Net zero energy facility with uncertainty handling

Abstract

There is provided a method for achieving a net energy target for building operation for a time period including a first sub-period prior to a current time and a second sub-period from the current time to an end of the time period. The method includes generating a first predicted range of energy consumption for a plurality of time steps in the second sub-cycle, a second predicted range of energy production for the plurality of time steps in the second sub-cycle is generated and a third predicted range of net energy amount for the plurality of time steps in the second sub-cycle is generated. The net energy amount is based on a difference between the energy consumption amount and the energy production amount. The method further includes providing a policy for the building operation based on the third predicted range and the net energy target.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims priority to and the benefit of U.S. Provisional Patent Application No. 63 / 436,008, filed on December 29, 2022, the entire disclosure of which is incorporated herein by reference. Background Art

[0003] The present disclosure generally relates to building management systems (BMS). A BMS is generally a system of devices configured to control, monitor, and manage equipment in or around a building or building area. A BMS may include, for example, an HVAC system, a security system, a lighting system, a fire alarm system, any other system capable of managing building functions or equipment, or any combination thereof. In some scenarios, a BMS is associated with a source of green energy, such as a photovoltaic energy system, which provides energy to other equipment and devices associated with the BMS. Summary of the invention

[0004] One embodiment of the present disclosure is a method for achieving a net energy target for building operations over a time period, the time period including a first sub-period before a current time and a second sub-period from the current time to the end of the time period. The method includes generating a first forecast range of energy consumption for a plurality of time steps in the second sub-period, generating a second forecast range of energy production for a plurality of time steps in the second sub-period, and generating a third forecast range of net energy for a plurality of time steps in the second sub-period. The net energy is based on the difference between the energy consumption and the energy production. The method also includes providing a strategy (e.g., a control strategy) based on the third forecast range and the net energy target.

[0005] In some embodiments, the method further includes providing a graphical user interface, the graphical user interface including a net energy map. The net energy map includes a first line showing actual net energy in a first sub-period, a second line showing planned net energy in a second sub-period, and an area of ​​a third forecast range based on the second sub-period.

[0006] In some embodiments, the method includes fitting a mean model to the historical energy consumption data, fitting a deviation model to the error in the output of the mean model, and converting the combination of the mean model and the deviation model into a Gaussian model. Generating a first forecast horizon of energy consumption for a plurality of time steps in the time period is performed using the Gaussian model.

[0007] In some embodiments, the method includes fitting a mean model to the historical energy production data, fitting a deviation model to the error in the output of the mean model, and converting the combination of the mean model and the deviation model into a Gaussian model. Generating a second forecast horizon of energy production at multiple time steps in the time period can be performed using the Gaussian model.

[0008] In some embodiments, the first prediction horizon and the second prediction horizon are associated with confidence intervals of a prediction model for predicting energy consumption and energy production. In some embodiments, providing a control strategy may include generating a net energy trajectory including a net energy target for a plurality of time steps based on the first prediction horizon and the second prediction horizon. Each net energy target indicates a target difference between cumulative energy consumption and cumulative energy production or an offset from the start of a time period to a corresponding time step in a plurality of time steps. Providing a strategy may also include, for a given time step, generating a set of reduction actions predicted to achieve a net consumption target for a given time step and implementing the set of reduction actions. Generating a net energy trajectory may include performing an optimization constrained by a net energy target.

[0009] In some embodiments, providing the strategy includes controlling building equipment serving the facility, the energy consumption corresponds at least in part to operation of the building equipment, and the energy production corresponds to green energy production at the facility.

[0010] In some embodiments, the method further comprises decomposing the first forecast horizon of energy consumption into energy consumption categories, wherein the energy consumption categories include heating and cooling. The strategy can be configured to drive the net energy amount at a final time step in the second sub-period to a net energy target.

[0011] In some embodiments, the net energy amount at a given time step in the plurality of time steps is the cumulative difference between the energy consumption amount and the energy production amount during the time period leading up to the given time step.

[0012] Another embodiment of the present disclosure is a system. The system includes an energy load operable to consume energy, a green energy source configured to generate energy, and a processing circuit. The processing circuit is programmed to generate a first prediction range of energy consumption of the energy load at multiple time steps in a second sub-period, a second prediction range of energy production of the green energy source at multiple time steps in the second sub-period, and a third prediction range of net energy at multiple time steps in the second sub-period. The net energy amount is based on the difference between the energy consumption amount and the energy production amount. The processing circuit is also programmed to control the energy load using a control strategy configured to drive one or more of the net energy amounts to a target.

[0013] In some embodiments, the processing circuit is also programmed to host a graphical user interface that includes a net energy graph that includes a first line showing actual net energy within a first sub-period, a second line showing planned net energy within a second sub-period, and an area of ​​a third forecast range based on the second sub-period.

[0014] In some embodiments, the processing circuit is further programmed to fit a mean model to the historical energy consumption data, fit a deviation model to the error in the output of the mean model, and convert the combination of the mean model and the deviation model into a Gaussian model. The processing circuit can be programmed to use the Gaussian model to generate a first prediction range of energy consumption at multiple time steps in the time period.

[0015] In some embodiments, the processing circuit is programmed to execute the control strategy by generating a net consumption trajectory including a plurality of net consumption targets for time steps based on the first prediction horizon and the second prediction horizon, wherein each net consumption target indicates a target difference or offset between total consumption and total production from the beginning of the time period to the corresponding time step. Executing the control strategy may also include generating, by the processing circuit and for a given time step in the plurality of time steps, a set of reduction actions predicted to achieve the net consumption target for the sub-period and implementing the set of reduction actions by controlling the energy load.

