Method and system for HVAC control with model-driven deep learning

Abstract

To provide an HVAC control system with model-driven deep learning.SOLUTION: A method includes the steps of: operating equipment to affect a variable state or condition of a space; and determining a set of learned weights for a neural network by modeling an estimated cost of operating the equipment over a plurality of simulated scenarios. Each simulated scenario includes simulated measurements relating to the space. The neural network is configured to generate simulated control dispatches for the equipment based on the simulated measurements. The method also includes the steps of: configuring the neural network for online control by applying the set of learned weights; applying actual measurements relating to the space to the neural network to generate a control dispatch for the equipment; and controlling the equipment in accordance with the control dispatch.SELECTED DRAWING: Figure 13

Description

[Technical Field]

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS This application is a continuation of U.S. Provisional Patent Application No. 62 / 673,479, filed May 18, 2018. and the specification of U.S. Provisional Patent Application No. 62 / 673,496 filed May 18, 2018. Both of these patent applications claim the benefit of and priority to the present application and are incorporated herein by reference in their entirety. INCORPORATED INTO THE SPECIFICATION. [Background technology]

[0002] This disclosure generally relates to a variable refrigerant flow (VRF) system that provides temperature control for a building. refrigerant flow) system, room air conditioning (RAC: room air r conditioning) system, or packaged air conditioning (PAC) Managing energy costs in aged air conditioning systems Minimizing the energy consumption of such systems is crucial to achieving thermal comfort. This can lead to discomfort for building occupants as the temperature cannot be maintained without increasing power, while Accurately matching residential preferences at all times usually leads to high energy costs. Reduce the energy consumption of VRF, RAC and PAC systems without causing discomfort to the patient. What is needed is a system and method for reducing [Prior art documents] [Patent documents]

[0003] [Patent Document 1] U.S. Patent Application Serial No. 15 / 953,324 [Patent Document 2] U.S. Patent Application Serial No. 15 / 426,962 Summary of the Invention [Means for solving the problem]

[0004] One embodiment of the present disclosure is a method. The method includes: and operating the device across multiple simulated scenarios. A set of learned weights for the neural network is determined by modeling the cost. Each simulated scenario includes simulated measurements related to a space. The network simulates control dispatch of equipment based on simulated measurement results. The method is also configured to generate a control dispatch. Neural networks for online control by applying a set of learned weights. The neural network is used to generate the process and equipment control dispatch that constitutes the work. A process for applying actual measurements related to the space of the work and controlling the equipment according to the control dispatch. and

[0005] In some embodiments, a set of learned weights is used across multiple simulated scenarios. The cost is determined as a set of weights that minimizes the estimated cost of operating the equipment.

[0006] In some embodiments, determining the set of learned weights comprises determining the set of learned weights in a state space of the space. identifying a thermal model and defining a cost function using the state-space thermal model; Each scenario's neural network generates a set of simulated measurements and weights for the scenario. generating a simulated control dispatch based on the assignment; Given the simulated measurement results of , the estimated cost of operating the equipment over the simulated duration of the scenario is and calculating the set of learned weights by using a cost function. The determining step also modifies a set of weights to drive the cost toward a minimum of the cost function. This modified set of learned weights is then used to optimize the model across multiple scenarios. determining as the set of weightings that results in the lowest cost.

[0007] In some embodiments, the control dispatch includes a temperature set point, a temperature schedule, a humidity Set points, airflow set points, power levels, on / off settings, damper positions, fan speeds, compressor frequency, or resource consumption allocation. In this embodiment, the device is connected to one or more of the air-side system or the water-side system. In some embodiments, the equipment includes a variable refrigerant flow system, a room air conditioner, or or packaged air conditioners.

[0008] Another embodiment of the present disclosure is a system. The system is adapted to respond to variable states or conditions of a space. HVAC equipment that can be operated to affect the performance of the system and collect measurements related to the space. One or more sensors configured in a single system and HVAC equipment across multiple simulated scenarios By modeling the estimated costs of the operations, a set of trained neural networks is generated. and an offline training system configured to determine the weightings for each model. The simulated scenario includes simulated measurements related to the space. The neural network then The system is configured to generate simulated control dispatches for HVAC equipment based on the results of the simulation. The system also has a neural network configured according to a set of learned weights. Measurements from one or more sensors to generate control dispatches for HVAC equipment The results are applied to a neural network to control HVAC equipment according to the control dispatch. and an online control circuit configured to:

[0009] In some embodiments, the offline training system may provide training over multiple simulated scenarios. The set of learned weights is used as the set of weights that minimizes the estimated cost of standing and operating the equipment. In some embodiments, offline training is configured to determine The system identifies a state-space thermal model of the space and calculates the cost function using the state-space thermal model. Define,the neural network, and simulate the measurement results for each scenario. and generating a simulated control dispatch based on a set of weightings, and Given a set of simulated measurements, operate the HVAC equipment over a simulated period of the scenario. The method is configured to calculate the estimated cost of producing the object by using a cost function. In an embodiment, the offline training system drives the cost towards the minimum of the cost function. The set of weights is modified to set the weights, and this modified set of learned weights is used as a The cost is determined as a set of weights that results in the minimum cost across a number of scenarios. .

[0010] In some embodiments, the HVAC equipment is an air-side system or a water-side system. In some embodiments, the online control circuitry includes one or more of the following: C equipment, and offline training systems can be used with one or more Includes cloud computing resources.

[0011] Another embodiment of the present disclosure is a system. The system includes: and one or more sensors configured to collect measurements related to the space. models multiple sensors and the estimated cost of operating a chiller across multiple simulated scenarios to determine a set of learned weights for the neural network by and an offline training system configured in accordance with the present invention. Each simulated scenario relates to a space. The neural network simulates the control of the cooling machine based on the simulated measurement results. The system is also configured to generate a set of learned weights. Generate control dispatches for chillers with neural networks configured according to the applying measurements from one or more sensors to a neural network to generate a an online control circuit configured to control the chiller in accordance with the control dispatch; include.

[0012] In some embodiments, the offline training system may provide training over multiple simulated scenarios. A set of learned weights is used to minimize the estimated cost of operating the chiller. In some embodiments, the offline training is configured to determine the weighting. The system identifies a state-space thermal model of the space and uses the state-space thermal model to calculate the cost function. Define,the neural network, and simulate the measurement results for each scenario. and generating a simulated control dispatch based on a set of weightings, and Given a set of simulated measurements, operate the chiller for a simulated period of the scenario. The estimated cost is calculated using a cost function. In this state, the offline training system drives the cost towards the minimum of the cost function. The set of weights is modified to achieve this, and this modified set of learned weights is used to The weightings are configured to determine the set of weights that yields the minimum cost over the scenario.

[0013] In some embodiments, the chiller is a room air conditioner, a packaged air conditioner, or a variable refrigerant In some embodiments, the flow control circuit includes one or more of the flow devices. is included locally with the chiller, and the offline training system is one or more In some embodiments, the cloud computing resources include a control dispatch contains the temperature set point. [Brief explanation of the drawings]

[0014] [Figure 1] FIG. 1 is an illustration of a building with an HVAC system in accordance with an illustrative embodiment. [Figure 2] FIG. 2 is a block diagram of a water side system that may be used to service the building of FIG. 1 in accordance with an exemplary embodiment. [Figure 3] FIG. 2 is a block diagram of an airside system that may be used to service the building of FIG. 1 in accordance with an exemplary embodiment. [Figure 4] 2 is a block diagram of a building management system (BMS) that may be used to monitor and control the building of FIG. 1 in accordance with an example embodiment. [Figure 5] FIG. 2 is a block diagram of another BMS that may be used to monitor and control the building of FIG. 1 in accordance with an example embodiment. [Figure 6] 1 is an illustration of a building served by a variable refrigerant flow system according to an exemplary embodiment. [Figure 7] FIG. 7 is a diagram of the variable refrigerant flow system of FIG. 6 according to an exemplary embodiment. [Figure 8] FIG. 2 is a detailed diagram of a variable refrigerant flow system according to an exemplary embodiment. [Figure 9] FIG. 1 is a block diagram of a window air conditioner according to an exemplary embodiment. [Figure 10] FIG. 1 is a block diagram of a room air conditioning system in accordance with an exemplary embodiment. [Figure 11] FIG. 1 is a block diagram of a packaged air conditioner system according to an exemplary embodiment. [Figure 12] FIG. 1 is a block diagram of a system manager in accordance with an example embodiment. [Figure 13] FIG. 1 is a block diagram of an offline training system in accordance with an illustrative embodiment; DETAILED DESCRIPTION OF THE INVENTION

[0015] Construction of HVAC systems and building management systems 1-5, in accordance with some embodiments, the systems and methods of the present disclosure Several building management systems (BMS) and HVAC systems are shown in which the In overview, FIG. 1 shows a building 10 equipped with an HVAC system 100. FIG. FIG. 3 is a block diagram of a water side system 200 that may be used to serve a water supply. 1 is a block diagram of an airside system 300 that may be used to service a building 10. FIG. 4 is a block diagram of a BMS that may be used to monitor and control building 10. FIG. 5 is a block diagram of another BMS that may be used to monitor and control building 10. do.

[0016] Building and HVAC Systems With particular reference to FIG. 1, a perspective view of a building 10 is shown. The building 10 is serviced by a BMS. A BMS typically controls equipment in or around a building or building area. A BMS is a system of devices configured to manage, monitor, and control the , HVAC systems, security systems, lighting systems, fire alarm systems, building machinery any other system, or any combination thereof, that is capable of managing a function or device It may include a combination.

[0017] The BMS serving the building 10 includes an HVAC system 100. 00 configured to provide heating, cooling, ventilation or other services to a building 10 Numerous HVAC devices (e.g., heaters, chillers, air handling units, pumps, fans) For example, the HVAC system 100 may include a water supply system. The water-side system is shown to include a water-side system 120 and an air-side system 130. The system 120 provides a heated or cooled fluid to the air handling units of the air-side system 130. The air side system 130 can use a heating or cooling fluid to The airflow provided to the object 10 can be heated or cooled. An example of an exemplary water-side system and air-side system that can be used is shown in FIG. This will be explained in more detail with reference to 3.

