Operational constraint optimization apparatuses, methods and systems

• US 10,352,884 B2
• Filed: 12/01/2015
• Issued: 07/16/2019
• Est. Priority Date: 01/30/2015
• Status: Active Grant

• Current Assignee: Schneider Electric USA, Inc. (Schneider Electric SE)
• Original Assignee: Schneider Electric USA, Inc. (Schneider Electric SE)
• Inventors: Turner, Larry A., Buda, Paul Robert, Brown, Scott Robert, Pollard, Gary Brent, Glasgow, David R.
• Primary Examiner(s): Karim, Ziaul
• Assistant Examiner(s): Kabir, Saad M
• Application Number:
  US14/956,019
• Publication Number:
  US 20160223214A1
• Time in Patent Office: 1,323 Days
• Field of Search: None
• CPC Class Codes: F24F 11/30 for purposes related to the operation of the system, e.g. for safety or monitoring
• CPC Class Codes: F24F 11/46 Improving electric energy efficiency or saving
• CPC Class Codes: F24F 11/52 Indication arrangements, e.g. displays
• CPC Class Codes: F24F 11/58 using Internet communication
• CPC Class Codes: F24F 11/62 characterised by the type of control or by internal processing, e.g. using fuzzy logic, adaptive control or estimation of values
• CPC Class Codes: F24F 11/63 Electronic processing
• CPC Class Codes: F24F 11/64 using pre-stored data
• CPC Class Codes: F24F 2110/10 Temperature
• CPC Class Codes: F24F 2110/20 Humidity
• CPC Class Codes: F24F 2130/00 Control inputs relating to environmental factors not covered by group F24F2110/00
• CPC Class Codes: F24F 2130/10 Weather information or forecasts
• CPC Class Codes: F24F 2140/60 Energy consumption
• CPC Class Codes: G01N 25/18 by investigating thermal conductivity by calorimetry G01N25/20; by measuring change of resistance of an electrically-heated body G01N27/18
• CPC Class Codes: G05B 13/021 in which a variable is automatically adjusted to optimise the performance
• CPC Class Codes: G05B 15/02 electric
• CPC Class Codes: G05B 2219/2642 Domotique, domestic, home control, automation, smart house
• CPC Class Codes: G05D 23/1917 using digital means

Abstract

A system for comfort based management of thermal systems, including residential and commercial buildings with active cooling and/or heating, is described. The system can operate without commissioning information, and with minimal occupant interactions, and can learn heat transfer and thermal comfort characteristics of the thermal systems so as to control the temperature thereof while minimizing energy consumption and maintaining comfort.

Claims

1. A processor-implemented method for adaptive temperature control of Heating, Ventilation, and Air Conditioning (HVAC) unit of a thermal system, the method comprising:

receiving, at a comfort agent system, one or more thermal device properties associated with one or more volumes of the thermal system;

receiving, at the comfort agent system, one or more weather estimates associated with a surrounding environment of the thermal system;

determining a root state associated with a temperature vector comprising a plurality of temperatures associated with the one or more volumes of the thermal system, and the root state being associated with a first time, and a plurality of transitions;

calculating a plurality of states including the root state, originating at the first time and extending over a duration of time, each state of the plurality of states being associated with a respective temperature vector, a respective time, and a respective plurality of transitions;

determining one or more constraint properties, the one or more constraint properties being associated with a time originating at the first time and extending over the duration of time;

selecting an optimal path comprising one or more of the plurality of states, the optimal path;

(1) originating with the root state;

(2) including transitions of the respective plurality of transitions, (3) defining a contiguous sequence of one or more of the plurality of states extending over the duration of time, each state in the sequence of one or more of the plurality of states being associated with a unique respective time; and

(4) minimizing a cost associated with the optimal path determined by a cost function;

calculating one or more control temperature vectors each comprising a plurality of control temperatures associated with the one or more volumes of the thermal system; and

communicating the one or more control temperature vectors to one or more thermal control units of the one or more volumes of the thermal system, to regulate temperature during an active cooling mode or an active heating mode of the one or more thermal control units;

wherein the method further comprises;

estimating a sequence of optimal temperature vectors, T_o,c,n, in cool state c, at sample n, where n;

[0,N), equal to the sequence of reference temperature vectors, T_k,c,n, associated with the one or more of the plurality of states of the optimal path, calculated as a function of;

a child state value vector, s_k,c,n+1, of a plurality of [0,V^J) child states, v, an action value vector, q_u,k,c,n, associated with the transition to a plurality of [0,V^J) child states, a time discount rate, γ

, a density, V, and J volumes of the thermal system, wherein the resulting state value vector is selected to minimize a vector norm of the plurality of candidate state value vectors, according to;

2. The processor-implemented method of claim 1, wherein the one or more thermal control units of the one or more volumes of the thermal system are associated with one or more HVAC units.