[0016] In some embodiments, the processing circuit is programmed to generate the net consumption trajectory by performing an optimization subject to a goal constraint. In some embodiments, the processing circuit is programmed to generate the set of reduction actions based on a decomposition of energy usage types of the energy load, the energy usage types including heating and cooling.

[0017] Another embodiment of the present disclosure is one or more non-transitory computer-readable media storing program instructions that, when executed by one or more processors, cause the one or more processors to perform operations. The operations include generating a first forecast range of energy consumption or carbon emissions for multiple time steps in a time period, generating a second forecast range of energy production or carbon capture for multiple time steps in a time period, and providing a control strategy based on the first forecast range of energy consumption or carbon emissions and the second forecast of energy production or carbon capture, the control strategy being configured to drive the cumulative difference between energy production or carbon capture and energy consumption or carbon emissions in the time period to a target.

[0018] In some embodiments, the operations further include fitting a mean model to the historical energy consumption or carbon emission data, fitting a deviation model to the error in the output of the mean model, and converting the combination of the mean model and the deviation model to a Gaussian model. Generating a first prediction horizon for a plurality of time steps in the time period is performed using the Gaussian model. In some embodiments, providing the control strategy includes implementing a reduction action based on the target.

[0019] Another embodiment of the present disclosure is a method. The method includes providing a net consumption trajectory, the net consumption trajectory including net consumption targets for one or more sub-periods of a time period. Each net consumption target indicates a target difference between total consumption and total production or offset from the beginning of a time period to the end of the sub-period. The method also includes, for a sub-period in the plurality of sub-periods, generating a set of reduction actions predicted to achieve the net consumption target for the sub-period. The method also includes implementing the set of reduction actions.

[0020] In some embodiments, the net consumption target is a net energy target that indicates a target difference between energy consumption and total energy production. A sub-period may begin after the start of a time period. In some embodiments, the total energy consumption corresponds to energy used by the facility, and the total energy production corresponds to green energy produced at the facility.

[0021] In some embodiments, the net consumption target is a net carbon target indicating a target difference between total carbon emissions and total carbon capture from the beginning of the time period to the end of the sub-period. The method may also include providing a user dashboard including a visualization of the net consumption trajectory and reduction actions, and updating the user dashboard as multiple sub-periods pass.

[0022] In some embodiments, the forecasting cost takes into account carbon emissions associated with implementing the reduction action. In some embodiments, providing the net consumption trajectory includes generating the net consumption trajectory as an output of a forecasting optimization constrained so that the net consumption trajectory has a target value at the end of the time period. Implementing the reduction action may include deploying a device of the building equipment specified by the reduction action. Implementing the reduction action may also include controlling the equipment to perform the reduction action.

[0023] In some embodiments, the device includes lighting devices and HVAC devices. The set of reduction actions includes lighting level changes and temperature set point changes. Providing the net consumption trajectory includes performing a predictive optimization at the beginning of the time period. Generating the set of reduction actions can be performed at the beginning of a sub-period.

[0024] In some embodiments, the total consumption includes a first portion associated with building equipment and a second portion associated with transportation vehicles. The method may also include using a neural network to predict an amount of energy production for a time period, wherein the net consumption trajectory is based on the amount of energy production for the time period. In some embodiments, generating the set of reduction actions includes using a digital twin of the facility, and wherein implementing the set of reduction actions includes operating equipment represented in the digital twin. In some embodiments, generating the set of reduction actions includes allocating portions of the total reduction amount across multiple types of equipment.

[0025] Another embodiment of the present disclosure is a system. The system includes: an energy load, which is operable to consume energy; a green energy source, which is configured to produce energy; and a processing circuit, which is programmed to provide a net energy trajectory, which includes a net energy target for one or more sub-periods of a time period. The net energy target indicates the difference between the energy consumed by the energy load and the energy produced by the green energy source. The processing circuit can also be programmed to generate a reduction action predicted to achieve the net energy target for one of the one or more sub-periods. The processing circuit can also be programmed to implement the reduction action by affecting the energy load.

[0026] In some embodiments, the net energy target for a sub-period indicates a target difference between cumulative energy consumption by energy loads and cumulative energy production by green energy sources from the start of the time period to the end of the sub-period. The sub-period may begin after the start of the time period. In some embodiments, the system is configured to generate reduction actions by performing a predictive optimization that uses cost forecasts for possible reduction actions and energy reduction forecasts for possible reduction actions.

[0027] In some embodiments, the processing circuit is programmed to perform a first predictive optimization and a second predictive optimization, the net energy trajectory is an output of the first predictive optimization and an input to the second predictive optimization, and the curtailment action is an output of the second predictive optimization. The first predictive optimization may be constrained so that the net energy trajectory has a zero value at the end of the time period. In some embodiments, the first predictive optimization is performed at the beginning of the time period, and the second predictive optimization is performed at the beginning of a sub-period, the beginning of the sub-period being after the beginning of the time period.

[0028] In some embodiments, wherein the energy load comprises HVAC equipment and the curtailment action comprises a set point change. In some embodiments, the processing circuit is programmed to generate a set of curtailment actions comprising using a digital twin of the facility, and wherein implementing the set of curtailment actions comprises operating equipment represented in the digital twin. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 is a diagram of a building equipped with an HVAC system according to some embodiments.

[0030] Figure 2 According to some embodiments, it can be used to serve Figure 1 Block diagram of the water side system of a building.

[0031] Figure 3 According to some embodiments, it can be used to serve Figure 1 Block diagram of the building's air side system.

[0032] Figure 4 According to some embodiments, it can be used to monitor and control Figure 1 Block diagram of a building management system (BMS) for a building.

[0033] Figure 5 According to some embodiments, it can be used to monitor and control Figure 1 Block diagram of another BMS for a building.

[0034] Figure 6 is a block diagram of a net zero energy facility according to some embodiments.