[0018] The HVAC system 100 includes a chiller 102, a boiler 104, and a rooftop air handling unit. The water supply system 120 is shown to include a boiler 104. and a refrigerator 102 is used to heat or cool a working fluid (e.g., water, glycol, etc.). The working fluid can be circulated to the AHU 106. In some embodiments, the HVAC devices of the waterside system 120 may be located within or adjacent to the building 10. They may be located in the vicinity (as shown in Figure 1) or at an off-site location such as a central plant. can be located in a facility (e.g., a chiller plant, steam plant, heat plant, etc.). The working fluid is supplied to the boiler 10 depending on whether heating or cooling is required in the building 10. The boiler 104 can be heated by a boiler 104 or cooled by a refrigerator 102. By burning a substance (e.g., natural gas) or by using an electric heating element The refrigerator 102 is used to absorb heat from the circulating fluid. A circulating fluid is in a heat exchange relationship with another fluid (e.g., a refrigerant) in a heat exchanger (e.g., an evaporator). The working fluid from the chiller 102 and / or boiler 104 can be placed in a pipe. 108 to AHU 106.

[0019] The AHU 106 is configured to place the working fluid in a heat exchange relationship with the airflow flowing through the AHU 106. (e.g., via one or more stages of cooling and / or heating coils) The airflow can be, for example, outside air, return air from within the building 10, or a combination of both. The HU 106 is a device that connects an airflow and a working fluid to provide heating or cooling to the airflow. For example, the AHU 106 may be configured to transfer heat over a heat exchanger containing a working fluid. one or more fans or blowers configured to move air through a heat exchanger or The working fluid is then passed through piping 110 to the refrigerator 102 or boiler 103. You can go back to 04.

[0020] The air supply system 130 is supplied by the AHU 106 via the supply air duct 112. The supplied airflow (i.e., supply airflow) can be delivered to the building 10 via return air duct 114. In some embodiments, the building 10 may provide return air to the AHU 106. , an air-side system 130 includes a plurality of variable air volume (VAV) units 116. For example, the airside system 130 may include a separate VAV unit for each floor or zone of the building 10. The VAV units 116 are shown to include individual zones of the building 10. A damper or other fan that can operate to control the amount of intake airflow provided to the fan. In another embodiment, the air side system 130 may include an intermediate VAV The supply air flow is directed to one of the buildings 10 without the use of unit 116 or other flow control elements. or to multiple zones (e.g., via supply ducts 112). Various sensors configured to measure attributes of the intake airflow (e.g., temperature sensors, pressure sensors) The AHU 106 may include a power supply (such as a power supply or a heater) to achieve setpoint conditions for the building zone. receive input from sensors located within the AHU 106 and / or within the building zones to , the flow rate, temperature or other attributes of the supply airflow in the AHU 106 may be adjusted.

[0021] Water supply system Referring now to FIG. 2, a block diagram of a water side system 200 according to some embodiments. In various embodiments, the water-side system 200 may be a water-side system that is integrated with the HVAC system 10. 100 or HVAC system 100. When implemented in the HVAC system 100, the water-side system 200 is a subset of HVAC devices in the HVAC system 100 (e.g., boiler 1 04, refrigerator 102, pumps, valves, etc.) to transfer the heated or cooled fluid to the AH The HVAC devices of the water side system 200 may be operable to supply water to the U106. within the building 10 (e.g., as part of the water-side system 120) or at a central plant, etc. It may be located off-site.

[0022] In FIG. 2, the water supply system 200 includes a number of sub-plants 202 to 212. Subplants 202-212 are shown as heater subplants 202-212. 02, Heat recovery chiller sub-plant 204, chiller sub-plant 206, cooling tower sub-plant Subplant 208, Thermal Energy Storage (TES) Subplant 210 and Cold Energy Storage (TES) subplant 212. Subplants 202-212 is the thermal energy load (e.g., hot and cold water, heating, cooling, etc.) of a building or campus. To provide this, they consume resources (e.g., water, natural gas, electricity, etc.) from utilities. For example, the heater subplant 202 may be configured to provide a heating system between the heater subplant 202 and the building 10. The water can be heated in a hot water loop 214 that circulates the water. The chiller subplant 206 circulates chilled water between the chiller subplant 206 and the building 10. The chiller can be configured to cool water in a chilled water loop 216. Subplant 204 is used to provide additional heating of hot water and additional cooling of cold water. The water loop 216 may be configured to transfer heat to the hot water loop 214. The cooling tower loop 218 absorbs heat from the chilled water of the chiller subplant 206 and The heat absorbed by 208 can be rejected or the heat absorbed can be transferred to the hot water loop 214. The TES subplant 210 and the cold TES subplant 212 are for later use. It is possible to store both hot and cold energy.

[0023] Hot water loop 214 and chilled water loop 216 provide heated and / or chilled water to the building 10 to an air handling unit (e.g., AHU 106) located on the roof of the building or to an individual floor or The air handler may heat the air or send it to a zone (e.g., VAV unit 116). or a heat exchanger through which water flows to provide cooling (e.g., a heating coil or cooling coil) The heated or cooled air serves the thermal energy load of the building 10. The water can then be pumped to individual zones of the building 10 for further heating or cooling. The reactor returns to subplants 202-212 for cooling.

[0024] Subplants 202-212 shall heat and cool water for circulation to the building. Although shown and described as such, it may be used instead of or to supply the thermal energy load. In addition to water, any other type of working fluid (e.g., glycol, CO2, etc.) can be used. It will be appreciated that in other embodiments, the subplants 202-212 may be Direct heating and / or cooling to a building or campus without the need for a transmission fluid These and other variations of the water side system 200 are provided in accordance with the present disclosure. This is within the scope of the teachings of the patent.

[0025] Each of the subplants 202-212 is configured to facilitate the function of the subplant. For example, the heater subplant 202 may include a heater in the hot water loop 214. A number of heating elements 220 (e.g., boilers, electric heaters) configured to add heat to the hot water via The heater subplant 202 is also shown to include several Also shown are pumps 222, 224, some of which are circulating hot water in the hot water loop 214 and flowing through the individual heating elements 220; The chiller subplant 206 is configured to control the speed of the chilled water in the chilled water loop 216. 2. The chiller 232 is shown to include a number of chillers 232 configured to remove heat from chilled water by The chiller subplant 206 also includes several pumps 234, 236. As shown, several pumps 234, 236 circulate chilled water in the chilled water loop 216. The chiller 232 is configured to circulate the chilled water and control the flow rate of the chilled water through each chiller 232.

[0026] The heat recovery chiller subplant 204 transfers heat from the chilled water loop 216 to the hot water loop 214. 226 (e.g., a refrigeration circuit) configured to The heat recovery chiller subplant 204 also includes several pumps 228, 2 30, and several pumps 228, 230 are shown as including a heat recovery heat exchanger. 226, hot and / or cold water is circulated through the individual heat recovery heat exchangers 226. The cooling tower subplant 208 is configured to control the flow rate of the water flowing through the condenser water loop. a number of cooling towers 238 configured to remove heat from the condenser water in the boiler 218; The cooling tower subplant 208 also includes several pumps 240. Several pumps 240 are also shown in the condenser water loop 218. configured to circulate condenser water and control the flow rate of condenser water through each cooling tower 238. will be done.

[0027] The hot TES subplant 210 is configured to store hot water for later use. 2. The thermal TES subplant 21 is shown to include a thermal TES tank 242. 0 controls the flow rate of hot water to or from the heated TES tank 242. The cooling / heating TES may also include one or more pumps or valves configured to control the The plant 212 is a chilled TES tank configured to store chilled water for later use. The cold TES subplant 212 is also shown to include a cold TES 244. The chilled water flow rate to or from the chilled TES tank 244 is configured to The system may also include one or more pumps or valves.

[0028] In some embodiments, the pumps of the water side system 200 (e.g., pump 222, 224, 228, 230, 234, 236 and / or 240) or water supply system One or more of the pipelines in the system 200 may be connected to the pump or pipeline. The shutoff valve controls the flow of fluid through the water supply system 200. It can be integrated with the pump or placed upstream or downstream of the pump to control In various embodiments, the waterside system 200 may include a specific More based on the configuration and type of load served by the water side system 200 may include many, fewer, or different types of devices and / or subplants. .

[0029] Air Supply System Referring now to FIG. 3, a block diagram of an airside system 300 according to some embodiments. In various embodiments, the air-side system 300 is an HVAC system. 100, or may supplement or replace the air-side system 130 in the HVAC system When installed within the HVAC system 100, the air supply side System 300 may be a subset of HVAC devices within HVAC system 100 (e.g., AHU106, VAV unit 106, ducts 112-114, fans, dampers, etc.) The air-side system 300 may include a water supply. By using the heated or cooled fluid supplied by the side system 200, It may operate to heat or cool the airflow supplied to the item 10 .

[0030] In FIG. 3, the air-side system 300 includes an economizer-type air handling unit (A The economizer type AHU is shown containing a 302. Vary the amount of outside and return air used by the air handling unit for cooling. For example, AHU 302 receives return air 304 from building zone 306 via return air duct 308. The supply air 310 can be delivered to the building zone 306 via a supply air duct 312 . In some embodiments, AHU 302 is a rooftop unit located on the roof of building 10. In some cases (e.g., AHU 106 as shown in Figure 1), both return air 304 and outdoor air 314 are used. The AHU 302 can be arranged in other ways to receive The exhaust damper 31 controls the amount of outside air 314 and return air 304 that form the supply air 310. 6. The mixing damper 318 and the outside air damper 320 may be configured to operate. Any return air 304 that does not pass through the mixing damper 318 is returned to the exhaust damper 311 as exhaust air 322. 6, which can be exhausted from AHU 302.

[0031] Each of the dampers 316 to 320 can be operated by an actuator. For example, the exhaust damper 316 can be operated by an actuator 324, and the mixing damper The damper 318 can be operated by an actuator 326, and the outside air damper 320 can be operated by an actuator 326. It can be operated by the actuator 328. The actuators 324 to 328 are , can communicate with the AHU controller 330 via a communication link 332. The computers 324 to 328 receive control signals from the AHU controller 330. A feedback signal can be provided to the controller 330. The feedback signal can be , e.g., an indication of the current actuator or damper position, given by the actuator the amount of torque or force obtained, diagnostic information (e.g., the amount of torque or force obtained by the actuators 324-328), results of diagnostic tests performed by the and / or the data collected, stored or used by the actuators 324-328. The AHU controller 330 may include other types of information or data that can be used to is implemented using one or more control algorithms (e.g., state-based algorithms, extremum-seeking control ESC control algorithm, proportional integral (PI) control algorithm, proportional integral derivative ( PID (Proportional Integral Derivative) control algorithms, Model Predictive Control (MPC) algorithms, Feedback control The actuators 324 to 328 are configured to be controlled using a control algorithm, etc. The economizer controller may be a

[0032] Still referring to FIG. 3, the AHU 302 includes a cooling unit located within the supply air duct 312. The cooling coil 334, the heating coil 336 and the fan 338 are shown. The fan 338 forces the supply air 310 through the cooling coil 334 and / or the heating coil 336. AH may be configured to pass through the building zone 306 and provide supply air 310 to the building zone 306. The U-controller 330 communicates with the air supply 310 via a communication link 340 to control the flow rate of the air supply 310. In some embodiments, the AHU controller 330 controls the heating or cooling applied to the supply air 310 by modulating the speed of the fan 338. or control the amount of cooling.

