Systems and methods for controlling the operation of a heating, ventilation, and air conditioning (HVAC) system

Abstract

A control system for controlling the operation of a heating, ventilation, and air conditioning (HVAC) system is provided. The control system includes an input interface configured to receive data indicating a target thermal state distribution in an environment; and a memory configured to store an air dynamics model (ADM) and an HVAC model. The control system further includes a processor configured to: invert the ADM to estimate boundary condition values at inlet locations that define a target thermal state at the inlet locations, the boundary condition values at the inlet locations resulting in the target thermal state distribution in the environment; determine target control parameters of actuators of the HVAC system that result in the target thermal state at the inlet locations using the HVAC model; and submit control commands to the HVAC system to operate the actuators of the HVAC system according to the control parameters.

Description

Technical Field

[0001] The present invention relates generally to heating, ventilation, and air conditioning (HVAC) systems and, more particularly, to systems and methods for controlling the operation of heating, ventilation, and air conditioning (HVAC) systems. Background Art

[0002] Airflow dynamics has a significant impact on building energy simulation (BES) for heating and cooling load calculations. Ventilation systems involving stratified airflow distribution are on the rise as they can achieve better thermal comfort and energy performance. Examples include displacement ventilation, natural ventilation, and advanced ventilation methods. These ventilation systems are provided for large spaces or spaces with high heat gain. Therefore, a combination of two dynamics, BES and airflow dynamics (represented by CFD), is required. Building Energy Simulation (BES) is used to predict heat loads, systems (buildings), responses to heat loads, and the resulting energy use, as well as building energy performance indicators such as occupant comfort and energy costs. Computational Fluid Dynamics (CFD) is used to predict airflow, temperature, and humidity distribution information in a room.

[0003] Each of these models is subject to certain limitations in its own right. To simplify computation, BES models (e.g., EnergyPlus, TRNSYS, ESP-r, IDA-ICE, BSIM) assume well-mixed indoor air and therefore do not account for stratification of airflow and temperature or non-uniform sources of heat load. This limitation leads to inaccurate predictions of building energy consumption and the closed-loop control performance of the HVAC system. Conversely, CFD models do not account for complex boundary conditions such as ambient weather (solar radiation), air handling units (fans), heat exchangers (evaporators), and vapor compression system models.

[0004] Simulating and designing energy-efficient heating, ventilation, and air conditioning (HVAC) systems is a challenging task due to the multi-scale, highly nonlinear, and complex nature of their dynamics.

[0005] Other approaches combine CFD with BES models for airflow dynamics and heat transfer prediction, incorporating the complex dynamics of phase-change refrigerant flow, ambient weather factors, solar radiation, wall heat loss, and other factors. However, these approaches only consider the feedback control provided by the BES and employ CFD simulations in a simplistic manner. Furthermore, current co-simulation methods ignore airflow optimization and, in the case of building-side optimization, employ a brute-force approach. Therefore, a system and method are needed to provide optimal design and control of HVAC systems. Summary of the Invention

[0006] Some embodiments aim to provide systems and methods for controlling the operation of a heating, ventilation, and air conditioning (HVAC) system. Furthermore, some embodiments aim to achieve a target thermal state profile and optimize the performance of the HVAC system. Additionally, or alternatively, some embodiments aim to determine one or more design variables for optimizing the design of the HVAC system.

[0007] Some embodiments are based on changing the state of the HVAC system's actuators to ensure thermal comfort for the occupants of the conditioned environment. Examples of actuator states include the speed of the HVAC compressor, the positions of various valves, the rotational position of the air vanes that direct exhaust air, and the like. To this end, HVAC actuator-based control aims to determine the state of the HVAC actuators to meet a set point desired by the occupants of the environment. HVAC actuator-based control determines the state of the HVAC actuators to meet the set point. Subsequently, air is output to the environment via an inlet based on the determined state of the HVAC actuators that meets the set point. In HVAC actuator-based control, the state of the HVAC actuators is the primary target of control, while the thermal state of the output air is secondary and treated as a byproduct. The thermal state of the output air includes one or a combination of the temperature, velocity, and humidity of the air output by the HVAC system to the environment. Because the thermal state of the output air is controlled by changing the state of the HVAC actuators, it may not be possible to directly determine the thermal state of the output air.

[0008] To this end, some embodiments invert this criterion by considering the thermal state of the output air as the primary control objective. First, the thermal state of the output air is determined that results in a desired setpoint. Then, the state of the HVAC actuators that results in the determined thermal state of the output air is determined. That is, the state of the HVAC actuators is determined with respect to the output of the HVAC system from the inlet, rather than with respect to the setpoint.

[0009] To this end, some embodiments are based on the recognition that the thermal state of the output air can be used as a boundary condition to estimate the thermal state distribution in the conditioned environment. An airflow dynamics model (ADM) can be used to estimate the thermal state distribution under boundary conditions. The ADM uses the Navier-Stokes equations and the energy equation to represent the aerodynamics in the environment, where computational fluid dynamics (CFD) calculations are used to solve the Navier-Stokes equations and the energy equation to estimate the distribution of the thermal state. The boundary conditions are defined in two ways, one is through the building envelope model (BEM), which uses information such as the dimensions, construction materials, lighting, etc. of the building / space; the other is the output of the HVAC system, such as the inlet air velocity, direction, humidity, and temperature. Given the ADM, BEM, and boundary conditions, computational fluid dynamics (CFD) calculations can be used to estimate the thermal state distribution. However, different boundary conditions can lead to the same thermal state distribution.

[0010] To this end, some embodiments are based on the recognition that boundary conditions related to the output of the HVAC system can be estimated that achieve a desired thermal state distribution and also optimize the performance of the HVAC system. Thus, the boundary conditions governed by the output of the HVAC system are used as control parameters to optimize the performance of the HVAC system, rather than as inputs.

[0011] To this end, some embodiments are based on the goal of estimating boundary condition values that define a target thermal state at the inlet location that results in an ideal thermal state distribution in the environment and that also optimizes the performance of the HVAC system to achieve the target thermal state at the inlet location.

[0012] To achieve the above relationships, a model of the HVAC system is used to establish a relationship linking the HVAC system's operation with the airflow dynamics and parameters to be optimized. To this end, some embodiments use an HVAC model, which is derived from the structure of the HVAC system, to achieve the above relationships. The HVAC model can be used to estimate outputs from the HVAC system (e.g., inlet air velocity, direction, humidity, and temperature). The estimated outputs from the HVAC system can be used as boundary conditions for the ADM, thereby linking the ADM and the HVAC model.

[0013] To this end, some embodiments are based on the recognition that both the HVAC model and the building envelope model (BEM) define the boundary conditions of the ADM. For example, the HVAC model provides the ADM with outputs from the HVAC system, while the building envelope model (BEM) provides boundary conditions for the temperature of all surfaces and / or the thermal state of the air at the walls of the environment when the environment is not conditioned by the HVAC system. Thus, given these boundary conditions and appropriate initial conditions, the ADM provides an estimate of a target thermal state at the inlet location that results in an ideal thermal state distribution in the environment. In this way, the operation of the HVAC system can be connected to the ideal thermal state distribution in the environment, and the performance of the HVAC system can be optimized while achieving the ideal thermal distribution in the environment. In some embodiments, optimizing the performance of the HVAC system includes optimizing the energy consumption of the HVAC system.

[0014] The ADM can be used to determine the thermal state distribution in a room given boundary conditions. However, some embodiments invert the ADM to estimate boundary condition values at inlet locations that define a target thermal state at the inlet locations, which results in the target thermal state distribution in the environment. Furthermore, using the HVAC model, target control parameters for actuators of the HVAC system corresponding to the target thermal state distribution in the environment are determined. Subsequently, control commands corresponding to the target control parameters are generated and submitted to the HVAC system. In particular, the control commands are submitted to a controller of the HVAC system to operate actuators and / or components of the HVAC system according to the target control parameters.