[0035] Figure 7 is a flow chart of a process for achieving net zero energy, according to some embodiments.

[0036] Figure 8 According to some embodiments Figure 7 A set of graphs showing process-related energy consumption and production data.

[0037] Fig. 9 is a block diagram of a system manager for a net zero energy facility, according to some embodiments.

[0038] Fig.10 is a flow chart of another process for achieving net zero energy, according to some embodiments.

[0039] Fig.11 According to some embodiments Fig.10 A set of diagrams related to the process of

[0040] Fig.12 is a flow diagram of a process related to energy consumption and / or production forecasting, for example, for achieving net-zero energy, according to some embodiments.

[0041] Fig.13 is a graph of energy consumption, energy production, and net energy forecasts according to some embodiments.

[0042] Fig.14 is a view in a graphical user interface according to some embodiments.

[0043] Fig.15 is another view in a graphical user interface according to some embodiments.

[0044] Fig.16 is yet another view in a graphical user interface according to some embodiments.

[0045] Fig.17 is a flow chart of a process for achieving a net energy goal according to some embodiments. DETAILED DESCRIPTION

[0046] Overview

[0047] Referring generally to the accompanying drawings, net zero energy facilities and systems, devices, and methods related thereto are shown in various embodiments. The innovations herein relate to providing facilities (buildings, campuses, etc.) that consume zero (or less) net total energy over a period of time (e.g., more than a year, more than a quarter, more than a month, etc.). Net zero total energy consumption may be advantageous for environmental reasons (e.g., to reduce carbon emissions and other pollution) and in order to comply with associated regulations and respond to various stakeholders. In some aspects, the innovations herein provide technical solutions to problems associated with climate change and environmental constraints on future facility projects.

[0048] A net zero energy facility is a facility that uses no more energy than the facility produces in a time period. Such facilities may include on-site energy sources, such as green energy sources, such as photovoltaic systems for collecting solar energy, one or more windmills for collecting wind energy, geothermal energy systems for converting geothermal activity into electricity, and the like. Such green energy sources may be carbon-free, pollution-free, renewable, and the like, and may be installed at the facility. Energy from such sources may be used by various energy loads of the facility, including HVAC devices, lighting devices, appliances, computing equipment (e.g., data centers), and the like. Net energy is the difference between energy production from the facility's energy sources and the facility's energy loads.

[0049] The facility can be connected to an energy grid that provides energy to or receives energy from the facility to address any excess energy or demand at a given time, such as when energy production and consumption are asynchronous. However, even where the facility uses grid energy, energy can be provided back to the grid, and there may be a technical goal to achieve net zero consumption from the grid over a time period. In such a scenario, a facility manager may be able to report net zero energy over a time period, even if grid energy is used at certain times (e.g., during times when renewable energy production is low). However, there are challenges in optimizing control facilities to adapt to changing conditions so as to achieve a net zero energy state over a desired time period. The innovations detailed below address such challenges.

[0050] Building HVAC systems and building management systems

[0051] Reference now Figures 1 to 5 , according to some embodiments, several building management systems (BMS) and HVAC systems are shown in which the systems and methods of the present disclosure may be implemented. In a brief overview, Figure 1 A building 10 equipped with an HVAC system 100 is shown. Figure 2 is a block diagram of a waterside system 200 that may be used to service a building 10 . Figure 3is a block diagram of an airside system 300 that may be used to serve a building 10 . Figure 4 is a block diagram of a BMS that can be used to monitor and control a building 10 . Figure 5 is a block diagram of another BMS that may be used to monitor and control the building 10 .

[0052] Buildings and HVAC Systems

[0053] Specific reference Figure 1 , showing a perspective view of a building 10. The building 10 is served by a BMS. A BMS is generally a system of devices configured to control, monitor, and manage equipment in or around a building or building area. The BMS may include, for example, an HVAC system, a security system, a lighting system, a fire alarm system, any other system capable of managing building functions or devices, or any combination thereof.

[0054] The BMS serving the building 10 includes an HVAC system 100. The HVAC system 100 may include a plurality of HVAC devices (e.g., heaters, chillers, air handling units, pumps, fans, thermal energy storage devices, etc.) configured to provide heating, cooling, ventilation, or other services to the building 10. For example, the HVAC system 100 is shown to include a water side system 120 and an air side system 130. The water side system 120 may provide a heated or cooled fluid to the air handling units of the air side system 130. The air side system 130 may use the heated or cooled fluid to heat or cool an air flow provided to the building 10. Reference Figures 2 to 3 Exemplary waterside and airside systems that may be used in HVAC system 100 are described in greater detail.

[0055] HVAC system 100 is shown to include chiller 102, boiler 104, and rooftop air handling unit (AHU) 106. Waterside system 120 can use boiler 104 and chiller 102 to heat or cool a working fluid (e.g., water, glycol, etc.) and can circulate the working fluid to AHU 106. In various embodiments, HVAC devices of waterside system 120 can be located in or around building 10 (e.g., Figure 1102 ), or located at an off-site location such as a central plant (e.g., a chiller plant, a steam plant, a heat generating plant, etc.). Depending on whether heating or cooling is required in the building 10, the working fluid can be heated in the boiler 104 or cooled in the chiller 102. The boiler 104 can add heat to the circulating fluid, for example, by burning a combustible material (e.g., natural gas) or using an electric heating element. The chiller 102 can place the circulating fluid in a heat exchange relationship with another fluid (e.g., a refrigerant) in a heat exchanger (e.g., an evaporator) to absorb heat from the circulating fluid. The working fluid from the chiller 102 and / or the boiler 104 can be delivered to the AHU 106 via a conduit 108.