[0033] The cooling coil 334 receives water from the water supply system 200 (e.g., chilled water) via piping 342. 216) and delivers the cooling fluid to the water-side system 200 via piping 344. The valve 346 allows the cooling fluid to flow back through the cooling coil 334. It can be placed along piping 342 or piping 344 to control the flow rate. In some embodiments, the cooling coil 334 modulates the amount of cooling applied to the supply air 310. can be independently activated and deactivated to 0, by the BMS controller 366, etc.) includes multiple stages of cooling coils.

[0034] The heating coil 336 is connected to the water supply system 200 (e.g., hot water) via piping 348. 214) and delivers the heated fluid to the water supply system 200 via piping 350. The valve 352 allows the return of the heated fluid through the heating coil 336. It can be placed along the piping 348 or piping 350 to control the flow rate. In some embodiments, the heating coil 336 modulates the amount of heating applied to the air supply 310. can be independently activated and deactivated to 0, by the BMS controller 366, etc., includes multiple stages of heating coils.

[0035] Each of the valves 346 and 352 can be controlled by an actuator. For example, valve 346 can be controlled by actuator 354, and valve 3 52 can be controlled by actuator 356. Actuators 354 to 356 communicates with the AHU controller 330 via communication links 358-360. The actuators 354 to 356 receive control signals from the AHU controller 330. and provide a feedback signal to the controller 330. In this embodiment, the AHU controller 330 controls the supply air duct 312 (e.g., cooling coils) 334 and / or downstream of the heating coil 336) from a temperature sensor 362 The AHU controller 330 receives the temperature readings. The building zone 306 may also receive temperature measurements from the associated temperature sensors 364 .

[0036] In some embodiments, the AHU controller 330 controls the heating provided to the supply air 310. or to modulate the amount of cooling (e.g., to achieve a setpoint temperature of the supply air 310). or to maintain the temperature of the supply air 310 within a setpoint temperature range), The valves 346 and 352 are operated via the controllers 354 to 356. The location is determined by the heating provided to the supply air 310 by the cooling coil 334 or the heating coil 336. or the amount of cooling and energy consumed to achieve the desired supply air temperature. The AHU 330 activates or deactivates the coils 334 to 336. by adjusting the speed of the fan 338, or by a combination of both. , the temperature of the supply air 310 and / or the building zone 306 can be controlled.

[0037] Still referring to FIG. 3, the airside system 300 includes a building management system (BM) S) is shown to include a controller 366 and a client device 368. The BMS controller 366 controls the air supply system 300, the water supply system 200, and the H for the VAC system 100 and / or other controllable systems supplying the building 10 System level controllers, application or data servers, head nodes, One or more computer systems (e.g., BMS controllers may include: Controller 366 can communicate with similar or different protocols (e.g., LON, BACnet, etc.) ) via communication link 370 to multiple downstream building systems or subsystems ( For example, HVAC systems 100, security systems, lighting systems, water supply systems In various embodiments, the AHU controller 3 30 and BMS controller 366 can be separate (as shown in FIG. 3) or integrated. In an integrated implementation, the AHU controller 330 may be integrated with the BMS controller. a software module configured to be executed by a processor of the controller 366; It can be.

[0038] In some embodiments, the AHU controller 330 receives the data from the BMS controller 366. It receives information (e.g., commands, setpoints, operating boundaries, etc.) from the BMS controller. information (e.g., temperature readings, valve or actuator positions, operating status) to the controller 366. For example, the AHU controller 330 provides the temperature sensor 362 ~364 temperature readings, equipment on / off status, equipment operational capabilities, and / or building BMS control is used to monitor and control variable states or conditions within Zone 306. Any other information that can be used by the BMS controller 366 can be provided.

[0039] The client device 368 may communicate with the HVAC system 100, its subsystems and / or or one or more hulls to control, view, or otherwise interact with the device. - Man-machine interface or client interface (e.g., graphical User interfaces, reporting interfaces, text-based computer interfaces Faces, customer-facing web services, serving pages to web clients The client device 368 may include a computer work station, client terminal, remote or local interface, or other The client device 36 may be any type of user interface device. The client device 36 may be a fixed terminal or a mobile device. 8, a desktop computer, a computer server having a user interface; Laptop computer, tablet, smartphone, PDA, or any other The client device 368 can be any type of mobile or non-mobile device. The BMS controller 366 and / or the AHU controller 372 communicate with the The controller 320 can communicate with the controller 330.

[0040] Building Management System Referring now to FIG. 4, in accordance with some embodiments, a building management system (BMS) 4 A block diagram of the BMS 400 is shown. The BMS 400 automatically monitors various building functions. The BMS 400 can be implemented in the building 10 to provide and control is shown to include a MS controller 366 and a number of building subsystems 428 The building subsystem 428 includes the building electrical subsystem 434, the information and communication technology (ICT) subsystem a service subsystem 436, a security subsystem 438, an HVAC subsystem 440, and a lighting subsystem 441. Lighting Subsystem 442, Elevator / Escalator Subsystem 432 and Fire Safety Subsystem In various embodiments, the building subsystem 430 is shown to include 28 may include fewer, additional, or alternative subsystems. For example, building subsystems. The system 428 may include a refrigeration subsystem, an advertising or signage subsystem, a cooking subsystem, , vending subsystem, printer or copy service subsystem, or building 10. Other uses of controllable devices and / or sensors to monitor and control may also include or alternatively include any type of building subsystem of In some embodiments, the building subsystem 428 may be any of the components described with reference to FIGS. The water-side system 200 and / or the air-side system 300 may be .

[0041] Each of the building subsystems 428 has the necessary components to complete its individual functions and control activities. It may also include any number of devices, controllers, and connections. 1-3, and the same components as HVAC system 100. For example, the HVAC subsystem 440 may include a chiller, a boiler, or any air handling units, economizers, field controllers, supervisory controllers, and air conditioners. actuators, temperature sensors, and temperature, humidity, airflow, or other variable conditions within the building 10. The lighting subsystem 442 may include any number of lighting devices. Fixtures, ballasts, light sensors, dimmers, or other devices that can control the amount of light provided to a building space The security subsystem 438 may include other devices configured to coordinate with the The system includes occupancy sensors, video surveillance cameras, digital video recorders, video processing servers, and intrusion detection devices. devices, access control devices and servers or other security-related devices. obtain.

[0042] Still referring to FIG. 4, the BMS controller 366 is connected to a communication interface 40 7 and BMS interface 409. user control, monitor, and / or The BMS controller 366 and external applications are connected to enable synchronization and coordination. applications (e.g., monitoring and reporting applications422, enterprise control applications, Applications 426, Remote Systems and Applications 444, Client Devices 4 48) and facilitate communication between the , the interface 407 is connected to the BMS controller 366 and the client device 448. The BMS interface 409 can also facilitate communication between the BMS controller and the Controller 366 and building subsystems 428 (e.g., HVAC, lighting, security, elevators) It can facilitate communication between the power grid (power grid, distribution grid, businesses, etc.).

[0043] The interfaces 407, 409 are connected to the building subsystem 428 or other external systems. Or a wired or wireless communication interface for performing data communication with the device ( (e.g., jack, antenna, transmitter, receiver, transceiver, wire terminal, etc.) In various embodiments, the interface 407, Communication via 409 may be direct (e.g., local wired or wireless communication); It may also be via a communications network 446 (e.g., a WAN, the Internet, a cellular network, etc.). For example, the interfaces 407 and 409 can be Ethernet ( for transmitting and receiving data over a (registered trademark) based communications link or network In another example, the interface 407 may include an Ethernet card and port for 409 includes a Wi-Fi transceiver for communicating over a wireless communication network In another example, one or both of the interfaces 407, 409 may be a cellular phone or In one embodiment, communication interface 407 may include a mobile phone or cellular phone communication transceiver. is a power line communication interface, and the BMS interface 409 is an Ethernet interface. In another embodiment, the communication interface 407 and the BMS interface Both ports 409 can be separate Ethernet interfaces or can be on the same Ethernet It is a network interface.

[0044] Still referring to FIG. 4, the BMS controller 366 may include a processing circuit 404. 4, the processing circuit 404 includes a processor 406 and a memory 408. The processing circuitry 404 is configured to allow the processing circuitry 404 and its various components to interface 40 7, 409 to transmit and receive data via the BMS interface 40 9 and / or communication interface 407. The processor 406 may be a general-purpose processor, an application-specific integrated circuit (ASIC), or one or more Field Programmable Gate Arrays (FPGAs), a group of processing components , or other suitable electronic processing components.

[0045] The memory 408 (e.g., memory, memory unit, storage device, etc.) may be any of the memory devices described in this application. Data and information to complete and / or facilitate the various processes, layers and modules involved and / or one or more devices for storing computer code (e.g., memory (RAM, ROM, flash memory, hard disk storage, etc.). 408 can be or include volatile or non-volatile memory. The memory 408 stores database components, object code components, scripts, and information structures and components to support various activities, or any other type of and information structures described in this application. According to some embodiments, memory 408 may be communicatively connected to the processor 406 via the processing circuitry 404, performing one or more processes described herein (e.g., processing circuitry 404 and / or or by the processor 406).

[0046] In some embodiments, the BMS controller 366 may be implemented as a single computer (e.g., , one server, one housing, etc.). In various other embodiments, B The MS controller 366 may be configured to operate on multiple servers or computers (e.g., in distributed locations). Furthermore, Figure 4 shows the BMS components 366. However, in some embodiments, the applications 422, 426 may be The data may be hosted within the controller 366 (e.g., in memory 408).

[0047] Still referring to FIG. 4, the memory 408 includes an enterprise integration layer 410, an automated measurement and Qualification (AM&V) layer 412, Demand Response (DR) layer 414, Fault Detection and Diagnostics (FDD) layer 416, an integrated control layer 418, and a building subsystem integration layer 420. Layers 410-420 receive input from building subsystem 428 and other data sources. determining an optimal control action for the building subsystem 428 based on the input; and and providing the generated control signal to the building subsystem 428. The following paragraphs describe each of the layers 410-420 of the BMS 400. This section describes some of the common functions performed by

[0048] The enterprise integration layer 410 integrates information to support various enterprise-level applications. and configure it to serve services to clients or local applications. For example, the enterprise control application 426 may have a graphical user interface. interface (GUI) or any number of enterprise-level business applications (e.g., to provide subsystem-wide control over systems (e.g., accounting systems, user identification systems, etc.). The enterprise control application 426 can be configured to 6. It can also be configured as a configuration GUI to configure the In yet another embodiment, the enterprise control application 4 26 is received on interface 407 and / or BMS interface 409 Optimize building performance (e.g., efficiency, energy use, comfort, or safety) based on inputs In order to achieve this, layers 410-420 can be linked together.