[0015] Some embodiments are based on the recognition that different combinations of target thermal state values at the inlet location result in target thermal state distributions in the environment. To this end, some embodiments select targets for the combination of thermal states based on a performance metric of the HVAC system. The performance metric of the HVAC system is defined by a multi-objective cost function. The multi-objective cost function is a combination of the operating cost of the HVAC system and the difference between the target thermal state distribution and the corresponding current thermal state distribution. Minimizing the multi-objective cost function includes iteratively minimizing the multi-objective cost function until a termination condition is satisfied.

[0016] According to some embodiments, each iteration of the iterative minimization process includes determining the sensitivity of the cost function to an update of the boundary conditions at the inlet location. The boundary conditions are updated in the direction of the sensitivity of the multi-objective cost function. Furthermore, the current distribution of thermal conditions is determined according to the ADM using the updated boundary conditions, and then the operating cost of the HVAC system resulting from the updated boundary conditions at the inlet location is determined. For example, a termination condition for the iterative minimization process is satisfied when the sensitivity of the multi-objective cost function is less than a first threshold, the value of the cost function is less than a second threshold, or the number of iterations is greater than a third threshold.

[0017] Some embodiments are based on the recognition that optimizing the performance of an HVAC system includes optimizing one or more design variables, including HVAC system design variables and environmental design variables, wherein the HVAC system design variables include the number of inlets, the location of the inlets on the walls of the environment, the size of the air conditioner (AC) diffuser, and the number of AC units, and wherein the environmental design variables include the thickness of the walls of the environment, the material of the wall insulation, the material of the window insulation, and the window shading.

[0018] Furthermore, some embodiments are based on the recognition that one or more design variables can be optimized by iteratively minimizing a multi-objective cost function. According to some embodiments, each iteration of the iterative minimization includes determining the sensitivity of the multi-objective cost function to an update of the one or more design variables. The one or more design variables are updated in the direction of the sensitivity of the multi-objective cost function. Furthermore, a current distribution of thermal conditions is determined according to the ADM using the updated one or more design variables, and then an operating cost of the HVAC system resulting from the updated one or more design variables is determined. The iterative minimization terminates when a termination condition is satisfied.

[0019] Therefore, one embodiment discloses a control system for controlling the operation of a heating, ventilation, and air conditioning (HVAC) system, the HVAC system being configured to condition an indoor environment by pushing air into the environment via a set of inlets at a set of locations arranged on one or more walls of the environment, wherein the thermal condition of the air pushed into the environment at the inlet locations comprises one or a combination of the temperature, velocity, and humidity of the air, the control system comprising: an input interface configured to receive data indicating a target thermal condition distribution in the environment, wherein the thermal condition at a location in the environment comprises one or a combination of the temperature, velocity, and humidity of the air; and a memory configured to store an airflow dynamics model AD. M and an HVAC model defining the dynamics of the HVAC system, the ADM defining a thermal state distribution in the environment, the thermal state distribution being affected by boundary conditions of the thermal state of air at the walls of the environment; a processor configured to: invert the ADM to estimate boundary condition values at the inlet positions that define the target thermal state at the inlet positions, the boundary condition values at the inlet positions resulting in the target thermal state distribution in the environment; using the HVAC model, determine target control parameters of the actuators of the HVAC system that result in the target thermal state at the inlet positions; and submit control commands to the HVAC system to operate the actuators of the HVAC system according to the control parameters.

[0020] Therefore, another embodiment discloses a method for controlling the operation of a heating, ventilation and air conditioning (HVAC) system, the HVAC system being configured to condition an indoor environment by pushing air into the environment via a set of inlets at a set of locations arranged on one or more walls of the environment, wherein the thermal state of the air pushed into the environment at the inlet locations includes one or a combination of the temperature, velocity and humidity of the air, wherein the method uses a processor coupled to a memory storing an airflow dynamics model ADM and an HVAC model defining the dynamics of the HVAC system, the ADM defining a thermal state distribution in the environment, the thermal state distribution being affected by boundary conditions of the thermal state of the air at the walls of the environment, and wherein the method, when executed by the processor, The processor is coupled to the stored instructions when the processor is coupled to the stored instructions to implement the steps of the method, the steps comprising: receiving data indicating a target thermal state distribution in the environment, wherein the thermal state at a location in the environment includes one or a combination of temperature, velocity, and humidity of air; inverting the ADM to estimate boundary condition values at the inlet location that define the target thermal state at the inlet location, the boundary condition values at the inlet location resulting in the target thermal state distribution in the environment; using the HVAC model, determining target control parameters of the actuators of the HVAC system that result in the target thermal state at the inlet location; and submitting control commands to the HVAC system to operate the actuators of the HVAC system according to the control parameters.

[0021] definition

[0022] When describing embodiments of the present invention, the following definitions apply throughout (including above).

[0023] A "vapor compression system" refers to a system that uses a vapor compression cycle to move a refrigerant through the system's components, based on the principles of thermodynamics, fluid dynamics, and / or heat transfer. Vapor compression systems can include, but are not limited to, heat pumps, refrigeration, and air conditioning systems. Vapor compression systems have applications beyond conditioning residential or commercial spaces. For example, a vapor compression cycle can be used to cool computer chips in high-performance computing applications.

[0024] A "radiant system" is a system that supplies heat directly to building materials (such as concrete floors) and heats the space and occupants primarily through radiant heat transfer. Radiant systems use a heat transfer medium (such as hot or cold water) that runs through tubes embedded in the building materials, but they can also use heating wires or a heat exchanger connected to a steam compressor.

[0025] An "HVAC" system refers to any building heating, ventilation, and air conditioning (HVAC) system that implements a vapor compression cycle. HVAC systems range from systems that only supply outdoor air to the building's occupants, to systems that only control the building's temperature, to systems that control both temperature and humidity.

[0026] "Component of a vapor compression system" means any component of a vapor compression system whose operation can be controlled by a control system. Such components include, but are not limited to: a variable speed compressor, which is used to compress and pump refrigerant through the system; an expansion valve, which is used to provide a pressure drop between the high-pressure and low-pressure portions of the system; and evaporating and condensing heat exchangers.

[0027] "Evaporator" refers to the heat exchanger in a vapor compression system in which the refrigerant passing through the heat exchanger evaporates over the length of the heat exchanger, resulting in a higher specific enthalpy of the refrigerant at the heat exchanger outlet than at the heat exchanger inlet, and the refrigerant generally changes from a liquid to a gas. There can be one or more evaporators in a vapor compression system.

[0028] A "condenser" in a vapor compression system is a heat exchanger in which the refrigerant passing through the heat exchanger condenses over the length of the heat exchanger, resulting in the refrigerant having a lower specific enthalpy at the heat exchanger outlet than at the heat exchanger inlet, and the refrigerant generally changes from a gas to a liquid. There may be one or more condensers in a vapor compression system.

[0029] A "vent" is a point at the edge of a system of passageways used for heating, ventilation, and air conditioning (HVAC) to deliver and remove air from a conditioned environment. Thus, vents are used to ensure acceptable indoor air quality and thermal comfort. The thermal conditions at the vent can include one or a combination of the temperature, velocity, and humidity of the air output by the HVAC system to the environment.

[0030] A "circuit" refers to a closed path interconnecting electrical signals between components such as a processor, memory, or actuator.

[0031] A "set point" is a desired value for a variable, such as building space temperature. The term "set point" applies to any specific value for a specific set of variables.

[0032] "Computer" means any device that can accept structured input, process the structured input according to specified rules, and produce as output the results of the processing. Examples of computers include general-purpose computers; supercomputers; mainframes; supermicrocomputers; minicomputers; workstations; microcomputers; servers; interactive televisions; hybrid combinations of computers and interactive televisions; and application-specific hardware that emulates computers and / or software. A computer may have a single processor or multiple processors, and the processors may operate in parallel and / or not in parallel. A computer also refers to two or more computers connected together by a network for transmitting or receiving information between the computers. One example of such a computer includes a distributed computer system for processing information by computers linked by a network.