3. The processor-implemented method of claim 1,

wherein the one or more thermal device properties includes at least one of:

a temperature, a humidity, an energy consumption of an HVAC unit, or a power consumption of an HVAC unit, and

wherein the thermal device properties are collected from one or more sensing devices.

4. The processor-implemented method of claim 1, wherein the one or more weather estimates includes at least one of:

a temperature, a cloud cover, a humidity, or a solar irradiance.

5. The processor-implemented method of claim 1, wherein at least one of the one or more weather estimates is received from at least one of:

one or more weather service providers, or one or more sensing devices associated with the surrounding environment of the thermal system.

6. The processor-implemented method of claim 1, wherein at least one of the one or more weather estimates is one of:

a forecast weather estimate, an historical weather estimate, or an historical weather observation.

7. The processor-implemented method of claim 1, wherein each state of the plurality of states is associated with a respective temperature vector, the respective temperature vector comprising a plurality of temperatures associated with the one or more volumes of the thermal system, wherein each temperature of the plurality of temperatures is calculated as a function of at least one of:

the one or more thermal device properties, or the one or more weather estimates.

8. The processor-implemented method of claim 1, wherein each state of the plurality of states is associated with a respective temperature vector that is calculated to define physically realizable transitions as a function of at least one of:

an energy, a power, a current, a temperature, a financial resource, a social resource, a resource availability, or a time.

9. The processor-implemented method of claim 1, wherein each state of the plurality of states is associated with a respective time that is defined at discrete constant or variable time intervals originating at the first time and extending over the duration of time.

10. The processor-implemented method of claim 1, wherein the plurality of states and associated transitions are contained in a collection, wherein the collection is a representation of at least one of:

a hypercube, a tree, or a mesh.

11. The processor-implemented method of claim 1, wherein two or more states of the plurality of states associated with the same respective time are merged into a single state if a vector norm of a difference between the respective temperature vectors associated with the two or more states is less than a temperature resolution.

12. The processor-implemented method of claim 1, wherein the one or more constraint properties includes at least one of:

an energy, a power, a current, a temperature, a financial resource, a social resource, a resource availability, or a time.

13. The processor-implemented method of claim 12, wherein the one or more constraint properties includes one or more effective temperature vectors, the effective temperature vector comprising a plurality of effective temperatures associated with the one or more volumes of the thermal system, wherein the effective temperature represents an estimate of a temperature that minimizes energy consumption due to active cooling or heating while maintaining comfort.

14. The processor-implemented method of claim 1, wherein each transition of the respective plurality of transitions is associated with a respective action value vector, the respective action value vector comprising a plurality of action values associated with the one or more volumes of the thermal system, wherein each respective action value vector represents an immediate cost of transitioning from a specific state to a child state calculated as a function of at least one of:

an energy, a power, a current, a temperature, a financial resource, a social resource, a resource availability, or a time.

15. The processor-implemented method of claim 14, wherein each state of the plurality of states is associated with a respective state value vector, the respective state value vector comprising a plurality of state values associated with the one or more volumes of the thermal system, wherein each respective state value vector represents the cost of being in a specific state calculated as a function of at least one of:

one or more state value vectors associated with one or more child states, one or more action value vectors associated with transitions to one or more child states, or a time discount rate.

16. The processor-implemented method of claim 15, wherein the selecting an optimal path comprises selecting transitions of the respective plurality of transitions to minimize a vector norm of a plurality of candidate state value vectors, where each candidate state value vector is calculated as a function of at least one of:

a state value vector associated with a child state, an action value vector associated with a transition to a child state, or a time discount rate.

17. The processor-implemented method of claim 1, wherein the cost function is a function of at least one of:

an energy, a power, a current, a temperature, a financial resource, a social resource, a resource availability, or a time.

18. The processor-implemented method of claim 1, wherein the calculating the one or more control temperature vectors is a function of at least one of:

the respective temperature vectors associated with the one or more of the plurality of states of the optimal path, or the one or more constraint properties.

19. The processor-implemented method of claim 1, wherein the method is implemented iteratively, asynchronously or synchronously, at a constant or variable frequency rate.