[0049] The building subsystem integration layer 420 connects the BMS controller 366 and the building subsystems 42 8. For example, building subsystem integration Layer 420 receives sensor data and input signals from building subsystem 428 and The building subsystem 428 may provide output data and control signals. The system integration layer 420 can be configured to manage communications between building subsystems 428. The building subsystem integration layer 420 can also support multiple multi-vendor / multi-protocol Translating communications (e.g., sensor data, input signals, output signals, etc.) across systems .

[0050] The demand response layer 414 responds to meeting demands of the building 10 by adjusting resource usage (e.g., electricity usage, natural gas usage, water usage, etc.) and / or the monetary cost of such resource usage The optimization can be configured to optimize usage based on time-of-use pricing, curtailment signals, and energy Energy availability, utility providers, distributed energy generation systems 424, Energy Storage 427 (e.g., Heat TES 242, Cold TES 244, etc.) ) or other data received from other sources. Other layers of the MS controller 366 (e.g., building subsystem integration layer 420, integrated control layer 418, etc. Inputs received from other layers include temperature, Carbon dioxide level, relative humidity level, air quality sensor output, occupancy sensor output, room schedule Inputs may include environmental or sensor inputs such as temperature, humidity, and the like. (e.g., expressed in kWh), heat load measurements, price information, forecast prices, levelized prices, utility rates It may also include inputs such as curtailment signals from the business and the like.

[0051] According to some embodiments, the demand response layer 414 responds to received data and signals. These responses are communicated to the control algorithms of the integrated control layer 418. change the control strategy, change the setpoint, or This may include activating / deactivating building equipment or subsystems under the demand response layer. 414 is control logic configured to determine when to utilize the stored energy. For example, the demand response layer 414 may also include a demand response layer 414 that may be configured to automatically A decision can be made to start using energy from storage 427.

[0052] In some embodiments, the demand response layer 414 responds to demand (e.g., prices, reduction signals, demand Energy demand based on one or more inputs representing Actively initiate cost-minimizing control actions (e.g., automatically changing setpoints) In some embodiments, the demand response layer 4 14 uses an equipment model to determine an optimal set of control actions. For example, the inputs, outputs and / or functions performed by various sets of building equipment The equipment model may include a thermodynamic model that describes the building equipment collection (e.g., radiators, refrigerator arrays, etc.) or individual devices (e.g., individual refrigerators, heaters, poles, etc.) It can represent a number of different things (e.g., a slur).

[0053] The demand response layer 414 stores one or more demand response policy definitions (e.g., databases, It can also contain or utilize additional policies (e.g., XML files). The definition is tailored to the user's application, desired comfort level, and specific building equipment. can adjust the control actions initiated in response to demand inputs based on other concerns. Edited or adjusted by a user (e.g., via a graphical user interface) For example, a demand response policy definition can determine which equipment is to be deployed in response to a particular demand input. When to turn a system or piece of equipment on or off, and for how long What setpoints can be changed and the range of allowable setpoint adjustments How long will the high demand set point last before returning to the normal scheduled set point? How long you hold the point, how close you are to your capacity limits, and what equipment mode or energy storage devices (e.g., thermal storage tanks, battery banks) Energy transfer charges to and from energy storage devices (e.g., maximum charges , alarm charges, other charge limit information, etc.), when to send out on-site generated energy (e.g. For example, fuel cells, motor-generator sets, etc. may be specified.

[0054] The integrated control layer 418 communicates with the building subsystem integration layer 420 and / or Alternatively, the data input or output of the demand response layer 414 may be configured to be used. The subsystem integration provided by the building subsystem integration layer 420 allows for the integrated control layer 418 organizes the subsystems 428 so that they behave as a single, integrated supersystem. In some embodiments, the control activities of the integrated control layer 41 can be integrated. 8 provides better comfort and energy savings than the separate subsystems can provide alone. from multiple building subsystems to provide greater comfort and energy savings For example, the integrated control layer 418 includes control logic that uses the inputs and outputs of the second subsystem. input from the first subsystem to make energy-saving control decisions for the system. The results of these decisions can be used in the building subsystem integration layer. You can send it back to 420.

[0055] The integrated control layer 418 is shown to be logically below the demand response layer 414 . The integrated control layer 418 cooperates with the demand response layer 414 to control the building subsystems 428 and their By enabling control of each control loop of the demand response layer 414, This configuration can advantageously provide enhanced For example, the integrated control layer 418 may Demand response driven upward adjustment of temperature setpoint (or any other function that directly or indirectly affects temperature) Another component that contributes to the overall building energy use is that of the refrigerator. Fan energy (or other energy used to cool a space) that would be lost The method can be configured to ensure that it does not result in an increase in the

[0056] The integrated control layer 418 controls the constraints (e.g., temperature, lighting levels) even while demand load shedding is in progress. The demand response layer 414 checks that the demand response system (e.g., bell) is properly maintained. Constraints can be configured to provide feedback to layer 414. Constraints can also be configured to provide feedback to layer 414. Safety, equipment operating limits and performance, comfort, fire codes, electrical codes, energy codes and It may also include set points or sensing boundaries associated with the same. 8 logically comprises a defect detection and diagnostics layer 416 and an automated measurement and verification layer 412. The integrated control layer 418 controls the building based on the outputs from multiple building subsystems. can be configured to provide calculated inputs (e.g., aggregations) to higher levels of can.

[0057] The automated measurement and verification (AM&V) layer 412 is connected to the integrated control layer 418 or the demand response layer 4 14 to verify that the control strategies dictated by the (e.g., AM&V layer 412, integrated control layer 418, building subsystem integration layer 4 20, using data aggregated by the FDD layer 416 or otherwise). AM& The calculations performed by the V layer 412 are based on the building system energy model and / or individual It can be based on an equipment model for a BMS device or subsystem. For example, The &V layer 412 passes the model prediction output to the building subsystem 4 to determine the accuracy of the model. 28.

[0058] The Fault Detection and Diagnostics (FDD) layer 416 includes a building subsystem 428, system devices (i.e., building equipment), as well as the demand response layer 414 and the integrated control layer 41 Configured to provide ongoing fault detection for the control algorithms used by the The FDD layer 416 can receive information from the integrated control layer 418 about one or more buildings. Receives data input directly from a subsystem or device or from another data source The FDD layer 416 can automatically diagnose and respond to detected defects. Responses to detected or diagnosed defects can be made by users, maintenance scheduling systems, or or a control algorithm configured to attempt to repair or address the defect. The method may include providing a warning message to the rhythm.

[0059] The FDD layer 416 provides detailed subsystem inputs available in the Building Subsystem Integration layer 420. Use force to identify the specific identity of the defective component or the cause of the defect (e.g., a loose In another exemplary embodiment, the present invention can be configured to output a damper linkage. The FDD layer 416 is configured to provide "fault" events to the integrated control layer 418. The control layer 418 executes control strategies and policies in response to received fault events. According to some embodiments, the FDD layer 416 (or integrated control engine or business rule) The policy (executed by the engine) is designed to maximize the lifespan of the device to reduce energy waste. To prolong life or ensure correct control response, It can shut down surrounding systems or directly control activity.

[0060] The FDD layer 416 is a data store for various different system data (or data stores for live data). FDD Layer 4 can be configured to store or access the data in a 16 uses some content of the data store to determine the equipment level (e.g., a specific cooling It can identify defects in other components (freezer, specific AHU, specific terminal unit, etc.) The content can be used to identify defects at the component or subsystem level. For example, building subsystem 428 may be configured to manage BMS 400 and its various components. It is possible to generate temporal (i.e., time series) data that indicates the performance of the building. The data generated by the system 428 may be measured or calculated, exhibiting statistical properties. The data may include previously acquired values ​​and may be used to determine the corresponding system or process (e.g., temperature control process, flow - how a control process is performing in terms of error from its setpoint These processes can provide information about how well the system is performing. It exposes when defects begin to deteriorate and encourages users to repair them before the deterioration becomes more severe. This can be checked by the FDD layer 416 to alert the

[0061] Referring now to FIG. 5, another building management system (BMS) according to some embodiments. 1 shows a block diagram of a BMS 500. The BMS 500 controls the HVAC system 100, the water Devices of supply side system 200, air side system 300, building subsystem 428 , as well as other types of BMS devices (e.g., lighting equipment, security equipment, etc.) and / or can be used to monitor and control HVAC equipment.

[0062] BMS 500 is a system architecture that facilitates automatic device discovery and device model distribution. Device discovery is performed across multiple different communication buses (e.g., system bus 554, zone buses 556-560, 564, sensor / actuator bus 566, etc. and across multiple different communication protocols and at multiple levels of the BMS 500. In some embodiments, device discovery involves tracking the status of devices connected to each communication bus. This is accomplished using an active node table that provides node information for each communication bus. by monitoring the corresponding active node table for new nodes. , it can monitor for new devices. When a new device is detected, BMS 500 initiates interaction with new devices without user interaction It can be used to communicate with other devices (e.g., send control signals, use data from devices).

[0063] Some devices in the BMS 500 use equipment models to identify themselves to the network. The equipment model is a device object that is used for integration with other systems. Object attributes, view definitions, schedules, trends and related BACnet value objects ( For example, analog values, binary values, multi-state values, etc.) Devices store their own equipment model. Other devices in the BMS 500 store their own equipment model. The device model may be stored in a separate location (e.g., in another device). The generator 508 can store an equipment model of the bypass damper 528. In some embodiments, the zone coordinator 508 may control the bypass damper 528 or the zone Automatically creates device models for other devices on the bus 558. The network operator can also create equipment models for devices connected to those zone buses. The equipment model of a device is the data points exposed by the device on the zone bus. Automatically based on device type, device type and / or other device attributes Some examples of automatic device discovery and device model distribution are described in more detail below. Discuss in.

[0064] Still referring to FIG. 5, the BMS 500 includes a system manager 502, several Some Zone Coordinators 506, 508, 510, 518 and some Zone Controllers The device is shown to include trolleys 524, 530, 532, 536, 548, and 550. The system manager 502 monitors data points within the BMS 500 and calculates the monitored changes. The numbers may be reported to various monitoring and / or control applications. 502 includes a data communication link 574 (e.g., BACnet IP, Ethernet, wired or wirelessly) to a client device 504 (e.g., a user device, device computers, laptops, mobile devices, etc.) The system manager 502 receives the client device data via a data communication link 574. A user interface can be provided to the device 504. The user interface can include: A user can monitor and / or control the BMS 500 via a client device 504. To enable control.