[0033] A "central processing unit (CPU)" or "processor" is a computer or the part of a computer that reads and executes software instructions.

[0034] "Memory" or "computer-readable medium" refers to any storage device for storing data accessible by a computer. Examples include: magnetic hard disks; floppy disks; optical disks, such as CD-ROMs or DVDs; magnetic tape; memory chips; and carrier waves for carrying computer-readable electronic data, such as those used to transmit and receive electronic mail or access the Internet; and computer memory, such as random access memory (RAM).

[0035] "Software" refers to the prescribed rules for operating a computer. Examples of software include: software; code segments; instructions; computer programs; and programming logic. The software of an intelligent system may be self-learning.

[0036] A "module" or "unit" refers to a basic component in a computer that performs a task or part of a task, and can be implemented by software or hardware.

[0037] "Controller," "control system," and / or "regulator" refers to a device or group of devices that manages, directs, directs, or regulates the behavior of other devices or systems. A controller can be implemented as hardware, a processor configured to operate using software, or a combination thereof. A controller can also be an embedded system.

[0038] The presently disclosed embodiments will be further explained with reference to the accompanying drawings. The drawings shown are not necessarily to scale, with emphasis generally being placed upon illustrating the principles of the presently disclosed embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 A schematic overview of the principles of some embodiments for controlling a heating, ventilation, and air conditioning (HVAC) system is shown.

[0040] Figure 2A block diagram of a system for controlling and optimizing the performance of a heating, ventilation, and air conditioning (HVAC) system is shown, according to some embodiments.

[0041] Figure 3 A block diagram illustrating inputs and outputs of a building envelope model (BEM) according to some embodiments.

[0042] Figure 4 A schematic diagram illustrating pre-processing and post-processing involved in a computational fluid dynamics (CFD) simulation to obtain the thermal state of a target region, according to some embodiments.

[0043] Figure 5 A schematic diagram illustrating obtaining thermal conditions at a target zone based on CFD simulation and optimal air conditioner input according to some embodiments is shown.

[0044] Figure 6 A schematic diagram illustrating minimizing a cost function to achieve a target thermal state in a target zone while improving the energy efficiency of an HVAC system, according to some embodiments.

[0045] Figure 7A A schematic diagram is shown for constructing and evaluating a cost function that includes both a target thermal state and energy consumption of an HVAC system, according to some embodiments.

[0046] Figure 7B A schematic diagram illustrating a vapor compression cycle as an example of a refrigerant cycle according to some embodiments is shown.

[0047] Figure 8A A block diagram is shown as an exemplary optimization method for implementing optimal air conditioner input according to a target thermal state with minimal energy, according to some embodiments.

[0048] Figure 8B Schematic diagram of collaborative simulation with feedback control of equipment and optimal control of airflow in a room based on thermal comfort and energy cost functions is shown.

[0049] Figure 9 A schematic diagram of an HVAC system including a controller in communication with the system employing the principles of some embodiments is shown.

[0050] Figure 10 A schematic diagram illustrating an iteration of minimization of a multi-objective cost function for designing an HVAC system, according to some embodiments.

[0051] Figure 11 A schematic diagram of an HVAC system designed for a room is shown, according to some embodiments. DETAILED DESCRIPTION

[0052] In the following description, for purposes of explanation, numerous specific details are set forth to provide a thorough understanding of the present disclosure. However, it will be apparent to those skilled in the art that the present disclosure may be practiced without these specific details. In other cases, devices and methods are shown in block diagram form only to avoid obscuring the present disclosure.

[0053] As used in this specification and claims, the terms "for example," "for example," and "such as," and the verbs "comprise," "have," "include," and their alternative forms, when used in conjunction with a list of one or more components or other items, should each be understood as open-ended, meaning that the list should not be construed to exclude other, additional components or items. The term "based on" means based, at least in part, on. Furthermore, it should be understood that the phrases and terminology used herein are for descriptive purposes only and should not be construed as limiting. Any generic names used in this description are for convenience only and have no legal or limiting effect.

[0054] Figure 1 A schematic diagram of the principles of some embodiments for controlling a heating, ventilation, and air conditioning (HVAC) system is shown. The embodiments are based on the recognition that the goal of HVAC control is to change the states of the actuators of the HVAC system to ensure thermal comfort for occupants of the environment to be conditioned. Examples of actuator states include the speed of the HVAC compressor, the positions of various valves, the rotational position of the air vanes that direct exhaust air, etc. To this end, the HVAC actuator-based control 100 is intended to determine the state 104 of the HVAC actuator to meet the set point 102 desired by the occupants of the environment. The HVAC actuator-based control 100 determines the state 104 of the HVAC actuator to meet the set point 102. Subsequently, air is output to the environment via an inlet based on the state 104 of the HVAC actuator determined to meet the set point 102.

[0055] In HVAC actuator-based control 100, the state 104 of the HVAC actuator is the primary target of control, while the thermal state 106 of the output air is secondary and treated as a byproduct. The thermal state 106 of the output air includes one or a combination of the temperature, velocity, and humidity of the air output by the HVAC system to the environment. Because the thermal state 106 of the output air is controlled by changing the state 104 of the HVAC actuator, it may not be possible to directly determine the thermal state of the output air.

[0056] Some embodiments invert this criterion by considering the thermal state 106 of the output air as the primary target of control. First, the thermal state of the output air that results in achieving the desired set point is determined (i.e., the thermal state 106 of the output air). Subsequently, the state 104 of the HVAC actuator that results in the determined thermal state of the output air is determined. That is, the state 104 of the HVAC actuator is determined for the output of the HVAC system from the inlet, rather than for the set point.

[0057] Such a transformation simplifies the internal control of the HVAC actuator and allows the important part of the HVAC control to be transferred from the dynamic domain of the HVAC actuator to the air flow dynamic domain. To this end, some embodiments use the thermal state 106 of the output air as a boundary condition to estimate the thermal state distribution in the conditioned environment. The air flow dynamic model (ADM) 110 can be used to estimate the thermal state distribution under the boundary conditions (i.e., Figure 1 The thermal state distribution in the room 114). The boundary conditions are defined in two ways, one by the building envelope model (BEM) 108, which uses information such as the dimensions of the building / space, construction materials, lighting, etc.; and the other by the output of the HVAC system, such as the inlet air rate, direction, humidity, and temperature. Given the ADM 110, the BEM 108, and the boundary conditions, computational fluid dynamics (CFD) calculations can be used to estimate the thermal state distribution. In other words, some boundary conditions are not related to the operation of the HVAC system (at least not directly), while other boundary conditions are directly dependent on the operation of the HVAC system. Examples of HVAC-related boundary conditions include the geometry of the room being conditioned, heat exchange through the walls of the room, etc. Examples of HVAC-related boundary conditions include the thermal state of the air that the HVAC system pushes into the room via the inlet. Knowing all the boundary conditions, the ADM can be used to determine the thermal state distribution within the room.

[0058] It is noteworthy that different combinations of boundary conditions can result in the same thermal state distribution. To this end, some embodiments are based on the recognition that boundary conditions associated with the output of the HVAC system can be estimated to achieve a desired thermal state distribution and also optimize the performance of the HVAC system. Thus, the boundary conditions are used as control parameters to optimize the performance of the HVAC system, rather than as inputs.

[0059] To this end, some embodiments are based on the goal of estimating boundary condition values that define a target thermal state at the inlet location that results in a desired thermal state distribution in the environment and also optimizes the performance of the HVAC system. To achieve this goal, a model of the HVAC system is used to establish a relationship connecting the operation of the HVAC system with the airflow dynamics and the parameters to be optimized.

[0060] To this end, some embodiments use an HVAC model 112 that is given the structure of the HVAC system to implement the above relationship. The HVAC model 112 can be used to estimate the outputs from the HVAC system (e.g., inlet air velocity, direction, humidity, and temperature). The estimated outputs from the HVAC system can be used as boundary conditions for the ADM 110, thereby connecting the ADM 110 and the HVAC model 112.