[0056] The AHU 106 can place the working fluid in heat exchange relationship with an airflow passing through the AHU 106 (e.g., via one or more stages of cooling coils and / or heating coils). For example, the airflow can be outdoor air, return air from within the building 10, or a combination of both. The AHU 106 can transfer heat between the airflow and the working fluid to provide heating or cooling to the airflow. For example, the AHU 106 can include one or more fans or blowers configured to pass the airflow through or through a heat exchanger containing the working fluid. The working fluid can then be returned to the chiller 102 or the boiler 104 via the conduit 110.

[0057] The air side system 130 may deliver the airflow supplied by the AHU 106 (i.e., supply airflow) to the building 10 via the supply duct 112, and may provide return air from the building 10 to the AHU 106 via the return duct 114. In some embodiments, the air side system 130 includes a plurality of variable air volume (VAV) units 116. For example, the air side system 130 is shown as including a separate VAV unit 116 on each floor or zone of the building 10. The VAV unit 116 may include a damper or other flow control element that may operate to control the amount of supply airflow provided to each zone of the building 10. In other embodiments, the air side system 130 (e.g., via the supply duct 112) delivers the supply airflow to one or more zones of the building 10 without using an intermediate VAV unit 116 or other flow control element. The AHU 106 may include various sensors (e.g., temperature sensors, pressure sensors, etc.) configured to measure properties of the supply airflow. The AHU 106 may receive input from sensors located within the AHU 106 and / or within the building zone, and may adjust the flow rate, temperature, or other properties of the supply air flow through the AHU 106 to achieve a set point condition for the building zone.

[0058] Water side system

[0059] Reference now Figure 2, a block diagram of a waterside system 200 is shown, according to some embodiments. In various embodiments, the waterside system 200 can supplement or replace the waterside system 120 in the HVAC system 100, or can be implemented separately from the HVAC system 100. When implemented in the HVAC system 100, the waterside system 200 can include a subset of the HVAC devices in the HVAC system 100 (e.g., boiler 104, chiller 102, pumps, valves, etc.), and can be operated to supply heated or cooled fluid to the AHU 106. The HVAC devices of the waterside system 200 can be located within the building 10 (e.g., as components of the waterside system 120), or located at an off-site location such as a central facility.

[0060] exist Figure 2 , the water side system 200 is shown as a central device having a plurality of sub-devices 202-212. The sub-devices 202-212 are shown to include a heater sub-device 202, a heat recovery chiller sub-device 204, a chiller sub-device 206, a cooling tower sub-device 208, a high temperature thermal energy storage (TES) sub-device 210, and a low temperature thermal energy storage (TES) sub-device 212. The sub-devices 202-212 consume resources (e.g., water, natural gas, electricity, etc.) of the utility to serve the thermal energy loads (e.g., hot water, cold water, heating, cooling, etc.) of the building or campus. For example, the heater sub-device 202 can be configured to heat water in a hot water loop 214 that circulates hot water between the heater sub-device 202 and the building 10. The chiller sub-device 206 can be configured to cool water in a cold water loop 216 that circulates cold water between the chiller sub-device 206 and the building 10. The heat recovery chiller sub-device 204 can be configured to transfer heat from the cold water loop 216 to the hot water loop 214 to provide additional heating for the hot water and additional cooling for the cold water. The condenser water loop 218 can absorb heat from the cold water in the chiller sub-device 206 and reject the heat absorbed in the cooling tower sub-device 208 or transfer the absorbed heat to the hot water loop 214. The high temperature TES sub-device 210 and the low temperature TES sub-device 212 can store high temperature thermal energy and low temperature thermal energy, respectively, for subsequent use.

[0061] The hot water loop 214 and the cold water loop 216 can deliver heated and / or cooled water to an air handler (e.g., AHU 106) located on the roof of the building 10 or to various floors or zones (e.g., VAV units 116) of the building 10. The air handler pushes the air through a heat exchanger (e.g., a heating coil or a cooling coil) through which the water provides heating or cooling to the air. The heated or cooled air can be delivered to various zones of the building 10 to serve the thermal energy load of the building 10. The water is then returned to the sub-equipment 202-212 for further heating or cooling.

[0062] Although the sub-devices 202-212 are shown and described as heating and cooling water for circulation to the building, it should be understood that any other type of working fluid (e.g., glycol, CO2, etc.) may be used in place of or in addition to water to serve the thermal energy load. In other embodiments, the sub-devices 202-212 may provide heating and / or cooling directly to the building or campus without the need for an intermediate heat transfer fluid. These and other variations of the waterside system 200 are within the teachings of the present disclosure.

[0063] Each of the sub-devices 202-212 can include a variety of devices configured to facilitate the functions of the sub-device. For example, the heater sub-device 202 is shown as including a plurality of heating elements 220 (e.g., boilers, electric heaters, etc.) configured to add heat to the hot water in the hot water loop 214. The heater sub-device 202 is also shown as including a plurality of pumps 222 and 224 configured to circulate the hot water in the hot water loop 214 and control the flow rate of the hot water through each of the heating elements 220. The chiller sub-device 206 is shown as including a plurality of chillers 232 configured to remove heat from the cold water in the cold water loop 216. The chiller sub-device 206 is also shown as including a plurality of pumps 234 and 236 configured to circulate the cold water in the cold water loop 216 and control the flow rate of the cold water through each of the chillers 232.

[0064] The heat recovery chiller sub-equipment 204 is shown to include a plurality of heat recovery heat exchangers 226 (e.g., refrigeration circuits) configured to transfer heat from the chilled water loop 216 to the hot water loop 214. The heat recovery chiller sub-equipment 204 is also shown to include a plurality of pumps 228 and 230 configured to circulate hot and / or chilled water through the heat recovery heat exchangers 226 and to control the flow rate of water through each heat recovery heat exchanger 226. The cooling tower sub-equipment 208 is shown to include a plurality of cooling towers 238 configured to remove heat from condensed water in the condensed water loop 218. The cooling tower sub-equipment 208 is also shown to include a plurality of pumps 240 configured to circulate condensed water in the condensed water loop 218 and to control the flow rate of condensed water through each cooling tower 238.