[0065] In some embodiments, the system manager 502 communicates with the system bus 554 It is connected to the zone coordinators 506 to 510 and 518. System Manager 502 Master-Slave Token Passing (MSTP) protocol or any other communication protocol The zone coordinators 506 to 510 communicate with each other via the system bus 554 using the protocol. , 518. The system bus 554 may also be configured to communicate with the System Manager 502 is connected to Constant Volume (CV) Rooftop Unit (RTU) 512, Input / Output Module an IOM (Integrated Operation Module) 514, a thermostat controller 516 (e.g., a TEC5000 series Leeds Thermostat Controller) and Network Automation Engine (NAE) or third party controller 520. The RTU 512 can be configured to communicate directly with the system manager 502. Other RTUs can be connected directly to the system bus 554 via intermediate devices. For example, the wired input 562 may communicate with the system manager 502. The three-party RTU 542 can be connected to the thermostat controller 516, The stat controller 516 is connected to the system bus 554 .

[0066] The system manager 502 manages user input for any device, including the device model. Zone coordinators 506-510, 518 and and thermostat controller 516 communicate with the system via system bus 554. The system manager 502 can then provide these device models. In this embodiment, the system manager 502 may be configured to manage connected devices (e.g., I Automatically create equipment models for the OM 514, third-party controller 520, etc. For example, the system manager 502 may request the device model of any device that responds to the device tree request. The system manager 502 can also create equipment models. can be stored in the system manager 502. 502 uses the equipment models created by the system manager 502 to It is possible to provide a user interface for devices that do not include their own device model. In some embodiments, the system manager 502 communicates with the system bus 554. and stores view definitions for each type of connected equipment, and then uses the stored view definitions A user interface for the device is generated.

[0067] Each of the zone coordinators 506 to 510 and 518 is connected to a zone bus 556, 558, 56 Zone controllers 524, 530-532, 536, 548-55 through 0, 564 0. The zone coordinators 506 to 51 0,518 is used to zone-balance using MSTP protocol or any other communication protocol. Zone controllers 524, 530 to 532, and 536 are connected via switches 556 to 560 and 564. , 548 to 550. In addition, the zone buses 556 to 560 and 564 , Zone Coordinators 506 to 510, 518, Variable Air Volume (VAV) RTU 52 2, 540, Changeover Bypass (COBP) RTU 526, 552, Bypass Damper 528 Connects with other types of devices such as 534, 544, 546 and PEAK controllers You can also do this.

[0068] Zone Coordinators 506-510 and 518 are responsible for the various zoning systems. In some embodiments, each zone can be configured to monitor and instruct. The zoning coordinators 506-510 and 518 are monitors for the separate zoning systems. and commands, and is connected to the zoning system via a separate zone bus. The zone coordinator 506 communicates with the VAV RTUs 522 and 523 via the zone bus 556. and a zone controller 524. The zone coordinator 508 , COBP RTU 526, bypass damper 528, COB via zone bus 558 It can be connected to the P zone controller 530 and the VAV zone controller 532. The zone coordinator 510 can communicate with the PEAK controller via the zone bus 560. 534 and VAV zone controller 536. The PEAK controller 544, bypass driver 518 are connected to the zone bus 564. Zone Controller 546, COBP Zone Controller 548 and VAV Zone Controller 550 It can be connected.

[0069] The single model of Zone Coordinator 506-510 and 518 can accommodate multiple different types Zoning systems (e.g., VAV zoning system, COBP zoning system) Each zoning system can be configured to handle RTUs, The device may include one or more zone controllers and / or bypass dampers, for example. , the zone coordinators 506 and 510 respectively to the VAV RTUs 522 and 540. Shown as a connected Verasys VAV engine (VVE). The deactivator 506 is directly connected to the VAV RTU 522 via the zone bus 556. whereas the zone coordinator 510 is provided to the PEAK controller 534. connected to a third party VAV RTU 540 via a wired input 568. The networkers 508 and 518 are connected to the COBP RTUs 526 and 552, respectively. rasys COBP Engine (VCE) Zone Coordinator 5 08 is directly connected to COBP RTU 526 via zone bus 558 The zone coordinator 518 receives the wired input provided to the PEAK controller 544. 570 to a third party COBP RTU 552.

[0070] Zone controllers 524, 530 to 532, 536, 548 to 550 are sensors / Individual BMS devices (e.g., sensors, actuators) can be connected via the SA bus. For example, the VAV zone controller 536 may communicate with the SA It is shown connected to networked sensors 538 via bus 566. The network controller 536 uses the MSTP protocol or any other communication protocol. 5, one SA bus 538 can be used to communicate with the network-connected sensor 538. Although only 66 is shown, each of the zone controllers 524, 530 to 532, 536, It should be understood that 48-550 can be connected to different SA buses. Each SA bus is The controller can be connected to various sensors (e.g., temperature sensor, humidity sensor, pressure sensor, light sensors, occupancy sensors, etc.), actuators (e.g., damper actuators, valve actuators, actuators, etc.) and / or other types of controllable equipment (e.g., refrigerators, heating It can be connected to a variety of devices (machines, fans, pumps, etc.).

[0071] Each zone controller 524, 530-532, 536, 548-550 is The zone controller 5 can be configured to monitor and control the product zone. 24, 530-532, 536, 548-550 are provided via their SA buses. The inputs and outputs can be used to monitor and control various building zones. For example, the zone controller 536 may use feedback in the temperature control algorithm. temperature input received from networked sensor 538 via SA bus 566 as (e.g., measured temperatures of building zones) can be used. Zone controller 524 , 530-532, 536, 548-550 are various types of control algorithms (e.g. For example, state-based algorithms, extremum-seeking control (ESC) algorithms, proportional-integral (PI) ) control algorithm, proportional-integral-derivative (PID) control algorithm, model predictive control ( MPC (Master Process Control) algorithms, feedback control algorithms, etc. are used to or variable states or conditions around the building 10 (e.g., temperature, humidity, airflow, lighting, etc.) It can be controlled.

[0072] Variable Refrigerant Flow System 6-7, a variable refrigerant flow (VRF) system according to some embodiments is shown. 2100 is shown. The VRF system 2100 includes one or more outdoor VRF units 2 102 and a plurality of indoor VRF units 2104. The outlet 2102 may be located outside the building and may operate to heat or cool a refrigerant. The VRF unit 2102 may be configured to generate liquid, vapor, and / or superheated vapor phases. Electricity may be consumed to convert refrigerant between different phases (VRF). The units 2104 may be distributed throughout various building zones within a building to heat or cool the Each indoor VRF unit 210 4 may provide temperature control for the particular building zone in which the indoor VRF unit 2104 is located. The term "indoor" refers to the fact that the indoor VRF unit 2104 is typically located inside a building. is used to represent, but in some cases, one or more indoor VRF units For example, to heat / cool patios, entrance halls, walkways, etc. (external to)

[0073] One advantage of the VRF system 2100 is that other indoor VRF units 2104 are in heating mode. Some indoor VRF units 2104 may operate in cooling mode while others operate in heating mode. For example, outdoor VRF unit 2102 and indoor VRF unit 210 Each of the four building zones can operate in heating mode, cooling mode, or off mode. may be independently controlled and may have different temperature set points. In some embodiments, each building , up to three outdoor VRF units 2102 located outside the building (for example, on the roof), and Up to 128 indoor VRF units distributed throughout the building (e.g., within different building zones) The building zone may include, among other possibilities, apartment units, office In some cases, various building zones may include commercial spaces, retail spaces, and common areas. The properties are owned, leased, or otherwise occupied by a variety of tenants. All services are provided by the VRF system 2100.

[0074] Many different configurations exist for the VRF system 2100. In some embodiments, The VRF system 2100 includes a single outdoor VRF unit 2102 with a single refrigerant return line and a single In a double piping system, the refrigerant outlet line is connected to the outdoor VR. All of the F units 2102 only produce either heated or cooled refrigerants. can be supplied through a single refrigerant outlet line and therefore can operate in the same mode. In the VRF system 2100, each outdoor VRF unit 2102 has a refrigerant return line, a temperature It is a triple piping system that connects to the hot refrigerant outlet line and the cold refrigerant outlet line. In piping systems, both heating and cooling are provided simultaneously via dual refrigerant outlet lines An example of a triple-pipe VRF system is described in detail with reference to FIG.

[0075] Referring now to FIG. 8, a block diagram illustrating a VRF system 2200 according to some embodiments is shown. A block diagram is shown. The VRF system 2200 includes an outdoor VRF unit 202, several 2206, and several indoor VRF units 2204. The outdoor VRF unit 202 includes a compressor 2208, a fan 2210, or Other power consumers configured to convert refrigerant between the liquid phase, the vapor phase, and / or the heated vapor phase. Indoor VRF units 2204 may include cooling components throughout the building. and may receive heated or cooled refrigerant from the outdoor VRF unit 202. Each indoor VRF unit 2204 is connected to the specific building zone in which the indoor VRF unit 2204 is located. The heat recovery unit 2206 can provide temperature control for the outdoor VRF unit 202. The flow of refrigerant between the indoor VRF unit 2204 and the ) to minimize the heating or cooling load served by the outdoor VRF unit 202. It can become.

[0076] The outdoor VRF unit 202 includes a compressor 2208 and a heat exchanger 2212. The compressor 2208 transfers the refrigerant to the heat exchanger 2212 and the indoor VRF unit 22 The compressor 2208 is controlled by the outdoor unit control circuit 214. At high frequencies, the compressor 2208 has a greater heat transfer capacity. is given to the indoor VRF unit 2204. The power consumption of the compressor 2208 is It increases in proportion to the frequency.

[0077] Heat exchanger 2212 acts as a condenser ( The VRF system 2200 may be used as a heating element (allowing the refrigerant to reject heat to the outside air) or as a When operating in the evaporator mode, it functions as an evaporator (allowing the refrigerant to absorb heat from the outside air). The fan 2210 provides airflow through the heat exchanger 2212. The velocity is adjusted to modulate the rate of heat transfer to or from the refrigerant in the heat exchanger 2212. This can be controlled (for example, by the outdoor unit control circuit 214).

[0078] Each indoor VRF unit 2204 includes a heat exchanger 2216 and an expansion valve 2218. Each of the heat exchangers 2216 is shown as a separate unit when the indoor VRF unit 2204 is in heating mode. When operating in a mode, the condenser (which allows the refrigerant to reject heat to the atmosphere in the room or zone) or as an evaporator when the indoor VRF unit 2204 operates in cooling mode. (allowing the refrigerant to absorb heat from the atmosphere within the room or zone). The fan 2220 provides airflow through the heat exchanger 2216. The speed of the fan 2220 is can be adjusted (e.g., For example, by the indoor unit control circuit 2222).