[0061] According to some embodiments, both the HVAC model 112 and the building envelope model (BEM) 110 define boundary conditions for the ADM 110. For example, the HVAC model 112 provides the ADM 110 with output from the HVAC system, while the building envelope model (BEM) provides boundary conditions for the temperatures of all surfaces and / or the thermal state of the air at the walls of the environment when the environment is not conditioned by the HVAC system. Thus, given these boundary conditions and appropriate initial conditions, the ADM 110 provides an estimate of the target thermal state at the inlet locations that results in a desired thermal state distribution in the environment. In this way, the operation of the HVAC system can be linked to the desired thermal state distribution in the environment, optimizing the performance 116 of the HVAC system while achieving the desired thermal distribution in the environment. In some embodiments, optimizing the HVAC system performance 116 includes optimizing the energy consumption of the HVAC system. In other embodiments, optimizing the HVAC system performance 116 includes optimizing design parameters (e.g., the number of inlets, the location of inlets at the walls of the environment, the size of the air conditioner (AC) diffusers, the number of AC units, etc.). Some embodiments are based on the recognition that HVAC system performance 116 can be optimized by iteratively minimizing a cost function that is a combination of the operating cost of the HVAC system and the difference between a desired thermal state profile and a current thermal state profile.

[0062] System Overview

[0063] Figure 2 A block diagram of a system 200 for controlling and optimizing the performance of a heating, ventilation, and air conditioning (HVAC) system, according to some embodiments, is shown. An HVAC system is configured to deliver air to an environment through a set of inlets disposed at a set of locations on one or more walls of the environment. The environment can be a room or space in a building, or an entire building in which the HVAC system is installed. In some embodiments, the environment can correspond to a space in a building where occupants reside or inhabit. The thermal condition of the air delivered to the environment at the inlet locations includes one or a combination of the air's temperature, velocity, and humidity.

[0064] The system 200 may have a number of interfaces that connect the system 200 to other systems and devices. For example, a network interface controller (NIC) 214 is adapted to connect the system 200 via a bus 212 to a network 216, which connects the system 200 to a set of sensors. Via the network 216, the system 200 receives data 218 indicating a target thermal condition distribution in the environment, either wirelessly or via wires. Furthermore, the system 200 includes a control interface 226 configured to submit control commands to a controller 228 of the HVAC system to operate the actuators 222 of the HVAC system.

[0065] System 200 includes an input interface 202, which is configured to receive data indicating a target thermal state distribution in an environment. In some embodiments, the target thermal state distribution is received for a portion of the environment, which is referred to as a target area in the environment. In this case, the target area is maintained according to the target thermal state distribution, while the remaining areas of the environment are maintained with different thermal state distributions. Therefore, in these cases, the HVAC system adjusts the environment, which results in an uneven distribution of at least two different thermal state values at two different locations in the environment. In some other embodiments, the target thermal state distribution is provided for the entire environment. In this case, a uniform thermal state distribution will be generated in which each location in the environment has the same thermal state. In another embodiment, the input interface 202 is configured to obtain environmental data indicating environmental geometry and energy exchange. The environmental data indicating energy exchange is based on one or more of wall insulation, window insulation, average external temperature, or solar radiation at the environment wall.

[0066] System 200 includes a processor 204 configured to execute stored instructions and a memory 206 storing instructions executable by processor 204. Processor 204 may be a single-core processor, a multi-core processor, a computing cluster, or any number of other configurations. Memory 206 may include random access memory (RAM), read-only memory (ROM), flash memory, or any other suitable memory system. Processor 204 is connected to one or more input and output devices via bus 212.

[0067] Additionally, the system 200 includes a building envelope model (BEM) 208 , a heating, ventilation, and air conditioning (HVAC) model 210 , and an air dynamics model (ADM) 224 . The aforementioned models are executed by the processor 204 .

[0068] According to some embodiments, the BEM 208 provides boundary conditions for the thermal state of the air at the walls of the environment when the environment is not conditioned by the HVAC system based on the environment data. The processor 204 initializes the boundary conditions by submitting the thermal state values outside the environment to the BEM 208. In some embodiments, the BEM 208 uses information such as the size of the building / environment, the construction materials, the lighting, etc. to define the boundary conditions for the thermal state of the air at the walls of the environment. Figure 3 Explain BEM 208 in detail.

[0069] In some embodiments, the HVAC model 210 defines the dynamics of the HVAC system. The HVAC model 210 can be used to estimate the outputs from the HVAC system (e.g., inlet air velocity, direction, humidity, and temperature). Furthermore, the estimated outputs from the HVAC system are fed into the ADM 224 as boundary conditions. The ADM 224 defines a thermal state distribution affected by boundary conditions of the thermal state of the air at the ambient walls. In some embodiments, the ADM 224 uses the Navier-Stokes equations and the energy equation to represent the dynamics of the ambient air. Furthermore, computational fluid dynamics (CFD) calculations are used to solve the Navier-Stokes equations and the energy equation to estimate the thermal state distribution.

[0070] To facilitate data exchange between the aforementioned models (208, 210, and 224), one embodiment defines a fixed synchronization time step for data exchange. Additionally or alternatively, according to some embodiments, the synchronization time step is larger than the integration time step of each model, which can be fixed or adaptive. A quasi-dynamic data synchronization scheme is used in the combined simulation, whereby the programs exchange data only during synchronization and retain the received data unchanged between synchronizations.

[0071] In some other embodiments, the processor 204 is configured to invert the ADM 224 to estimate boundary condition values at the inlet locations that define target thermal states that result in a target thermal state distribution in the environment. Different combinations of target thermal state values at the inlet locations result in the target thermal state distribution in the environment. To this end, the processor 204 is configured to select a combination of thermal states based on a performance metric of the HVAC system. The performance metric of the HVAC system is defined by a multi-objective cost function. The multi-objective cost function is a combination of the operating cost of the HVAC system and the difference between the target thermal state distribution and the corresponding current thermal state distribution. The multi-objective cost function is iteratively minimized by the processor 204 until a termination condition is satisfied.

[0072] In some other embodiments, processor 204 is configured to determine design variables by minimizing a multi-objective cost function. The design variables include HVAC system design variables and environmental design variables. The HVAC system design variables include the number of inlets, the location of the inlets on the walls of the environment, the size of the air conditioning (AC) diffuser, and the number of AC units. According to some embodiments, the environmental design variables include the thickness of the walls of the environment, the insulation material of the walls, the insulation material of the windows, and the shading of the windows.

[0073] Additionally, the system 200 includes an output interface 220 configured to output one or more design variables. In some embodiments, the output interface 220 is configured to submit control commands to the HVAC system.

[0074] Figure 3 A block diagram illustrates the inputs and outputs of a building envelope model (BEM) 314, according to some embodiments. Buildings include walls, windows, roofs, and other components that play a key role in determining comfort. Various information, such as weather data 300, building location 302, building geometry 304, building materials 306, occupancy 308, internal loads 310, and lighting 312, are provided as inputs 322 to the BEM 314. Weather data 300 can be obtained from meteorological agencies, such as the National Weather Service in the United States. Weather data 300 provides information such as outside air temperature and humidity, weather patterns, and so on. Inputs 322 influence the energy consumption of the HVAC system.

[0075] To this end, in some embodiments, BEM 314 uses input 322 to determine building energy consumption 320. Additionally, in some embodiments, building envelope model 314 uses weather data 300 and building location 302 to calculate heating loads and resulting energy consumption and thermal comfort 316 based on first principles equations.