20. A processor-implemented method for adaptive temperature control of a Heating, Ventilation, and Air Conditioning (HVAC) unit of thermal system, the method comprising:

receiving, at a comfort agent system, one or more thermal device properties associated with one or more volumes of the thermal system;

receiving, at the comfort agent system, one or more weather estimates associated with a surrounding environment of the thermal system;

determining a root state associated with a temperature vector comprising a plurality of temperatures associated with the one or more volumes of the thermal system, and the root state being associated with a state value vector comprising a plurality of state values that represent a cost of being in a specific state, a first time, and a plurality of transitions, each transition associated with an action value vector comprising a plurality of action values which represent the immediate cost of transitioning from a specific state to a child state;

calculating a plurality of states, including the root state, originating at the first time and extending over a duration of time, each state of the plurality of states being associated with a respective temperature vector, a respective state value vector comprising a plurality of state values which represent the cost of being in a specific state, a respective time, and a respective plurality of transitions, each transition associated with an action value vector comprising a plurality of action values which represent the immediate cost of transitioning from a specific state to a child state;

determining one or more constraint properties, wherein the one or more constraint properties include at least one of;

an energy, a power, a current, a temperature, a financial resource, a social resource, a resource availability, or a time, the one or more constraint properties being associated with a time originating at the first time and extending over the duration of time;

selecting an optimal path comprising one or more of the plurality of states, the optimal path;

(1) originating with the root state;

(2) including transitions of the respective plurality of transitions, (3) defining a contiguous sequence of one or more of the plurality of states extending over the duration of time, each state in the sequence of one or more of the plurality of states being associated with a unique respective time; and

(4) minimizing the cost associated with the optimal path determined by a cost function, wherein the cost function is a function of at least one of;

an energy, a power, a current, a temperature, a financial resource, a social resource, a resource availability, or a time;

calculating one or more control temperature vectors each comprising a plurality of control temperatures associated with the one or more volumes of the thermal system, based on at least one of;

the one or more respective temperature vectors associated with the one or more of the plurality of states of the optimal path, or the one or more constraint properties; and

communicating the one or more control temperature vectors to one or more thermal control units of the one or more volumes of the thermal system, to regulate temperature during an active cooling mode or an active heating mode of the one or more thermal control units, wherein each thermal control unit is associated with one or more (HVAC) units;

wherein the method further comprises;

estimating a sequence of optimal temperature vectors, T_o,c,n, in cool state c, at sample n, where n;

[0,N), equal to the sequence of reference temperature vectors, T_k,c,n, associated with the one or more of the plurality of states of the optimal path, calculated as a function of;

a child state value vector, s_k,c,n+1, of a plurality of [0,V^J) child states, v, an action value vector, q_u,k,c,n, associated with the transition to a plurality of [0,V^J) child states, a time discount rate, γ

, a density, V, and J volumes of the thermal system, wherein the resulting state value vector is selected to minimize a vector norm of the plurality of candidate state value vectors, according to;

21. The processor-implemented method of claim 20, wherein each state of the plurality of states is associated with a respective temperature vector, the respective temperature vector comprising a plurality of temperatures associated with the one or more volumes of the thermal system, wherein each temperature of the plurality of respective temperatures is calculated as a function of at least one of:

the one or more thermal device properties, or one or more weather estimates.

22. The processor-implemented method of claim 20, wherein each state of the plurality of states is associated with a respective temperature vector that is calculated to define physically realizable transitions as a function of at least one of:

an energy, a power, a current, a temperature, a financial resource, a social resource, a resource availability, or a time.

23. The processor-implemented method of claim 20, wherein the selecting an optimal path comprises selecting transitions of the respective plurality of transitions to minimize a vector norm of a plurality of candidate state value vectors, where each candidate state value vector is calculated as a function of at least one of:

a state value vector associated with a child state, an action value vector associated with to a transition a child state, or a time discount rate.

24. The processor-implemented method of claim 20, wherein the state value vector, s_u,c,n, associated with a specific state, u, in cool state c, at sample n, where n:

[0,N), is calculated as a function of;

a child state value vector, s<sub>k,c,n−

1</sub>, of a plurality of [0,V^J) child states, v, an action value vector, q_u,k,c,n, associated with the transition to a plurality of [0,V^J) child states, a time discount rate, γ

, a density, V, and J volumes of the thermal system, wherein the resulting state value vector is selected to minimize a vector norm of the plurality of candidate state value vectors, according to;

25. The processor-implemented method of claim 20, wherein the action value vector, q_u,v,c,n, associated with the transition from a specific state, u, to a child state, v, in cool state c, at sample n, where n:

[0,N), with a sample period, T_S, a density, V, and J volumes of the thermal system, is selected from a plurality of V^J quantized vectors, Ē

_u,v,c,n, as a function of at least one of;

an available energy vector, Ē

_M,c, or an available power vector, P_M,c, according to;