[0079] In Figure 8, the indoor VRF unit 2204 is shown operating in cooling mode. In this mode, refrigerant is supplied to the indoor VRF unit 2204 via the cooling line 2224. The refrigerant is expanded by expansion valve 2218 to a cold, low-pressure state and is then pumped into rooms within the building. or through a heat exchanger 2216 (which functions as an evaporator) to absorb heat from the zone. The heated refrigerant then flows back through return line 2226 to the outdoor VRF unit 202. The refrigerant is then compressed into a hot, high-pressure state by the compressor 2208. The cooled refrigerant flows through the condenser 2212 (which functions as a condenser) and rejects heat to the outside air. Cooling line 2224 can be supplied to indoor VRF unit 2204. In cooling mode, , the flow control valve 2228 may be closed and the expansion valve 230 may be fully open.

[0080] In heating mode, the refrigerant is delivered to the indoor VRF unit 22 in a hot state via heating line 2232. 04. The hot refrigerant flows through heat exchanger 2216 (which functions as a condenser) The refrigerant then passes through the cooling line 2224 to reject the heat to the atmosphere within the room or zone of the building. The refrigerant flows back through the expansion valve to the outdoor VRF unit (opposite to the flow direction shown in Figure 8). The expanded refrigerant is then passed through heat exchanger 2212 ( The heated refrigerant flows through the compressor (which acts as an evaporator) and absorbs heat from the outside air. The compressed air is then sent to the indoor VRF unit via the heated line 2302 in a hot compressed state. In the heating mode, the flow control valve 2228 controls the flow rate of the compressor. The heater 2208 may be completely opened so that refrigerant from the heater 2208 can flow into the heating line 2302 .

[0081] As shown in FIG. 8, each indoor VRF unit 2204 includes an indoor unit control circuit 2222. The indoor unit control circuit 2222 responds to the building zone temperature setpoints or In response to other requests to provide heating / cooling to the zone, the fan 2220 and expansion valve 2 218. The fan control circuit 2222 may generate a signal to turn the fan 2220 on and off. The control circuit 2222 also controls the heat transfer capacity required by the indoor VRF unit 2204. The frequency of the compressor 2208 corresponding to the capacity of the indoor VRF unit 22 is determined. 04 must provide a certain amount of heating or cooling capacity. Once the control circuit 2222 is determined, the indoor unit control circuit 2222 determines the required capacity. a compressor frequency request including a corresponding compressor frequency, and This is transmitted to the control circuit 214.

[0082] The outdoor unit control circuit 214 receives signals from one or more indoor unit control circuits 2222. receive compressor frequency requests and then sum these compressor frequency requests, e.g. In some embodiments, the compressor The frequency has an upper limit so that the compressor overall frequency cannot exceed the upper limit. The outdoor unit control circuit 214 controls, for example, the DC inverter compressor motor of the compressor. The compressor provides the full frequency as the input frequency to the compressor. Therefore, the indoor unit control circuit 2222 and the outdoor unit control circuit 214 cooperate to: Modulates compressor frequency to match heating / cooling demand. Outdoor unit control circuit 214 also controls the valve positions of the flow control valve 2228 and the expansion valve 230, the refrigerant power setpoint, the refrigerant flow rate setpoint, and the refrigerant pressure setpoint (measured, for example, by pressure sensor 2306). pressure differential setpoint), on / off commands, staging commands, or Other signals that affect the operation of the compressor 2208, as well as the fan speed setpoint, fan power setpoint, airflow setpoint, on / off command, or fan 2210 operation. The control signal provided to the fan 2210 may include other signals.

[0083] The indoor unit control circuit 2222 and the outdoor unit control circuit 214 are connected to a control circuit 221 4, 2222, data of one or more control signals generated by or provided to For example, the indoor unit control circuit 2222 may store and / or provide the generated data history. Compressor demand frequency, fan on / off time and indoor VRF unit 22 04 may store and / or provide a log of on / off times. 4 is the compressor demand frequency and / or compressor total frequency and compressor A log of execution times may be stored and / or provided.

[0084] The VRF system 2200 transmits energy via an outdoor meter 2252 and an indoor meter 2254. 2250. According to an embodiment, the energy grid 2250 may be any source of electrical power (e.g., a utility). Electric grid maintained by a utility company and supplied with power by one or more power plants The outdoor meter 2252 is a meter for measuring the consumption of the outdoor VRF unit 202 over time. The indoor meter 2254 measures the power consumption, for example, in kilowatt-hours (kWh). The VRF system measures the power consumption of the unit 2204 over time, for example, in kWh. 2200 is a method for calculating the energy consumption charges charged by the utility company that supplies the electricity to the outdoor The power consumption is measured based on the power consumption measured by the indoor meter 2252 and / or the indoor meter 2254. The price (e.g., dollars per kWh) of electricity can fluctuate over time.

[0085] The VRF system 2200 also includes a system manager 502. See FIGS. As will be described in more detail below in light of the above, the system manager 502 also maintains occupant comfort. The VRF system 2200 is configured to minimize energy consumption costs while

[0086] Window air conditioner 9, a window air conditioner 2300 is shown in accordance with an exemplary embodiment. 2300 is configured to be mounted to a window of a building so as to span the exterior wall 2302 of the building. Therefore, the window air conditioner 2300 can be used to separate indoor (i.e., inside the building) and outdoor (i.e., outside the building) air. ) and / or receive air from them. 0 is sometimes referred to in the art as a room air conditioner.

[0087] The window air conditioner 2300 acts as a heat pump, transferring heat from the indoor air to the outdoor air. As shown in Figure 1, the window air conditioner 2300 takes in indoor air and outputs cooled air to the room. The window air conditioner 2300 also takes in outside air and outputs exhaust air to the outside of the building. 2300 traverses the compressor, condenser, evaporator, and outer wall 2302 (i.e. It may include one or more fans to facilitate the transfer of heat (from inside to outside). Thus, the window air conditioner 2300 is configured to reduce the temperature of the indoor air toward a temperature setpoint. can be.

[0088] The window air conditioner 2300 uses energy when operating to transfer heat across the exterior wall 2302. The window air conditioner 2300 consumes power from the energy grid 2250. The window air conditioner 2300 operates based on, for example, a temperature set point. Can it be controlled to operate at different power levels to provide different levels of cooling to the building? The window air conditioner 2300 may also be turned on and off as needed. The air conditioner 2300 uses more power to provide more cooling and less cooling. It consumes less power to provide.

[0089] The system manager 502 provides control signals for the window air conditioner 2300 and The window air conditioner 2300 is communicatively coupled to the window air conditioner 2300 to receive data from the window air conditioner 2300. For example, the system manager 502 may provide a temperature setpoint to the window air conditioner 2300. Manager 502 is described in more detail with reference to Figures 12-13. In some embodiments, In some embodiments, the system manager 502 is integrated into the window air conditioner 2300. The system manager 502 can remotely (e.g., cloud service) manage multiple window air conditioners 2300. (on the server) to operate and / or service.

[0090] Room air conditioning system Referring now to FIG. 10, a room air conditioning system 2400 is shown in accordance with an exemplary embodiment. The room air conditioning system 2400 provides cooling for the rooms in the building. The outdoor unit 2402 includes an indoor unit 2404. The unit 2404 is separated from the outdoor unit 2402 by the exterior wall 2302 of the building. , located outside the building, while the indoor unit 2404 is located inside the building. The indoor unit 2404 can be mounted on the indoor surface of the exterior wall 2302. The external unit 2402 is communicatively coupled to exchange control signals and data. The interior unit 2404 may also receive power via the exterior unit 2402, and vice versa. It's nice.

[0091] The outdoor unit 2402 receives power from the energy grid 2250 to cool the refrigerant. Next, the refrigerant is transported from the outdoor unit 2402 to the indoor unit 2404 through the outer wall 406. The refrigerant is pumped through a pipe 2408 that passes through the room. A fan 2410 transfers heat from the room to the refrigerant. Air is blown from the room over the pipe 2408 to transport the refrigerant. It flows back to the outdoor unit 2402, where it is re-cooled and circulated back to the indoor unit 2404. Therefore, the room air conditioning system 2400 distributes heat from the room to the outside across the exterior wall 2302. Then operate the transfer.

[0092] The outdoor unit 2402 and the indoor unit 2404 are configured to track the room temperature setpoint. For example, the outdoor unit 2402 may be controlled to vary the speed and / or volume of refrigerant flow. Controlled to operate at different power levels to provide different refrigerant temperatures to the indoor unit 2404 The fan 2410 can be controlled to operate at various speeds. 2400 can also be controlled to turn on and off as needed. System 2400 consumes more power while providing more cooling to the room. Consumed from grid 2250.

[0093] The system manager 502 provides control signals for the room air conditioner system 2400 and outdoor unit 2402 and / or Or, the system manager 50 is communicatively coupled to the indoor unit 2404. 2 can provide temperature set points to the room air conditioner system 2400. The system manager 502 12-13. In some embodiments, the system manager The heater 502 is integrated into the outdoor unit 2402 and / or the indoor unit 2404 . In some embodiments, the system manager 502 may include multiple room air conditioner systems 2400. are operated and / or serviced remotely (e.g., on a cloud server).

[0094] Packaged Air Conditioner Referring now to FIG. 11, a packaged air conditioner system 2500 according to an exemplary embodiment is shown. The packaged air conditioner system 2500 includes a packaged air conditioner 2504, an intake air conditioner 2506, an intake air conditioner 2508, an intake air conditioner 2509, an intake air conditioner 2510, an intake air conditioner 2511, an intake air conditioner 2512, an intake air conditioner 2513, an intake air conditioner 2514, an intake air conditioner 2515, an intake air The packaged air conditioner 2504 includes a room air vent 2506, and a cold air duct 2508. The intake vent 2506 and the cold air duct 2508 are located outside the package. To allow air to flow between the packaged air conditioner 2504 and the interior of the building, 2504 and penetrates the exterior wall 2302 of the building.

[0095] The packaged air conditioner system 2500 draws air from the interior of the building through intake vent 2506 into the room. It consumes power from the energy grid 2250 to draw in the air and cool it down. The packaged air removes heat from the indoor air and delivers cool air to the cool air duct 2508. The air conditioning system 2500 expels heat to the outside air. The cold air duct 2508 cools the building's indoor temperature. To lower the temperature, cool air is allowed to flow across the exterior wall 2302 and into the building's atmosphere. do.