[0076] Because the building envelope model 314 is used to assess the thermal performance of a building over a full year, it is not integrated with the CFD simulation. With the BEM, it is sometimes assumed that the conditioned environment is well-mixed, and inhomogeneous distributions of velocity, temperature, pressure, and concentration are ignored to speed up the CFD simulation. To this end, some embodiments are based on the goal of combining the building envelope model 314 with the CFD simulation. In some embodiments, the building envelope model 314 provides boundary conditions 318 to the CFD, and the CFD simulates the airflow dynamics in the environment based on the provided boundary conditions 318. In addition, the CFD sends average airflow and heat transfer information back to the building envelope model 314, thus completing a closed-loop analysis.

[0077] Figure 4A schematic diagram illustrates the pre-processing 400 and post-processing 402 involved in a computational fluid dynamics (CFD) simulation to obtain the thermal state of a target area, according to some embodiments. CFD is a branch of fluid mechanics that uses numerical analysis and data structures to solve and analyze equations related to fluid flow. Mathematical models of physical cases and numerical methods are used in software tools (such as CFD) to analyze fluid flow. For example, the Navier-Stokes equations are specified as the mathematical model of the physical case. This describes the changes in the physical properties of both fluid flow and heat transfer. The fluid in this case may be air. The mathematical model varies depending on the content of the problem (such as heat transfer, mass transfer, phase change, chemical reaction, etc.). Here, the room is referred to as the environment in which the occupants live. The pre-processing 400 step includes using computer-aided design (CAD) to define the geometry and physical boundaries of the problem, thereby extracting the room geometry 404. Furthermore, a mesh can be generated 406 based on the extracted geometry, and operating parameters, including boundary conditions, can be set 408 for the air conditioner. The air conditioner can correspond to an HVAC system. Post-processing 402 includes obtaining a thermal state 412 at the target region based on the pre-processing step 400 and a CFD solution 410 of the governing fluid dynamics equations.

[0078] In some embodiments, the geometry can be the architecture of the room and can be extracted, for example, from an architectural blueprint of the room. Furthermore, the volume occupied by the fluid in the room is determined. After extracting the room geometry, a grid is generated 406 based on the extracted geometry and the air conditioner locations. In some embodiments, the extracted geometry is divided into discrete cells (i.e., a grid).

[0079] The mesh can be uniform or non-uniform, structured or unstructured, and composed of a combination of hexahedral, tetrahedral, prism, pyramidal, or polyhedral elements. The optimal number of mesh points in the domain is selected so that important flow structures (such as circulation in the room, active rising or falling flow structures near air conditioners or occupants, etc.) are captured in the mesh with high resolution. For example, the number of mesh points around the inlet can be increased to better capture the high-rate dynamics near the inlet. Other embodiments include increasing the number of elements near the heat load, occupants, or outlets. The optimal mesh can be obtained through mesh sensitivity analysis. To perform a mesh sensitivity analysis, the number of nodes is systematically increased, for example, by doubling the number of elements. For each mesh, a value of interest is monitored. The optimal mesh is one where the value of interest does not change as the number of mesh points is increased. For applications in structural fluid dynamics, such values of interest can be the average temperature in the target area, heat transfer through walls, the average temperature at the outlet, etc.

[0080] In addition, in pre-processing 400, other physical information required for CFD simulation and optimization is estimated by setting operating parameters 408 including boundary conditions of the air conditioner. This includes the fluid behavior and characteristics at all boundary surfaces of the specified room. The nominal conditions of the air conditioner can be used as the initial values of the air conditioner input in the form of boundary conditions. The boundary conditions of the field (such as velocity, pressure) specify the value of the function itself, or the value of the normal derivative of the function, or the form of the curve or surface that assigns the normal derivative and the variable itself, or the relationship between the value of the function and the derivative of the function in a given area. The boundary conditions of the velocity of solid surfaces such as walls are set to zero. The boundary conditions of temperature are based on the heat transfer between the room and its outside. This heat transfer depends on the thermophysical properties of the room walls, such as the conductivity of the various layers used in the building envelope, the thickness of each layer, and the outside air temperature. The thermophysical properties (such as density or thermal diffusivity) of the fluid in the room can be selected based on an appropriate thermal state model.

[0081] In addition, CFD can solve the governing equations 410 for fluid dynamics. Here, the governing equations are the Navier-Stokes equations. CFD solves the Navier-Stokes equations along with conservation of mass and energy. The system of equations based on the Navier-Stokes equations has been shown to represent the mechanical behavior of Newtonian fluids (such as air) and is implemented for flow simulations in environments or rooms. Discretization of the Navier-Stokes equations is a reconstruction of the equations so that they can be applied to computational fluid dynamics. The governing equations for Navier-Stokes with heat transfer are as follows:

[0082]

[0083]

[0084]

[0085] in,

[0086] is the divergence operator,

[0087] is the gradient operator, and

[0088] is the Laplace operator.

[0089] p, V, T are pressure, velocity and temperature distributions respectively.

[0090] Furthermore, equations such as relative humidity ψ can also be combined with the above equations.

[0091] Equations 1a-1c can be expressed as N(p, V, T, ψ) = 0, and computational fluid dynamics (CFD) can be used to solve Equations 1a-1c to obtain solutions.

[0092] In post-processing 402, the thermal state of the target area is obtained based on the solution 410 of Equations 1a-1c obtained by CFD. This thermal state includes the temperature, air velocity, and humidity of the air / fluid in the target area. Furthermore, the obtained thermal state can be used to estimate the target thermal state. Subsequently, the boundary conditions of the airflow dynamics model are modified to achieve the target thermal state distribution in the environment and minimize the energy consumption of the air conditioner.

[0093] Figure 5 Schematic diagram of obtaining the thermal state at the target area based on CFD simulation and optimal air conditioner input according to some embodiments is shown. Air conditioner control input 500 is, for example, inlet temperature and air conditioner flow rate, flow direction angle, humidity, etc. In some embodiments, air conditioner control input 500 is used as boundary conditions for CFD simulation 502. CFD simulation 502 can generate the thermal state at the target area (such as reference Figure 4 described above).

[0094] In addition, the air conditioner input 500 is optimized. The air conditioner input optimization 504 is based on the minimum energy consumption of the HVAC system and the target thermal state distribution. In some embodiments, the air conditioner input optimization 504 is based on the minimum energy consumption and the thermal state of the target area from the CFD simulation 502. Therefore, the optimal air conditioner input is such an input that provides the target thermal state distribution in the target area with the minimum energy consumption of the HVAC system. After the CFD simulation 502 and the air conditioner input optimization 504, the optimal thermal state 506 at the target area is determined. The optimal thermal state at the target area is determined based on the optimal value of the air conditioner input obtained from the optimization 504. In addition, the thermal state 508 of the output air is updated according to the determined optimal thermal state.

[0095] Some embodiments are based on the recognition that, when CFD is used to obtain the thermal state in a target region, a certain cost function needs to be minimized to achieve the target thermal state in the target region with minimal energy consumption.

[0096] Figure 6A schematic diagram is shown of minimizing a cost function to achieve a target thermal state in a target area while improving the energy efficiency of an HVAC system, according to some embodiments. Some embodiments are based on the recognition that there are many different combinations of HVAC system outputs that can result in meeting the target thermal state 600. To this end, some embodiments aim to find not just any suitable output combination, but rather a combination that minimizes the energy consumption of the HVAC system. That is, among all output combinations that can meet the target thermal state 600 in the target area, the present embodiment selects the combination that requires the least energy to achieve. In some embodiments, a cost function optimization is used to solve for the output combination 602. For example, the cost function optimization can be performed iteratively 604 based on the sensitivity of the cost function to the operation of the HVAC system. An exemplary sensitivity-based optimization is the direct adjoint loop (DAL) method.

[0097] Some embodiments are based on the understanding that the cost function can be evaluated based on a well-mixed thermal state model of the room, which assumes that the entire domain consists of a single temperature value. However, this understanding is incorrect and an oversimplification in many cases, as thermal stratification is common in buildings. Buildings are complex, multi-scale, multi-physics, and highly uncertain dynamic systems with a variety of interfering factors. In CFD analysis, the entire building is considered as a comprehensive system, and the airflow dynamics are effectively simulated.