[0065] The high temperature TES sub-device 210 is shown to include a high temperature TES tank 242 configured to store hot water for subsequent use. The high temperature TES sub-device 210 may also include one or more pumps or valves configured to control the flow rate of hot water into or out of the high temperature TES tank 242. The low temperature TES sub-device 212 is shown to include a low temperature TES tank 244 configured to store cold water for subsequent use. The low temperature TES sub-device 212 may also include one or more pumps or valves configured to control the flow rate of cold water into or out of the low temperature TES tank 244.

[0066] In some embodiments, one or more of the pumps in the waterside system 200 (e.g., pumps 222, 224, 228, 230, 234, 236, and / or 240) or pipes in the waterside system 200 include isolation valves associated therewith. The isolation valves may be integrated with the pumps or positioned upstream or downstream of the pumps to control fluid flow in the waterside system 200. In various embodiments, the waterside system 200 may include more, fewer, or different types of devices and / or sub-equipment based on the specific configuration of the waterside system 200 and the type of loads served by the waterside system 200.

[0067] Air side system

[0068] Reference now Figure 3 , a block diagram of an air side system 300 is shown, according to some embodiments. In various embodiments, the air side system 300 can supplement or replace the air side system 130 in the HVAC system 100, or can be implemented separately from the HVAC system 100. When implemented in the HVAC system 100, the air side system 300 can include a subset of the HVAC devices in the HVAC system 100 (e.g., AHU 106, VAV unit 116, ducts 112-114, fans, dampers, etc.), and can be located in or around the building 10. The air side system 300 can operate to heat or cool the airflow provided to the building 10 using the heated or cooled fluid provided by the water side system 200.

[0069] exist Figure 3 , the air side system 300 is shown to include an economizer type air handling unit (AHU) 302. The economizer type AHU can change the amount of outdoor air and return air used by the air handling unit for heating or cooling. For example, the AHU 302 can receive return air 304 from a building area 306 via a return air duct 308, and can deliver supply air 310 to the building area 306 via a supply air duct 312. In some embodiments, the AHU 302 is a rooftop unit (e.g., such as a rooftop unit) located on the roof of the building 10 or otherwise positioned to receive the return air 304 and the outdoor air 314. Figure 1 106 shown). The AHU 302 can be configured to operate an exhaust damper 316, a blend damper 318, and an outdoor air damper 320 to control the amount of outdoor air 314 and return air 304 that combine to form the supply air 310. Any return air 304 that does not pass through the blend damper 318 can be exhausted from the AHU 302 through the exhaust damper 316 as exhaust air 322.

[0070] Each of the dampers 316-320 may be operated by an actuator. For example, the exhaust damper 316 may be operated by an actuator 324, the blend damper 318 may be operated by an actuator 326, and the outdoor air damper 320 may be operated by an actuator 328. The actuators 324-328 may communicate with the AHU controller 330 via a communication link 332. The actuators 324-328 may receive control signals from the AHU controller 330 and may provide feedback signals to the AHU controller 330. The feedback signals may include, for example, an indication of a current actuator or damper position, an amount of torque or force applied by the actuator, diagnostic information (e.g., results of diagnostic tests performed by the actuators 324-328), status information, debug information, configuration settings, calibration data, and / or other types of information or data that may be collected, stored, or used by the actuators 324-328. The AHU controller 330 may be an economizer controller configured to control the actuators 324-328 using one or more control algorithms (e.g., a state-based algorithm, an extremum seeking control (ESC) algorithm, a proportional integral (PI) control algorithm, a proportional integral derivative (PID) control algorithm, a model predictive control (MPC) algorithm, a feedback control algorithm, etc.).

[0071] Still refer to Figure 3 , the AHU 302 is shown to include a cooling coil 334, a heating coil 336, and a fan 338 positioned within the supply air duct 312. The fan 338 can be configured to force the supply air 310 through the cooling coil 334 and / or the heating coil 336 and provide the supply air 310 to the building zone 306. The AHU controller 330 can communicate with the fan 338 via a communication link 340 to control the flow rate of the supply air 310. In some embodiments, the AHU controller 330 controls the amount of heating or cooling applied to the supply air 310 by adjusting the speed of the fan 338.

[0072] The cooling coil 334 may receive cooled fluid from the waterside system 200 (e.g., from the chilled water loop 216) via the conduit 342 and may return the cooled fluid to the waterside system 200 via the conduit 344. A valve 346 may be positioned along the conduit 342 or the conduit 344 to control the flow rate of the cooled fluid through the cooling coil 334. In some embodiments, the cooling coil 334 comprises multiple stages of cooling coils that may be independently activated and deactivated (e.g., by the AHU controller 330, by the BMS controller 366, etc.) to adjust the amount of cooling applied to the supply air 310.

[0073] The heating coil 336 may receive heated fluid from the waterside system 200 (e.g., from the hot water loop 214) via line 348 and may return the heated fluid to the waterside system 200 via line 350. A valve 352 may be positioned along the pipe 348 or the pipe 350 to control the flow rate of the heated fluid through the heating coil 336. In some embodiments, the heating coil 336 comprises a multi-stage heating coil that may be independently activated and deactivated (e.g., by the AHU controller 330, by the BMS controller 366, etc.) to adjust the amount of heating applied to the supply air 310.