[0096] The packaged air conditioner 2504 can be controlled to track the temperature set point of the building. For example, the packaged air conditioner 2504 may provide cold air at various temperatures and / or various flow rates. The packaged air conditioner 2502 can operate at various power levels to supply the cooling air to the cold air duct 2508. 504 may operate at a higher rate of power consumption and / or for more time. The energy grid 225 provides more cooling to the room by operating between 0 to consume more power.

[0097] The system manager 502 provides control signals for the room air conditioner system 2400 and to receive data from the packaged air conditioner 2504 For example, the system manager 502 may communicate with the package to set temperature set points. The system manager 502 can provide the information to the air conditioner 2504. In some embodiments, the system manager 502 may In some embodiments, the packaged air conditioner 2504 is integrated into the A plurality of room air conditioner systems 2400 can be remotely operated (e.g., on a cloud server) and / or Or serve.

[0098] System Manager with Neural Networks Referring now to FIG. 12, a block diagram of the system manager 502 according to an exemplary embodiment. FIG. 12 illustrates a system communicatively coupled to an instrument 600 and a sensor 602. 5 shows the system manager 502. According to various embodiments, the device 600 may be configured as shown in FIGS. Various HVAC equipment (e.g., HVAC system 100, water supply system 200, air supply 6-7, the VRF system 2100 of FIG. 8, system 2200, the window air conditioner 2300 of FIG. 9, the room air conditioning system 400 of FIG. 10, and / or or packaged air conditioner system 2500 of FIG. 11. The device 600 may include a plurality of window air conditioners 2300, room air conditioning systems 2400, and / or package For example, the equipment 600 may be a room, a building, or a campus operable to affect one or more variable states or conditions (such as the temperature inside the building) If the device 600 includes multiple devices, one or more of the multiple devices may be The device 600 can display its offline or online status in a system The system manager 502 may be configured to provide the

[0099] The sensors 602 take measurements that aid in the operation of the device 600 and the system manager 502. The sensors 602 include temperature sensors, humidity sensors, air velocity sensors, occupancy The sensor 602 may include a temperature sensor, such as a sensor for detecting the indoor temperature of the building, the outdoor temperature outside the building, the temperature sensor, etc. The sensor 602 may measure the humidity inside the building, the humidity outside the building, the number of people inside the building, etc. The sensor 602 also collects measurement data and provides it to the system manager 502. , the measurement data may be provided to the device 600.

[0100] The system manager 502 is shown to include a dispatch generation circuit 604. The patch generation circuit 604 transmits real-time measurements from the sensor 602 to the device 600. and generating economical dispatches of the equipment 600. As used herein, the term "measurement and status data" refers to a process for processing these measurements and status data. In addition, economic dispatch (control dispatch) is performed by controlling the control input (e.g., temperature setting) of the equipment 600. Fixed points, schedules, humidity set points, airflow set points, power levels, on / off settings, Any set of parameters (damper position, fan speed, compressor frequency, resource consumption allocation) Or refers to a group.

[0101] The dispatch generation circuit 604 distributes measurement and equipment status information and occupant comfort information. to link economic dispatch that minimizes utility prices while achieving A neural network uses a network of artificial neurons. The network is allowed to send signals to neighboring neurons. The artificial neuron receives the signal. The neuron then processes this signal and then transmits an output to various neighboring neurons. The output of a neuron can be calculated according to the sum of its inputs. The strength of the signal being received is influenced by a set of learned weights in the neural network. Neurons are connected and trained to convert input data into output signals. In some embodiments, the neural network The network is a convolutional neural network that uses layers and pooling to improve efficiency. As will be described in more detail below with reference to Figure 13, the learned weights are The training may be provided to the dispatch generation circuit 604 by an offline training system.

[0102] The neural network operated by the dispatch generation circuit 604 generates the measurements and is substantially more efficient than traditional methods of generating economic dispatch based on Therefore, the dispatch generation circuit 604 may , the system manager 502 controls, for example, the HVAC system 100, the water supply system 200, , air supply side system 300, VRF system 2100, VRF system 2200, window air air conditioner 2300, room air conditioning system, and / or packaged air conditioner system 2500; Both allow sophisticated dispatching in real-time online given available computational resources. This could enable deployment in locations that were previously too complex for network control.

[0103] In contrast to traditional computational programs in which deterministic rules are specified by the user, neural A network is typically a network that performs a desired association between input data and output signals. Training a neural network is the process of neural A set of learned weights that force the network to make the desired associations between inputs and outputs This involves assisting neural networks in learning neural networks. The network typically determines the input and output of the system that the neural network is trying to model. Use a dataset of real-world training data (e.g., measured data) However, heating and / or cooling may be performed in accordance with the techniques described herein. In the context of providing a comprehensive set of training data, it is often months or years of data. may require additional power, but it is important to consider all the conditions that the building may face (e.g. extreme weather events, rare Therefore, machine learning and neural networks Conventional approaches to networking are inadequate for use with the system manager 502 and the device 600. However, as will be described in more detail below with reference to FIG. 13, the present disclosure is Model-driven deep learning for devices (e.g., device 600) to address these challenges A system and method for:

[0104] Referring now to FIG. 13, an offline training system 70 according to an exemplary embodiment 0 is shown. The offline training system 700 is used for online control Generate one or more sets of learned weights to provide to the dispatch generation circuit 604. The system is configured to perform a model-driven deep learning process to The training system 700 includes a training optimization program circuit 702 and a system model. The circuit includes a simulator circuit 704.

[0105] The system model simulator circuit 704 calculates the economic dispatch (e.g., equipment set points and and operating point) and the system (i.e., device 600, and 600) to perform a simulation of Each scenario consists of a set of simulated measurements (e.g., internal temperature, external temperature) and equipment status (e.g., an indication of which devices 600 are online or offline) The system model simulator circuit 704 includes economic dispatch, simulated measurement results, and equipment status to the system model. The system model controls the building equipment. The same or similar to one or more models used in the model predictive control technique for For example, the system model is available in its entirety for reference. Patent Document 1, filed April 14, 2018, which is incorporated herein by reference, MANAGEMENT SYSTEM WITH SYSTEM IDENTIFICA TION USING MULTI-STEP AHEAD ERROR PREDIC This can be specified as the system model described in "SYSTEM MODEL 1".

[0106] The system model simulator circuit 704 calculates this cost over time. The simulation is run using the model for a period of time. The costs are calculated as 600 electricity consumption, utility prices for resource consumption over this period, various incentive-based demand response programs, demand power rates, and / or The economic cost function described in Patent Document 2, filed February 7, 2017, which is incorporated herein by reference, Costs may be determined by various other terms or variables. For example, costs may be determined by incorporating occupant comfort into costs. The cost function includes a penalty function (e.g., if the temperature is outside the comfortable temperature range, the cost is As used herein, the term "cost" refers to any such Therefore, the system model simulator circuit 704 may Receives economic dispatches and scenarios from the scheduling optimization program circuit 702 and calculates the cost is returned to the training optimization program circuit 702. In other words, the system model simulator The regulator circuit determines how the building system responds to a prescribed economic dispatch under a prescribed scenario. Simulate how you will respond and forecast the costs that will be incurred over a given period of time do.

[0107] The training optimization program circuit 702 performs various simulated measurements, as described in more detail below. to determine learned weights that minimize the cost of the measurement results and various equipment statuses. The training optimization program circuit 702 is configured as follows: Generates learned weights for use by the training optimization program The learned weights generated by the weighting circuit 702 are then fed to the dispatch generation circuit 604. The economic dispatch generated by the new system controls the equipment 600 to minimize costs. The training optimization program circuit 702 may adjust the neural network. and a system model simulator circuit 704, and an offline dispatch generation circuit 706. 06, a scenario circuit 708, and a weighting selection circuit 710.

[0108] The scenario circuit 708 generates various scenarios and dispatches them offline. The scenario circuit 708 provides the scenario to the generation circuit 706. The real time dispatch is shown as an input to the online dispatch generation circuit 604 in FIG. It includes the time status and the respective values ​​of the measurement results. For example, the scenario It may include values ​​such as indoor temperature, outdoor temperature, relative humidity, as well as indicators of equipment that may be connected. Therefore, the scenario circuit 708 controls the dispatch generation circuit 604 for online control. The same types of inputs that are provided to the offline dispatch generation circuit 706 are provided to the The scenario may also be provided to the system model simulator circuit 704. 08 is a scenario that provides a wide range of scenarios to the offline dispatch generation circuit 706. The user may change the status and value of the measurement.

[0109] The weighting selection circuit 710 is used by the offline dispatch generation circuit 706. The weights are configured to be changed to encourage convergence to the learned weights. The selection circuit 710 receives the costs from the system model simulator circuit 704 and selects the weighted costs. The selection circuit 710 selects new weightings to provide to the offline dispatch generation circuit 706. The weight selection circuit 710 determines the learned weights. To do this, we use gradient descent or stochastic gradient descent. In such cases, the ochastic gradient descent method can be used. In this case, the learned weights correspond to the cost extremum (e.g., minimum cost).

[0110] In some embodiments, a set of learned weights is determined for each set of device statuses. That is, for each combination of online and offline devices, a weighted selection circuit 710 is a corresponding set of learned weights (e.g., for online and offline devices) For online control, a different neural network model is determined for each combination. The display patch generation circuit 604 determines the current set of device statuses and accordingly A corresponding set of learned weights can then be applied. The real-time measurements can then be Neural networks are adjusted by their learned weights to generate dispatches. It can be handled by the network.

[0111] As shown in FIG. 13, the scenario provided by the scenario circuit 708 is The neural network of the dispatch generator 706 is an input to the economic dispatch is the output of the neural network. The weighting selection circuit 710 selects the system model The weighting is determined based on the costs generated by the simulator circuit 704. 708 includes a large number of scenarios (scenarios that may be difficult to find in real-world data). The training system 700 can then perform the training without having to wait for the actual event (including the actual event) to occur in the real world. (including scenarios that may be difficult to find in real-world data) Furthermore, the system model simulator circuit 704 allows the It replaces the need to wait for global cost data by generating simulated costs. Therefore, the scenario circuit 708 and the system model simulator circuit 704 cooperate to Use of data-based data to accelerate neural network training Therefore, real-world data is insufficient to train neural networks. Even if the training system 700 is used by the dispatch generation circuit 604, Reliably generate learned weights for

[0112] Referring again to FIG. 12, the learned weights generated by training system 700 are The assignment is provided to the dispatch generation circuit 604. The dispatch generation circuit 604 Apply the learned weights in the neural network of the dispatch generation circuit 604 Therefore, the dispatch generation circuit 604 generates a set of real-time equipment status and and measurements (i.e., real-world data that operates online to control a real set of building equipment) world scenario) and calculate costs (e.g., resource consumption costs or some other cost function) The neural network and the and dispatch generation circuit 604 requires substantially less computational power than traditional online control methods. Since the system manager 502 may need more power, it may be necessary to configure the control system using traditional methods. It is more efficient, consumes less energy, and is cheaper than conventional The processor may include a computing component.