[0098] In particular, some embodiments are based on the recognition that the thermal conditions in the environment affect the target thermal conditions in the target zone and the energy consumption of the air conditioner.

[0099] Figure 7A A schematic diagram of constructing and evaluating a cost function that includes both a target thermal state and the energy consumption of an HVAC system is shown, according to some embodiments. A target thermal state 700 is obtained. In addition, a thermal state 702 based on a CFD simulation is obtained. The cost function evaluation produces a numerical value that represents the degree of match between the CFD simulated thermal state 702 and the target thermal state along the sight line of sight of different beams at various heights. When considering the energy consumption 704 of an air conditioner based on a refrigerant cycle, multiple thermal states can have different energy budgets. For example, in some embodiments, a large fan speed value in the refrigerant cycle can be used to determine the control input of the air conditioner to achieve the target thermal state, which may be suboptimal for the operation of the HVAC system.

[0100] Figure 7BA schematic diagram of a vapor compression cycle, according to some embodiments, is shown as an example of a refrigerant cycle. A refrigeration cycle is a concept and mathematical model for heat pumps and refrigerators. One example of a refrigerant cycle is the vapor compression cycle, which is used in most household refrigerators and many large commercial and industrial refrigeration systems, using a refrigerant as the working fluid. A refrigerant is a substance or mixture, typically a fluid such as a fluorocarbon, that undergoes a phase change from liquid to gas and back to liquid during the vapor compression cycle. Vapor compression utilizes a circulating liquid refrigerant as a medium that absorbs and removes heat from the room to be cooled, then rejects the heat elsewhere. All of these systems have four components: a compressor 708; a condenser 710; a thermal expansion valve (also known as a throttle or metering device) 712; and an evaporator 714. Additionally, a fan 716 circulates the warm air within the enclosed space through coils or pipes carrying a mixture of cold refrigerant liquid and vapor. Therefore, energy consumption is associated with the compressor 708, condenser 710, thermal expansion valve 712, and evaporator 714. Cost function evaluation 706 may also consider this energy consumption in addition to the target thermal state.

[0101] In some embodiments, the multi-objective cost function J is defined as:

[0102] J=∫∫∫∫ Ω (TT comf ) 2 +(VV comf ) 2 +(ψ-ψ comf ) 2 dxdydzdt ..........(2)

[0104] in,

[0105] T comf , V comf , ψ comf Determined from a thermal comfort model that itself uses the thermal states T, V, ψ for training purposes. Here Ω is the region of interest.

[0106] Figure 8AA block diagram is shown of an exemplary optimization method for implementing optimal air conditioner inputs based on a target thermal state with minimal energy, according to some embodiments. Initial values 800 for the air conditioner are based on nominal operating conditions of the refrigerant cycle as boundary conditions for the Navier-Stokes equations with heat transfer. For example, the initial values can be derived from the current operating state of the HVAC system. The initial values 800 for the air conditioner are used to generate a CFD solution 802 to the control equations. Additionally, a CFD solution 804 to the adjoint equations is determined. The CFD solution 804 to the adjoint equations is used to evaluate the sensitivity 806 of the cost function.

[0107] ADM 224 is first performed by deriving the adjoint Navier-Stokes and heat transfer problems at the level of partial differential equations (PDEs) based on the Euler-Lagrange method. The goal is to minimize a cost function subject to the constraints of ADM 224 or any other suitable constraints. Other constraints may include the geometry of the room, the maximum rate that the HVAC system can provide, the minimum or maximum temperature that the HVAC system can provide, etc. When deriving the adjoint equations, a frozen turbulence assumption may be used, which ignores changes in the turbulent field relative to the cost function in the analysis. Second, as with conventional CFD problems, the PDE equations are discretized in the operational domain using various numerical methods (e.g., finite volume, finite element, finite difference, etc.).

[0108] The sensitivity 806 of the cost function is considered to be the gradient of the cost function. In addition, the gradient descent method is used to update the operating parameters 808 of the air conditioner. A convergence criterion 810 is checked. One embodiment of this convergence criterion is the change of the cost function between successive iterations. Another embodiment is the size of the sensitivity or gradient of the cost function relative to the design variables. If the convergence criterion is not reached, the next iteration is started and the CFD solution 802 of the control equation is determined in the iteration. When the thermal state in the target area is equal to or approximately equal to the target thermal state, the convergence criterion is met. When the convergence criterion is met, a final estimation 812 of the air conditioner operating parameters is performed.

[0109] The cost function is relative to any operating parameter ξ i The sensitivity can be expressed as:

[0110]

[0111] The set of operating parameters that need to be estimated is represented by (ξ1,ξ2,...ξ n ) is represented by the optimization method. The optimization method uses the enhanced objective function L, which is

[0112] L=J+∫ Ω (p a ,V a ,T a ,ψa )N(p,V,T,ψ)dΩ.............(4)

[0113] N(p, V, T, ψ) = 0 is the Navier-Stokes equation with heat transfer and mass conservation.

[0114] Considering ξ i The change of L can be expressed as

[0115]

[0116] To determine The accompanying variables are selected to satisfy the following conditions:

[0117]

[0118] Therefore, the direct adjoint loop (DAL) method involves the Lagrange multiplier (V a ,p a ,T a ), which represent the accompanying rate and pressure, so that The air conditioner inputs determined by the optimization are selected as V, T, ψ, namely, inlet velocity, inlet angle, inlet temperature and inlet humidity.

[0119] For example, to determine δJ / δV and δJ / δT, we can set ξ i =V or ξ i =T to use the optimization method. In addition, (p a ,V a ,T a ,ψ a ) are the accompanying pressure, velocity, temperature, and humidity used in step 804. The accompanying variables can be viewed as purely mathematical terms. In some embodiments, the accompanying variables provide or represent the influence of any source term on the function of interest, i.e., the Navier-Stokes equations for heat transfer.

[0120] The adjoint variable can be used to determine the sensitivity of the cost function to any operating parameter

[0121]

[0122] By using a simple steepest descent algorithm, ξ i Can be updated to

[0123]

[0124] where λ is a positive constant representing the step size.

[0125] Figure 8BA schematic diagram of a co-simulation of feedback control of HVAC in a room based on thermal comfort and energy cost functions and optimal control of airflow is shown. This illustrates an embodiment of how a CFD model with adjointly optimized ADM 224 can be synchronized with the dynamics of HVAC 210 and BEM 208. 814 shows an initial guess for solving the HVAC model using the Building Energy Simulation (BES) model. The results of this simulation are saved to disk at regular time intervals (e.g., Δt1). Simultaneously, the ADM model solves the Navier-Stokes equations for heat transfer using the initial guess, and the results are saved to disk 816. At this point, the commutative variables are calculated. For example, the exchangeable variables correspond to the rate and temperature at a given sensor location or outlet using an ADM, or the rate and temperature at a room inlet using an HVAC model coupled to a BES.

[0126] The simultaneous solution of 814 and 816 continues until the first checkpoint t1 is reached. At this point, the adjoint equation 818 is solved backward in time. This adjoint equation requires the CFD forward simulation data u(t i ), where an example of this data is the temperature, velocity, and humidity at each location in the room at different times. Compared to the 814 / 816 coupling, the 816 / 818 approach handles the data exchange between the CFD and adjoint solvers at each simulation time step. The solution of 818 produces adjoint variables, which are used to calculate the cost function with respect to the control variable (called x3(t i )) sensitivity. Such steps can be iterated until The ideal accuracy is achieved, which is the optimal exchange value given by CFD to BES. After exchanging data between CFD and BES, internal optimization / control is implemented in BES, for example using PI feedback control. The co-simulation of HVAC model, BES and ADM and the internal optimization of BES and CFD (adjoint method) are carried out until the final time t n+1 .