[0074] Each of valves 346 and 352 may be controlled by an actuator. For example, valve 346 may be controlled by actuator 354, and valve 352 may be controlled by actuator 356. Actuators 354-356 may communicate with AHU controller 330 via communication links 358-360. Actuators 354-356 may receive control signals from AHU controller 330 and may provide feedback signals to controller 330. In some embodiments, AHU controller 330 receives a measurement of supply air temperature from a temperature sensor 362 positioned in supply air duct 312 (e.g., downstream of cooling coil 334 and / or heating coil 336). AHU controller 330 may also receive a measurement of the temperature of building zone 306 from a temperature sensor 364 located in building zone 306.

[0075] In some embodiments, the AHU controller 330 operates valves 346 and 352 via actuators 354-356 to adjust the amount of heating or cooling provided to the supply air 310 (e.g., to achieve a set point temperature of the supply air 310 or to maintain the temperature of the supply air 310 within a set point temperature range). The position of the valves 346 and 352 affects the amount of heating or cooling provided to the supply air 310 through the cooling coil 334 or the heating coil 336 and can be related to the amount of energy consumed to achieve the desired supply air temperature. The AHU 330 can control the temperature of the supply air 310 and / or the building zone 306 by activating or deactivating the coils 334-336, adjusting the speed of the fan 338, or a combination of both.

[0076] Still refer to Figure 3, the airside system 300 is shown to include a building management system (BMS) controller 366 and a client device 368. The BMS controller 366 may include one or more computer systems (e.g., servers, supervisory controllers, subsystem controllers, etc.) that function as a system-level controller, application or data server, head node, or master controller for the airside system 300, the waterside system 200, the HVAC system 100, and / or other controllable systems serving the building 10. The BMS controller 366 may communicate with multiple downstream building systems or subsystems (e.g., the HVAC system 100, security system, lighting system, waterside system 200, etc.) via a communication link 370 according to similar or different protocols (e.g., LON, BACnet, etc.). In various embodiments, the AHU controller 330 and the BMS controller 366 may be separate (e.g., Figure 3 In an integrated embodiment, the AHU controller 330 may be a software module configured to be executed by a processor of the BMS controller 366.

[0077] In some embodiments, the AHU controller 330 receives information (e.g., commands, set points, operating boundaries, etc.) from the BMS controller 366 and provides information (e.g., temperature measurements, valve or actuator positions, operating states, diagnostics, etc.) to the BMS controller 366. For example, the AHU controller 330 may provide the BMS controller 366 with temperature measurements from the temperature sensors 362-364, equipment on / off states, equipment operating capacities, and / or any other information that may be used by the BMS controller 366 to monitor or control variable states or conditions within the building zone 306.

[0078] The client device 368 may include one or more human-machine interfaces or client interfaces (e.g., a graphical user interface, a reporting interface, a text-based computer interface, a client-oriented web service, a web server that provides pages to a web client, etc.) for controlling, viewing, or otherwise interacting with the HVAC system 100, its subsystems, and / or devices. The client device 368 may be a computer workstation, a client terminal, a remote or local interface, or any other type of user interface device. The client device 368 may be a fixed terminal or a mobile device. For example, the client device 368 may be a desktop computer, a computer server with a user interface, a laptop computer, a tablet computer, a smart phone, a PDA, or any other type of mobile or non-mobile device. The client device 368 may communicate with the BMS controller 366 and / or the AHU controller 330 via a communication link 372.

[0079] Building Management System

[0080] Reference now Figure 4 , a block diagram of a building management system (BMS) 400 is shown, according to some embodiments. The BMS 400 can be implemented in a building 10 to automatically monitor and control various building functions. The BMS 400 is shown to include a BMS controller 366 and a plurality of building subsystems 428. The building subsystems 428 are shown to include a building electrical subsystem 434, an information and communication technology (ICT) subsystem 436, a security subsystem 438, an HVAC subsystem 440, a lighting subsystem 442, an elevator / escalator subsystem 432, and a fire safety subsystem 430. In various embodiments, the building subsystems 428 may include fewer, additional, or alternative subsystems. For example, the building subsystems 428 may additionally or alternatively include a refrigeration subsystem, an advertising or signage subsystem, a cooking subsystem, a vending subsystem, a printer or copy service subsystem, or any other type of building subsystem that uses controllable devices and / or sensors to monitor or control the building 10. In some embodiments, the building subsystem 428 includes the waterside system 200 and / or the airside system 300, as shown in FIG. Figures 2 to 3 as described.

[0081] Each of the building subsystems 428 may include any number of devices, controllers, and connections for performing its respective functions and control activities. The HVAC subsystem 440 may include many of the same components as the HVAC system 100, as described with reference to FIG. Figures 1 to 3 As described. For example, HVAC subsystem 440 may include chillers, boilers, any number of air handling units, economizers, field controllers, supervisory controllers, actuators, temperature sensors, and other devices for controlling temperature, humidity, airflow, or other variable conditions within building 10. Lighting subsystem 442 may include any number of lighting fixtures, ballasts, lighting sensors, dimmers, or other devices configured to controllably adjust the amount of light provided to a building space. Security subsystem 438 may include occupancy sensors, video surveillance cameras, digital video recorders, video processing servers, intrusion detection devices, access control devices and servers, or other security-related devices.

[0082] Still reference Figure 4, the BMS controller 366 is shown to include a communication interface 407 and a BMS interface 409. The interface 407 can facilitate communication between the BMS controller 366 and external applications (e.g., monitoring and reporting applications 422, enterprise control applications 426, remote systems and applications 444, applications resident on client devices 448, etc.) to allow a user to control, monitor, and adjust the BMS controller 366 and / or subsystems 428. The interface 407 can also facilitate communication between the BMS controller 366 and client devices 448. The BMS interface 409 can facilitate communication between the BMS controller 366 and building subsystems 428 (e.g., HVAC, lighting security, elevators, power distribution, enterprise, etc.).