[0113] Configuration of an exemplary embodiment The construction and arrangement of the systems and methods shown in the various exemplary embodiments are merely exemplary. Although only a few embodiments have been described in detail in this disclosure, many variations (e.g., For example, the size, dimensions, structure, shape and proportions of various elements, parameter values, mounting Changes in the way elements are arranged (e.g., changes in the method, use of materials, color, orientation, etc.) are possible. For example, reversing the position of elements. or otherwise vary, changing the nature or number of individual elements or positions. All such modifications are therefore expressly incorporated by reference herein. Any process or method step sequence or The sequence may also be varied or rearranged according to alternative embodiments. Other substitutions, variations in the design, operating conditions and arrangement of the exemplary embodiments may be made without departing from the scope of the present invention. Modifications, variations and omissions may be made.

[0114] As used herein, the term "circuitry" refers to a circuit configured to perform the functions described herein. In some embodiments, each "circuit" may include hardware as defined herein. may include a machine-readable medium for configuring hardware to perform the functions described in Circuitry includes, but is not limited to, processing circuitry, network interfaces, peripheral devices, input embodied as one or more circuit components including force devices, output devices, sensors, etc. In some embodiments, the circuitry may be one or more analog circuits, electronic circuits, (e.g., integrated circuits (ICs), discrete circuits, systems on chips (SOCs) tem on a chip) circuits), communication circuits, hybrid circuits, and any other In this regard, "circuit" may take the form of any of the types described herein. may include any type of component for performing or facilitating the performance of the operations described. For example, the circuits described herein may include one or more transistors, logic gates (e.g., NAND, AND, NOR, or XOR, NOT, XNOR, etc.), resistors, multipliers, It may include a multiplexer, a resistor, a capacitor, an inductor, a diode, a wire, and so on.

[0115] A "circuit" may also refer to one or more memory devices communicatively coupled to one or more memory devices. In this regard, one or more processors may include or otherwise accessible to one or more processors. In some embodiments, one or more processors may execute instructions accessible to the processor. may be embodied in a variety of ways. One or more processors may implement at least the functions described herein. In some embodiments, the method may be constructed in a manner sufficient to perform the operations described in , one or more processors may be shared by multiple circuits (e.g., circuits A and B). Circuit B may include or otherwise share the same processor, as in some exemplary implementations. In an embodiment, the data is stored or otherwise accessed through various areas of memory. Alternatively or additionally, one or more processors may Or it may be structured to perform some operations independently of multiple coprocessors. In another exemplary embodiment, two or more processors may be configured to operate in independent, parallel, pipelined fashion. coupled via a bus to allow for parallel or multi-threaded instruction execution Each processor may be one or more general-purpose processors, application-specific integrated circuits (AS), IC:application specific integrated circuit it), Field Programmable Gate Array (FPGA), mmable gate array, digital signal processor (DSP: dig instructions provided by a digital signal processor, or memory It may also be implemented as any other suitable electronic data processing component configured to execute the or multiple processors may be single-core processors, multi-core processors (e.g., duplex processors), Alcore processor, triple-core processor, quad-core processor, etc.), microphone In some embodiments, one or more processors may take the form of a The processor(s) may be external to the device, e.g., one or more processors may be remote processors (e.g., Alternatively or additionally, one or The multiple processors may be internal and / or local to the device. A given circuit or component may be locally (e.g., a local server, a local computing as part of a cloud-based system) or remotely (e.g., as part of a cloud-based server) To that end, the "circuitry" described herein may be located in one or more or may include components distributed throughout multiple locations. The present invention also contemplates methods, systems, and program products on any machine-readable medium for Embodiments of the present disclosure may be implemented using existing computer processors to perform this or other purposes. A dedicated computer processor for a suitable system incorporated for the purpose Alternatively, it may be implemented by a hardwired system. The state is a facility for holding or having stored machine-executable instructions or data structures. This includes a program product that includes a machine-readable medium, whether general-purpose or special-purpose. Any available software that can be accessed by a computer or other machine with a processor By way of example, such machine-readable media may include RAM, ROM, EPR, ROM, EEPROM, CD-ROM or other optical disk storage devices, magnetic disk storage devices or other magnetic storage device, or in the form of machine-executable instructions or data structures. can be used to hold or store program code, and can be general-purpose or special-purpose Any other medium accessible by a computer or other machine with a processor Combinations of the above are also included within the scope of machine-readable media. For example, a particular function or group of functions can be implemented on a general-purpose computer, a special-purpose computer, or or special-purpose processing machine for execution.

Claims

1. 1. A method comprising: Operating equipment to affect variable states or conditions of a space; determining a set of learned weights for a neural network by modeling a plurality of estimated costs of operating the equipment over a plurality of simulated scenarios using a model distinct from the neural network, each simulated scenario including simulated measurements related to the space, the neural network being configured to generate a simulated control dispatch including a setpoint for the equipment based on the simulated measurements, the model being configured to receive inputs including the plurality of simulated scenarios and the simulated control dispatches, and to output the estimated costs based on the plurality of simulated scenarios and the simulated control dispatches, the estimated costs being used to modify the set of learned weights; constructing a neural network for online control by applying the set of learned weights; applying the spatially related actual measurements to the neural network to generate a control dispatch for the device; controlling the device according to the control dispatch; A method comprising:

2. The method of claim 1 , wherein the set of learned weights is determined as a set of weights that minimizes the estimated cost of operating the equipment across the plurality of simulated scenarios.

3. Determining the set of learned weights comprises: identifying a state space thermal model of the space as the one model; defining a cost function using the state-space thermal model; generating, by the neural network for each simulated scenario, a simulated control dispatch based on the simulated measurements for the simulated scenario and a set of weightings; calculating the estimated cost of operating the equipment over a simulated time period of the simulated scenario given the simulated control dispatch and the set of simulated measurements by using the cost function; 2. The method of claim 1, comprising:

4. Determining the set of learned weights further comprises: modifying the set of weights to drive the estimated cost toward a minimum of the cost function; determining the set of learned weights as the set of weights that results in a minimum cost across the plurality of simulated scenarios; 4. The method of claim 3, comprising:

5. 10. The method of claim 1, wherein the control dispatch includes one or more of a temperature setpoint, a temperature schedule, a humidity setpoint, an airflow setpoint, a power level, an on / off setting, a damper position, a fan speed, a compressor frequency, or a resource consumption allocation.

6. The method of claim 1 , wherein the equipment comprises one or more of an airside system or a waterside system.

7. The method of claim 1 , wherein the equipment comprises one or more of a variable refrigerant flow system, a room air conditioner, or a packaged air conditioner.

8. 1. A system comprising: HVAC equipment operable to affect a variable state or condition of the space; one or more sensors configured to collect measurements related to the space; an offline training system configured to determine a set of learned weights for a neural network by modeling a plurality of estimated costs of operating the HVAC equipment over a plurality of simulated scenarios using a model distinct from the neural network, each simulated scenario including simulated measurements related to the space, the neural network configured to generate simulated control dispatches including setpoints for the HVAC equipment based on the simulated measurements, the model configured to receive inputs including the plurality of simulated scenarios and the simulated control dispatches and to output the estimated costs based on the plurality of simulated scenarios and the simulated control dispatches, the estimated costs being used to modify the set of learned weights; and an online control circuit configured to apply the measurements from the one or more sensors to the neural network to generate a control dispatch for the HVAC equipment, the neural network configured according to the set of learned weights and configured to control the HVAC equipment according to the control dispatch; and Including, the system.

9. 9. The system of claim 8, wherein the offline training system is configured to determine the set of learned weights as a set of weights that minimizes the estimated cost of operating the HVAC equipment across the plurality of simulated scenarios.

10. The offline training system includes: identifying a state space thermal model of the space; defining a cost function using the state-space thermal model; generating, by the neural network and for each simulated scenario, a simulated control dispatch based on the simulated measurements of the simulated scenario and a set of weightings; calculating the estimated cost of operating the HVAC equipment over a simulated time period of the simulated scenario given the simulated control dispatch and the set of simulated measurements by using the cost function; The system of claim 8 configured to:

11. The offline training system includes: modifying the set of weights to drive the estimated cost toward a minimum of the cost function; determining the set of learned weights as the set of weights that results in a minimum cost across the plurality of simulated scenarios; The system of claim 10 configured to:

12. The system of claim 8 , wherein the HVAC equipment comprises one or more of an air-side system or a water-side system.

13. the online control circuitry is contained locally with the HVAC equipment; The system of claim 8 , wherein the offline training system comprises one or more cloud computing resources.

14. 1. A system comprising: a chiller operable to affect the temperature of the space; one or more sensors configured to collect measurements related to the space; an offline training system configured to determine a set of learned weights for a neural network by modeling a plurality of estimated costs of operating the chiller across a plurality of simulated scenarios using a model distinct from the neural network, each simulated scenario including simulated measurements related to the space, the neural network configured to generate a simulated control dispatch including a setpoint for the chiller based on the simulated measurements, the model configured to receive inputs including the plurality of simulated scenarios and the simulated control dispatches, and to output the estimated costs based on the plurality of simulated scenarios and the simulated control dispatches, the estimated costs being used to modify the set of learned weights; and an online control circuit configured to apply the measurements from the one or more sensors to the neural network to generate a control dispatch for the chiller, the neural network configured according to the set of learned weights and configured to control the chiller according to the control dispatch; and Including, the system.

15. 15. The system of claim 14, wherein the offline training system is configured to determine the set of learned weights as a set of weights that minimizes the estimated cost of operating the chiller across the plurality of simulated scenarios.

16. The offline training system includes: identifying a state space thermal model of the space; defining a cost function using the state-space thermal model; generating, by the neural network and for each simulated scenario, a simulated control dispatch based on the simulated measurements of the simulated scenario and a set of weightings; calculating the estimated cost of operating the chiller over a simulated time period of the simulated scenario given the simulated control dispatch and the set of simulated measurements by using the cost function; The system of claim 14 configured to:

17. The offline training system includes: modifying the set of weights to drive the estimated cost toward a minimum of the cost function; determining the set of learned weights as the set of weights that results in a minimum cost over the simulated scenarios; 17. The system of claim 16 configured to:

18. The system of claim 14 , wherein the chiller comprises one or more of a room air conditioner, a packaged air conditioner, or a variable refrigerant flow device.

19. the online control circuitry is contained locally with the chiller; The system of claim 14 , wherein the offline training system comprises one or more cloud computing resources.

20. The system of claim 14 , wherein the control dispatch includes a temperature set point.

Images