[0127] Control of HVAC systems

[0128] Figure 9A schematic diagram of an HVAC system 900 is shown, which includes a controller 910 in communication with the system 200 employing the principles of some embodiments. The controller 910 can be connected to the system 200 via the control interface 226. The controller 910 includes a compressor control 902, an evaporator fan control 904, an expansion valve control 906, and a condenser fan control 908. The system 200 is operatively connected to the controller 910 of the HVAC system either wirelessly or via wires. In addition, the system 900 includes a set of sensors to determine one or a combination of signals indicating measurements of the environment and measurements of the operation of the HVAC system 900.

[0129] Based on embodiments of the present invention, system 200 can determine boundary conditions for HVAC outputs that achieve a target thermal state profile while minimizing energy consumption by HVAC system 900. To this end, processor 204 of system 200 uses an HVAC model to determine target control parameters for actuators of HVAC system 900 that correspond to the target thermal state profile in the environment. Furthermore, processor 204 generates control commands corresponding to the target control parameters. Processor 204 submits the control commands to HVAC system 900. Specifically, the control commands are submitted to controller 910 of HVAC system 900 to operate actuators and / or components of HVAC system 900 according to the target control parameters.

[0130] To this end, in one embodiment, the controller 910 receives a control command from the system 200. In some embodiments, the control command includes an optimal speed for the compressor 708. In this case, the compressor control device 902 changes the current compressor speed to the optimal speed according to the control command. In some embodiments, the control command includes a position of the expansion valve 712. In this case, the expansion valve control device 906 actuates the expansion valve to the position defined by the control command.

[0131] HVAC system 900 includes evaporator fan 716 and condenser fan 912, which are not activated to achieve a target heat profile or target control parameters. Control commands include optimal speeds for evaporator fan 716 and condenser fan 912, respectively, so that fans 716 and 912 consume less energy while achieving the target control parameters. Evaporator fan control 904 controls evaporator fan 716 based on the control command including the optimal speed for evaporator fan 716. Similarly, evaporator fan control 904 controls condenser fan 912 based on the control command including the optimal speed for evaporator fan 716.

[0132] Because the control commands determined by system 200 correspond to target control parameters that result in a target thermal state distribution, when HVAC system 900 is operated according to the control commands, the target thermal state distribution in the environment is achieved. Furthermore, because the actuators and / or components of HVAC system 900 are controlled according to the aforementioned optimal values, the target thermal state distribution is achieved with minimal energy consumption by HVAC system 900. This allows for increased thermal comfort for occupants while simultaneously reducing energy consumption by HVAC system 900.

[0133] Furthermore, in some cases, occupants may change the target thermal profile of the thermal state, the layout of the environment may change over time, and the external temperature may vary seasonally and / or daily. In such cases, system 200 provides control commands that achieve the changed target thermal profile with minimal energy consumption. Thus, system 200 offers a significant advantage in controlling HVAC system 900 based on the dynamic changes in the environment and varying levels of occupant comfort.

[0134] HVAC system design

[0135] In a complex target thermal profile (e.g., a warehouse), different sections are maintained at different thermal conditions. Consider installing a simple HVAC system in the warehouse (e.g., with a single entrance located in a corner of the warehouse). It will be appreciated that such a complex target thermal profile cannot be achieved with such a simple HVAC system. It will also be appreciated that such a complex target thermal profile can be achieved, but at an unreasonable cost to the HVAC system.

[0136] To this end, some embodiments are based on the goal of optimizing the design of an HVAC system for a target thermal state distribution in a target environment. Figure 1 Such an HVAC system is implemented according to the principles explained in . Here, the design variables are optimized. In some embodiments, the design variables include the number of inlets, the location of the inlets at the target environment wall, the size of the air conditioner (AC) diffuser and the number of AC units, the inlet orientation (the direction in which the air is blown), etc. The design variables are optimized so that the optimized design variables generate the target thermal state distribution in an efficient manner. In addition, the HVAC system is installed according to the optimized design variables. In other words, for example, the number of inlets, the location of the inlets and the orientation of the inlets are determined, which enables the target thermal distribution to be achieved in an optimal manner. Subsequently, the determined number of inlets, the location of the inlets and the orientation of the inlets are installed, resulting in the optimal design and structure of the HVAC system.

[0137] Additionally or alternatively, optimizing the design variables includes determining other components of the HVAC system (e.g., the type of compressor, the number of AC units, and the shape of the diffuser) that produce the desired output for the determined inlet. Thus, a refrigeration cycle that produces the target thermal state profile can also be determined.

[0138] To this end, processor 204 is configured to process the environmental data using BEM 208 to estimate the thermal state of the air at the walls of the indoor environment. In some embodiments, BEM 208 is utilized to estimate the thermal state at the inlet location. Furthermore, processor 204 is configured to determine one or more of the aforementioned design variables by minimizing a multi-objective cost function. Processor 204 iteratively minimizes the multi-objective cost function until a termination condition is satisfied.

[0139] Figure 10 A schematic diagram illustrates an iterative process for minimizing a multi-objective cost function for designing an HVAC system, according to some embodiments. The multi-objective cost function is a combination of the operating cost of the HVAC system and the difference between the target thermal state profile and the current thermal profile. The processor 204 determines the sensitivity of the multi-objective cost function 1000 to update the design variables 1002. The design variables are updated in the direction of the sensitivity of the multi-objective cost function.

[0140] Additionally, a current thermal state distribution is determined by the processor 204. In some embodiments, the current thermal state distribution is determined 1004 based on the ADM 1006 having boundary conditions, including the thermal state at the inlet location defined by the BEM 208 and thermal states outside the inlet location resulting from operation of the HVAC system. In some other embodiments, the processor 204 is configured to determine the current distribution of thermal states based on the ADM using updated design variables.

[0141] In one embodiment, processor 204 is further configured to determine the operating cost of the HVAC system 1008 resulting in the updated design variables. Processor 204 iteratively minimizes the multi-objective cost function until a termination condition is satisfied. In some embodiments, the termination condition is satisfied when the sensitivity of the cost function is less than a first threshold. In other embodiments, the termination condition is satisfied when the value of the cost function is less than a second threshold. In yet other embodiments, the termination condition is satisfied when the number of iterations exceeds a third threshold. These thresholds may be predefined.

[0142] Figure 11A schematic diagram of an HVAC system designed for room 1110 according to some embodiments is shown. Occupant 1114 resides in area 1116 of room 1110. Area 1116 is referred to as an occupied area. System 200 is associated with the HVAC system and receives a desired thermal state for occupant 1114. Furthermore, since no occupant resides in area 1112, area 1112 of room 1110 is referred to as an unoccupied area. System 200 is configured to design the HVAC system so that only the occupied area of room 1110 is conditioned (cooled or heated) according to the desired thermal state. For example, the location and angle of vents can be considered as design variables, and the optimal location and angle of vents 1118 can be determined. Vents 1118 only regulate occupied area 1116. Because the HVAC system is designed so that only the occupied area of room 1110 is conditioned, rather than regulating the entire room 1110, the energy consumption of the HVAC system is also optimized. Thus, exemplary embodiments facilitate the design of an efficient HVAC system.

[0143] The following description provides only exemplary embodiments and is not intended to limit the scope, applicability, or configuration of the present disclosure. Instead, the following description of exemplary embodiments will provide those skilled in the art with an enabling description for implementing one or more exemplary embodiments. Various changes that can be made to the function and arrangement of elements are contemplated without departing from the spirit and scope of the subject matter disclosed in the appended claims.

[0144] Specific details are given in the following description to provide a thorough understanding of the embodiments. However, it will be understood by those skilled in the art that the embodiments may be practiced without these specific details. For example, the systems, processes, and other elements of the disclosed subject matter may be shown as components in block diagram form so as not to obscure the embodiments in unnecessary detail. In other cases, well-known processes, structures, and techniques may be shown without unnecessary detail to avoid obscuring the embodiments. In addition, the same reference numerals and numbers in the various figures represent the same elements.