[0083] Interfaces 407, 409 may be or include wired or wireless communication interfaces (e.g., sockets, antennas, transmitters, receivers, transceivers, wire connectors, etc.) for data communication with building subsystems 428 or other external systems or devices. In various embodiments, communication through interfaces 407, 409 may be direct communication (e.g., local wired or wireless communication) or through a communication network 446 (e.g., WAN, Internet, cellular network, etc.). For example, interfaces 407, 409 may include an Ethernet card and port for sending and receiving data through an Ethernet-based communication link or network. In another example, interfaces 407, 409 may include a Wi-Fi transceiver for communicating through a wireless communication network. In another example, one or both of interfaces 407, 409 may include a cellular or mobile phone communication transceiver. In one embodiment, communication interface 407 is a power line communication interface, and BMS interface 409 is an Ethernet interface. In other embodiments, both communication interface 407 and BMS interface 409 are Ethernet interfaces or the same Ethernet interface.

[0084] Still reference Figure 4 , the BMS controller 366 is shown to include a processing circuit 404, which includes a processor 406 and a memory 408. The processing circuit 404 can be communicatively connected to the BMS interface 409 and / or the communication interface 407, so that the processing circuit 404 and its various components can send and receive data through the interfaces 407, 409. The processor 406 can be implemented as a general-purpose processor, an application-specific integrated circuit (ASIC), one or more field programmable gate arrays (FPGA), a group of processing components, or other suitable electronic processing components.

[0085] Memory 408 (e.g., memory, memory unit, storage device, etc.) may include one or more devices (e.g., RAM, ROM, flash memory, hard disk storage, etc.) for storing data and / or computer code for completing or facilitating the various processes, layers, and modules described in this application. Memory 408 may be or include volatile memory or non-volatile memory. Memory 408 may include database components, object code components, script components, or any other type of information structure for supporting the various activities and information structures described in this application. According to some embodiments, memory 408 is communicatively connected to processor 406 via processing circuit 404 and includes computer code for executing (e.g., via processing circuit 404 and / or processor 406) one or more processes described herein. One or more non-transitory computer-readable media may store instructions that, when executed by one or more processors, perform the operations disclosed herein.

[0086] In some embodiments, the BMS controller 366 is implemented within a single computer (e.g., one server, one house, etc.). In various other embodiments, the BMS controller 366 may be distributed across multiple servers or computers (e.g., may exist in dispersed locations). Further, although Figure 4 The applications 422 and 426 are shown as existing external to the BMS controller 366 , but in some embodiments, the applications 422 and 426 may be hosted within the BMS controller 366 (eg, within the memory 408 ).

[0087] Still reference Figure 4 , the memory 408 is shown to include an enterprise integration layer 410, an automated measurement and verification (AM&V) layer 412, a demand response (DR) layer 414, a fault detection and diagnostics (FDD) layer 416, an integrated control layer 418, and a building subsystem integration layer 420. The layers 410-420 may be configured to receive inputs from building subsystems 428 and other data sources, determine optimal control actions for the building subsystems 428 based on the inputs, generate control signals based on the optimal control actions, and provide the generated control signals to the building subsystems 428. The following paragraphs describe some of the common functions performed by each of the layers 410-420 in the BMS 400.

[0088] The enterprise integration layer 410 can be configured to provide information and services to clients or local applications to support a variety of enterprise-level applications. For example, the enterprise control application 426 can be configured to provide subsystem cross-control to a graphical user interface (GUI) or to any number of enterprise-level business applications (e.g., billing system, user identification system, etc.). The enterprise control application 426 can additionally or alternatively be configured to provide a configuration GUI for configuring the BMS controller 366. In yet other embodiments, the enterprise control application 426 can work with the layers 410-420 to optimize building performance (e.g., efficiency, energy usage, comfort, or safety) based on input received at the interface 407 and / or the BMS interface 409.

[0089] The building subsystem integration layer 420 may be configured to manage communications between the BMS controllers 366 and the building subsystems 428. For example, the building subsystem integration layer 420 may receive sensor data and input signals from the building subsystems 428 and provide output data and control signals to the building subsystems 428. The building subsystem integration layer 420 may also be configured to manage communications between the building subsystems 428. The building subsystem integration layer 420 translates communications (e.g., sensor data, input signals, output signals, etc.) across multiple multi-vendor / multi-protocol systems.

[0090] The demand response layer 414 can be configured to optimize resource usage (e.g., electricity usage, natural gas usage, water usage, etc.) and / or the monetary cost of such resource usage in response to meeting the needs of the building 10. The optimization can be based on time-of-use prices, curtailment signals, energy availability, or other data received from a utility provider, a distributed energy generation system 424, from an energy storage device 427 (e.g., a high-temperature TES 242, a low-temperature TES 244, etc.), or from other sources. The demand response layer 414 can receive inputs from other layers of the BMS controller 366 (e.g., a building subsystem integration layer 420, an integrated control layer 418, etc.). The inputs received from other layers can include environmental or sensor inputs, such as temperature, carbon dioxide levels, relative humidity levels, air quality sensor outputs, occupancy sensor outputs, room schedules, etc. The inputs can also include inputs such as electricity usage (e.g., expressed in kWh), heat load measurements, pricing information, scheduled pricing, smoothed pricing, curtailment signals from utilities, and the like.

[0091] According to some embodiments, the demand response layer 414 includes control logic for responding to the data and signals it receives. These responses may include communicating with control algorithms in the integrated control layer 418 in a controlled manner, changing control strategies, changing set points, or activating / deactivating building equipment or subsystems. The demand response layer 414 may also include control logic configured to determine when to utilize stored energy. For example, the demand response layer 414 may determine to use energy from the energy storage device 427 just before the start of a peak usage period.

