Predictively controlling an environmental control system

Patent metadata

Field	Value
Publication number	US-2016201933-A1
Application number	US-201514597196-A
Country	US
Kind code	A1
Filing date	Jan 14, 2015
Priority date	Jan 14, 2015
Publication date	Jul 14, 2016
Grant date	—

Abstract

In an embodiment, an electronic device may include a power source configured to provide operational power to the electronic device and a processor coupled to the power source. The processor may be configured to generate temperature predictions using a model of a structure and possible control scenarios, determine a value of the temperature predictions and the respective possible control scenarios using a cost function, the cost function comprising weighted factors related to an error between a setpoint temperature and the temperature predictions, a length of runtime for an environmental control system (e.g., an HVAC system), and a length of environmental control system cycles. The processor may also be configured to select the control scenario with the highest value to apply to control the environmental control system. The control scenarios may be generated using upper confidence bound for trees (UCT).

First claim

1 . A non-transitory computer readable medium storing instructions thereon, the instructions, when executed by a processor of an electronic device, configured to:
generate temperature predictions of temperatures of a structure for a forthcoming period of time using a model of a structure and possible control scenarios to control an environmental control system for the forthcoming period of time; determine a value of the temperature predictions and the respective possible control scenarios using a cost function, the cost function comprising weighted factors related to an error between a setpoint temperature and the temperature predictions, a length of runtime for the environmental control system based on the control scenarios, and a length of environmental control system cycles based on the control scenarios; and select the control scenario with the highest value to apply to control the environmental control system.

2 . The computer-readable medium of claim 1 , wherein the selected control scenario provides improved environmental control system efficiency and temperature comfort relative to bang-bang control.

3 . The computer-readable medium of claim 1 , wherein the possible control scenarios are generated using upper confidence bound for trees (UCT) by iteratively simulating a plurality of control scenarios including a control action at each of a plurality of time steps over a period of time and assigning values to each of the control actions based on the value determined by the cost function.

4 . The computer-readable medium of claim 3 , wherein UCT filters out control scenarios that generate a value below a threshold while simulating the plurality of control scenarios.

5 . The computer-readable medium of claim 3 , wherein the selected control scenario comprises a set of control actions with the highest upper confidence bound on performance according to the values assigned by UCT.

6 . The computer-readable medium of claim 3 , wherein an interval length between the plurality of time steps is modified based on the efficiency of the environmental control system, wherein the interval length is set shorter for more efficient environmental control systems and longer for less efficient environmental control systems.

7 . The computer-readable medium of claim 1 , wherein the weights associated with each factor of the cost function are adjusted based on user preference.

8 . The computer-readable medium of claim 1 , wherein the control scenario comprises a control action to execute at each of a plurality of time steps for a period of time.

9 . The computer-readable medium of claim 8 , wherein the model accounts for upcoming setpoint changes and weather forecasts.

10 . A method, comprising:
generating, via a processor, temperature predictions of temperatures of a structure for a forthcoming period of time using a model of the structure and possible control scenarios to control an environmental control system for the forthcoming period of time; determining, via the processor, a value of the temperature predictions and the respective possible control scenarios using a cost function, the cost function balancing a weighted error between a setpoint temperature and the temperature predictions against a weighted runtime of the environmental control system based on the control scenarios and a weighted length of environmental control system cycles based on the control scenarios; and selecting, via the processor, the control scenario with the highest value to apply to control the environmental control system.

11 . The method of claim 10 , wherein the processor only applies the control scenario while indoor temperature is within a maintenance band surrounding the setpoint temperature and applies bang-bang control when the indoor temperature exceeds the maintenance band.

12 . The method of claim 11 , comprising expanding the maintenance band when an upcoming setpoint change is detected to enable the control scenario to execute preemptive actions including pre-cooling, pre-heating, or pre-drifting.

13 . The method of claim 10 , wherein the number of control scenarios generated is based on configuration settings, an amount of battery of an electronic device in which the processor is disposed, an amount of time available to perform the simulations, whether the electronic device is in use by a user, or some combination thereof.

14 . A method, comprising:
assigning, via a first processor, a value to each of a plurality of possible control scenarios useable to control an environmental control system for a forthcoming period of time and respective temperature predictions of temperatures of a structure for the forthcoming period of time generated by a model of the structure using a cost function that comprises factors related to a weighted error between a setpoint temperature and the temperature predictions, a weighted runtime of an environmental control system determined based on the control scenarios, and a weighted length of environmental control system cycles determined based on the control scenarios; and selecting, via the first processor, the control scenario with the highest value to apply to control the environmental control system.

15 . The method of claim 14 , comprising waking up the first processor, via a second processor, if the first processor is sleeping when an expected amount of time has elapsed, an indoor temperature has been reached, or it is time to switch environmental control system states.

16 . The method of claim 14 , comprising determining whether the difference in actual temperature and the predicted temperature meets or exceeds a threshold and, if so, cancelling the selected control scenario and applying another control strategy, such as bang-bang control.

17 . An electronic device, comprising:
a power source configured to provide operational power to the electronic device; and a processor coupled to the power source, the processor being configured to:
generate temperature predictions of temperatures of a structure for a forthcoming period of time using a model of the structure and possible control scenarios to control an environmental control system for the forthcoming period of time;
determine a value of the temperature predictions and the respective possible control scenarios using a cost function, the cost function comprising weighted factors related to an error between a setpoint temperature and the temperature predictions, a length of runtime for the environmental control system based on the control scenarios, and a length of environmental control system cycles based on the control scenarios; and
select the control scenario with the highest value to apply to control the environmental control system.

18 . The electronic device of claim 17 , wherein the model comprises a thermal model of the structure and analyzes factors comprising current indoor temperature, indoor temperature on a previous time step, indoor temperature rate, current environmental control system state, number of seconds into a current environmental control system state, time of day, outdoor temperature, current schedule mode, and target temperature.

19 . The electronic device of claim 17 , wherein the model is updated at least once a day with 18 days of past data.

20 . The electronic device of claim 17 , wherein the error is determined by an absolute error between indoor temperature and the setpoint temperature, or the error is determined by a one-sided temperature error on the side that is less comfortable.

Assignees

• Google Inc

Inventors

• Hester Todd Andrew
• Fisher Evan Jarman
• Khandelwal Piyush

Classifications

• F24F11/48

  prior to normal operation, e.g. pre-heating or pre-cooling · CPC title

• F24F11/526

  giving audible indications · CPC title

• F24F11/58

  using Internet communication · CPC title

• F24F11/64

  using pre-stored data · CPC title

• F24F2120/00

  Control inputs relating to users or occupants · CPC title