[0145] Moreover, each embodiment can be described as a process depicted as a flowchart, flow chart, data flow diagram, structure diagram or block diagram. Although a flowchart can describe an operation as a sequential process, many operations can be performed in parallel or simultaneously. In addition, the order of the operations can be rearranged. When the operation of the process is completed, it may terminate, but there may be other steps that are not discussed or included in the figure. In addition, not all operations in any particularly described process will occur in all embodiments. The process can correspond to a method, function, procedure, subroutine, subprogram, etc. When the process corresponds to a function, the termination of the function can correspond to the function returning to the calling function or main function.

[0146] Furthermore, embodiments of the disclosed subject matter may be implemented, at least in part, manually or automatically. Manual or automatic implementation may be performed or at least assisted by the use of a machine, hardware, software, firmware, middleware, microcode, hardware description language, or any combination thereof. When implemented in software, firmware, middleware, or microcode, program code or code segments that perform the necessary tasks may be stored in a machine-readable medium. A processor may perform the necessary tasks.

[0147] The various methods or processes outlined herein can be encoded as software that can be executed on one or more processors using any of a variety of operating systems or platforms. Furthermore, such software can be written using any of a number of suitable programming languages and / or programming or scripting tools, and can also be compiled into executable machine language code or intermediate code that is executed on a framework or virtual machine. Generally, in various embodiments, the functionality of the program modules can be combined or distributed as desired.

[0148] Embodiments of the present disclosure may be implemented as methods for which examples have been provided. The activities performed as part of the method may be ordered in any suitable manner. Thus, embodiments may be constructed in which activities are performed in an order different from that illustrated, which may include performing some activities simultaneously even though they are shown as being performed sequentially in an illustrative embodiment.

[0149] Although the present disclosure has been described with reference to certain preferred embodiments, it will be understood that various other changes and modifications may be made within the spirit and scope of the present disclosure. It is therefore intended that the appended claims encompass all such changes and modifications as fall within the true spirit and scope of the present disclosure.

Claims

1. A control system for controlling the operation of a heating, ventilation, and air conditioning (HVAC) system configured to condition an indoor environment by pushing air into the environment via a set of inlets disposed at a set of locations on one or more walls of the environment, wherein: The thermal condition of the air pushed to the environment at the inlet location includes one or a combination of the temperature, velocity and humidity of the air, the control system comprising: an input interface configured to accept data indicating a target thermal condition distribution in the environment, wherein the thermal condition at a location in the environment comprises one or a combination of temperature, velocity, and humidity of air; a memory configured to store an airflow dynamics model ADM and an HVAC model defining the dynamics of the HVAC system, the ADM defining a thermal state distribution in the environment subject to boundary conditions of the thermal state of the air at the walls of the environment; A processor configured to: inverting the ADM to estimate boundary condition values at the inlet location that define a target thermal state at the inlet location, the boundary condition values at the inlet location resulting in the target thermal state distribution in the environment; using the HVAC model, determining target control parameters of actuators of the HVAC system that result in the target thermal condition at the inlet location; and submitting control commands to the HVAC system to operate actuators of the HVAC system according to the control parameters, wherein different combinations of target thermal state values at the inlet locations result in target thermal state distributions in the environment, and wherein the processor selects a thermal state combination based on a performance metric of the HVAC system, wherein the performance metric is defined by a multi-objective cost function that is a combination of an operating cost of the HVAC system and a difference between a target thermal profile and a corresponding current thermal profile, such that the processor is configured to determine the target thermal state by minimizing the multi-objective cost function, wherein the processor iteratively minimizes the multi-objective cost function until a termination condition is satisfied, wherein, to perform the iteration, the processor is configured to: determining a sensitivity of the cost function to boundary conditions for updating the entry position; updating the boundary conditions at the inlet location in the direction of the sensitivity; determining a current thermal state distribution according to the ADM using the updated boundary conditions; and An operating cost of the HVAC system resulting in updated boundary conditions at the inlet location is determined.

2. The control system according to claim 1, wherein: When the sensitivity of the multi-objective cost function is less than a first threshold, the value of the cost function is less than a second threshold, or the number of iterations is greater than a third threshold, the termination condition is met.

3. The control system according to claim 1, wherein: The memory further stores a building envelope model (BEM) defining boundary conditions for a thermal state of air at a wall of the environment when the environment is not conditioned by the HVAC system, and wherein the processor initializes the boundary conditions by submitting thermal state values outside the environment to the BEM.

4. The control system according to claim 1, wherein: The target thermal state distribution is non-uniform, having at least two different thermal state values at two different locations in the environment.

5. The control system according to claim 1, wherein: The target thermal condition profile is provided for a portion of the environment, and wherein the HVAC system conditions the environment to result in a non-uniform thermal profile having at least two different thermal condition values at two different locations in the environment.

6. The control system according to claim 1, wherein: The ADM represents the dynamics of air in the environment using Navier-Stokes equations and an energy equation, wherein computational fluid dynamics (CFD) calculations solve the Navier-Stokes equations and the energy equation to estimate the thermal state distribution.

7. A method for controlling the operation of a heating, ventilation, and air conditioning (HVAC) system configured to condition an indoor environment by pushing air into the environment via a set of inlets disposed at a set of locations on one or more walls of the environment, wherein: The thermal condition of the air pushed into the environment at the inlet location includes one or a combination of the temperature, velocity, and humidity of the air, wherein the method uses a processor coupled to a memory storing an airflow dynamics model ADM and an HVAC model defining the dynamics of the HVAC system, the ADM defining a thermal condition distribution in the environment, the thermal condition distribution being affected by boundary conditions of the thermal condition of the air at the walls of the environment, the processor being coupled to stored instructions when executed by the processor to implement the steps of the method, the steps comprising: receiving data indicating a target thermal condition distribution in the environment, wherein the thermal condition at a location in the environment comprises one or a combination of temperature, velocity, and humidity of air; inverting the ADM to estimate boundary condition values at the inlet location that define a target thermal state at the inlet location, the boundary condition values at the inlet location resulting in the target thermal state distribution in the environment; using the HVAC model, determining target control parameters of actuators of the HVAC system that result in the target thermal condition at the inlet location; and submitting control commands to the HVAC system to operate actuators of the HVAC system according to the control parameters, wherein different combinations of target thermal state values at the inlet locations result in a target thermal state distribution in the environment, and wherein the method further comprises selecting a thermal state combination based on a performance indicator of the HVAC system, wherein the performance indicator is defined by a multi-objective cost function of a combination of an operating cost of the HVAC system and a difference between a target thermal profile and a corresponding current thermal profile, such that the processor is configured to determine the target thermal state by minimizing the multi-objective cost function, The method further comprises iteratively minimizing the multi-objective cost function until a termination condition is satisfied, wherein, to perform the iteration, the iterative minimization comprises: determining a sensitivity of the cost function to boundary conditions for updating the entry position; updating the boundary conditions at the inlet location in the direction of the sensitivity; determining a current thermal state distribution according to the ADM using the updated boundary conditions; and An operating cost of the HVAC system resulting in updated boundary conditions at the inlet location is determined.

8. The method according to claim 7, wherein: When the sensitivity of the multi-objective cost function is less than a first threshold, the value of the cost function is less than a second threshold, or the number of iterations is greater than a third threshold, the termination condition is met.

9. The method according to claim 7, wherein: The memory further stores a building envelope model (BEM) defining boundary conditions for a thermal state of air at a wall of the environment when the environment is not conditioned by the HVAC system, and wherein the processor initializes the boundary conditions by submitting thermal state values outside the environment to the BEM.

10. The method according to claim 7, wherein: The target thermal state distribution is non-uniform, having at least two different thermal state values at two different locations in the environment.

11. The method according to claim 7, wherein: The target thermal condition profile is provided for a portion of the environment, and wherein the HVAC system conditions the environment to result in a non-uniform thermal profile having at least two different thermal condition values at two different locations in the environment.

Images