readable instructions ( 131 ) stored therein for causing a control processor ( 122 ) to perform operations of generating control commands to coordinate roof top unit (RTU) operation across a spatial segment of a building, the operations comprising:
a server ( 120 ) selecting a first segment ( 106 , 108 ) from a plurality of building segments, each building segment being a volumetric portion of the building, and the server located remote from the building;
accessing a first model algorithm ( 134 ) corresponding to the first segment;
training the first model by applying at least one of historical data ( 133 ), segment layout records ( 138 ), and heating/ventilation/cooling (HVC) source configuration ( 136 ), the segment layout data, and the HVC configuration each containing data corresponding to the first segment;
pushing the trained first model to a wireless area controller edge device ( 114 ), the wireless area controller edge device being in communication with one or more thermostats ( 110 ) positioned in the first segment;
optimizing the trained first model using aggregated monitored data, the monitored data provided by sensors ( 110 , 112 ) distributed in the first segment;
generating control commands to coordinate the operation of the one or more thermostats in the first segment;
providing the generated control commands to the one or more thermostats in the first segment;
receiving monitored data from first segment sensors;
aggregating the received monitored data; and
the wireless area controller edge device performing at least one of an optimization and an update to the first model with the aggregated received data.
10. The non-transitory computer readable medium of claim 9 , including executable instructions that cause the control processor to perform the method by including the first model predicting the RTU operational performance for the first segment.
11. The non-transitory computer readable medium of claim 9 , including executable instructions that cause the control processor to perform the method by providing monitored time series data from the sensors, and using the monitored time series data during at least one of the training and the optimizing of the first model.
12. The non-transitory computer readable medium of claim 9 , including executable instructions that cause the control processor to perform the method by applying a predetermined boundary range in generating the control commands.
13. The non-transitory computer readable medium of claim 9 , including executable instructions that cause the control processor to perform the method by performing the optimizing of the first model at predetermined time intervals, the interval based on a monitoring interval applied to acquiring monitored data at the sensors.
14. The non-transitory computer readable medium of claim 9 , including executable instructions that cause the control processor to perform the method by responding to a request from the wireless area controller for the server to provide an updated model for a corresponding segment.
15. The non-transitory computer readable medium of claim 9 , including executable instructions that cause the wireless area controller to receive communications from other segments in the building, and creating respective aggregations of data based on monitored data received from sensors in the other segments.
16. A system for generating control commands to coordinate roof top unit (RTU) operation across a first spatial segment of a building, the system comprising:
a first plurality of sensors ( 110 , 112 ) distributed throughout the first spatial segment ( 106 , 108 );
one or more other pluralities of sensors distributed throughout other spatial segments of the building;
sensors within each of the first plurality and the one or more other pluralities of sensors forming respective groupings of sensors for each respective spatial segment of the building;
a wireless area controller ( 114 ) in communication with each of the sensor groupings, the wireless area controller configured to receive monitored data from each sensor;
a server ( 120 ) remote from the wireless area controller, and in communication with the wireless area controller across an electronic communication network ( 140 );
a data store ( 130 ) in communication with the server, the data store containing historical data records ( 133 ), segment layout records ( 138 ) for each respective sensor grouping, heating/ventilation/cooling (HVC) source configuration ( 136 ) for each respective sensor grouping, and executable instructions ( 131 );
the server including a control processor ( 122 ) configured to access the executable instructions, the executable instructions configured to cause the control processor to perform a method including:
selecting a first segment from a plurality of building segments, each building segment being a volumetric portion of the building, and the server located remote from the building;
accessing a first model algorithm ( 134 ) corresponding to the first segment;
a model training unit ( 124 ) training the first model by applying at least one of the historical data, the segment layout record, and the HVC configuration, the segment layout record and the HVC configuration each containing data corresponding to the first segment;
pushing the trained first model to a wireless area controller edge device ( 114 ), the wireless area controller edge device being in communication with one or more thermostats ( 110 ) positioned in the first segment;
an optimization engine unit ( 126 ) optimizing the trained first model using aggregated monitored data, the monitored data provided by sensors distributed in the first segment;
generating control commands to coordinate the operation of the one or more thermostats in the first segment;
providing the generated control commands to the one or more thermostats in the first segment;
receiving monitored data from first segment sensors;
aggregating the received monitored data; and
the wireless area controller edge device performing at least one of an optimization and an update to the first model with the aggregated received data.
17. The system of claim 16 , the executable instructions configured to cause the control processor to perform the method by including:
the first model predicting the RTU operational performance for the first segment; and
determining a set point range, the set point range based on a predetermined threshold window.
18. The system of claim 16 , the executable instructions configured to cause the control processor to perform the method by including applying a predetermined boundary range in generating the control commands.
19. The system of claim 16 , the executable instructions configured to cause the control processor to perform the method by including performing the optimizing of the first model at predetermined time intervals, the interval based on a monitoring interval applied to acquiring monitored data at the sensors.
20. The system of claim 16 , the executable instructions configured to cause the control processor to perform the method by including responding to a request from the wireless area controller for the server to provide an updated model for a corresponding segment.