[0092] In some embodiments, the demand response layer 414 includes control modules configured to proactively initiate control actions (e.g., automatically change set points) that minimize energy costs based on one or more demand-representative or demand-based inputs (e.g., prices, curtailment signals, demand levels, etc.). In some embodiments, the demand response layer 414 uses equipment models to determine an optimal set of control actions. Equipment models may include, for example, thermodynamic models that describe inputs, outputs, and / or functions performed by various sets of building equipment. Equipment models may represent a collection of building equipment (e.g., sub-equipment, an array of chillers, etc.) or individual devices (e.g., individual chillers, heaters, pumps, etc.).

[0093] The demand response layer 414 may further include or utilize one or more demand response policy definitions (e.g., a database, XML file, etc.). The policy definitions may be edited or adjusted by a user (e.g., via a graphical user interface) so that control actions initiated in response to a demand input may be customized for the user's application, desired comfort level, specific building equipment, or based on other concerns. For example, a demand response policy definition may specify which equipment may be turned on or off in response to a particular demand input, how long a system or piece of equipment should be off, which set points may be changed, what is the allowable set point adjustment range, how long to maintain a high demand set point before returning to a normally scheduled set point, how close to a capacity limit, which equipment modes to utilize, energy transfer rates to and from an energy storage device (e.g., a thermal storage tank, a battery bank, etc.) (e.g., maximum rate, alarm rate, other rate boundary information, etc.), and when to dispatch on-site energy generation (e.g., via a fuel cell, an electric generator set, etc.).

[0094] The integrated control layer 418 may be configured to use data inputs or outputs from the building subsystem integration layer 420 and / or the demand response layer 414 to make control decisions. Due to the subsystem integration provided by the building subsystem integration layer 420, the integrated control layer 418 may integrate the control activities of the subsystems 428 so that the subsystems 428 behave as a single integrated super-system. In some embodiments, the integrated control layer 418 includes control logic that uses inputs and outputs from multiple building subsystems to provide a higher level of comfort and energy savings than the individual subsystems can provide on their own. For example, the integrated control layer 418 may be configured to use inputs from a first subsystem to make energy-saving control decisions for a second subsystem. The results of these decisions may be transmitted back to the building subsystem integration layer 420.

[0095] The integrated control layer 418 is shown as being logically located below the demand response layer 414. The integrated control layer 418 can be configured to enhance the effectiveness of the demand response layer 414 by enabling the building subsystems 428 and their corresponding control loops to be controlled in coordination with the demand response layer 414. This configuration can advantageously reduce disruptive demand response behavior compared to conventional systems. For example, the integrated control layer 418 can be configured to ensure that a demand response-driven upward adjustment to the set point of the chilled water temperature (or another component that directly or indirectly affects the temperature) does not result in an increase in fan energy (or other energy used to cool the space) that would otherwise result in total building energy use greater than the energy saved at the chiller.

[0096] The integrated control layer 418 may be configured to provide feedback to the demand response layer 414 so that the demand response layer 414 checks whether constraints (e.g., temperature, lighting levels, etc.) are properly maintained, even when required load shedding is in progress. Constraints may also include set points or sensed boundaries related to safety, equipment operating limits and performance, comfort, fire codes, electrical codes, energy codes, etc. The integrated control layer 418 is also logically located below the fault detection and diagnostics layer 416 and the automated measurement and verification layer 412. The integrated control layer 418 may be configured to provide calculated inputs (e.g., summaries) to these higher levels based on outputs from one or more building subsystems.

[0097] The automated measurement and verification (AM&V) layer 412 can be configured to verify that the control strategies commanded by the integrated control layer 418 or the demand response layer 414 are operating properly (e.g., using data aggregated by the AM&V layer 412, the integrated control layer 418, the building subsystem integration layer 420, the FDD layer 416, or other layers). The calculations performed by the AM&V layer 412 can be based on the building system energy model and / or equipment models for the various BMS devices or subsystems. For example, the AM&V layer 412 can compare the model predicted output with the actual output of the building subsystem 428 to determine the accuracy of the model.

[0098] The fault detection and diagnostics (FDD) layer 416 may be configured to provide continuous fault detection for the building subsystems 428, building subsystem devices (i.e., building equipment), and control algorithms used by the demand response layer 414 and the integrated control layer 418. The FDD layer 416 may receive data inputs from the integrated control layer 418, directly from one or more building subsystems or devices, or from another data source. The FDD layer 416 may automatically diagnose and respond to detected faults. The response to a detected or diagnosed fault may include providing a warning message to a user, a maintenance scheduling system, or a control algorithm configured to attempt to repair the fault or resolve the fault.

[0099] The FDD layer 416 may be configured to use the detailed subsystem inputs available at the building subsystem integration layer 420 to output a specific identification of a faulty component or a cause of a fault (e.g., a loose damper linkage). In other exemplary embodiments, the FDD layer 416 is configured to provide a "fault" event to the integrated control layer 418, which executes control strategies and policies in response to the received fault event. According to some embodiments, the FDD layer 416 (or strategies executed by an integrated control engine or business rules engine) may shut down a system or direct control activities around a faulty device or system to reduce energy waste, extend equipment life, or ensure an appropriate control response.

[0100] The FDD layer 416 can be configured to store or access a variety of different system data repositories (or data points of live data). The FDD layer 416 can use some of the contents of the data repository to identify faults at the equipment level (e.g., a specific chiller, a specific AHU, a specific terminal unit, etc.) and use other contents to identify faults at the component or subsystem level. For example, the building subsystem 428 can generate temporal (i.e., time series) data indicating the performance of the BMS 400 and its various components. The data generated by the building subsystem 428 can contain measured or calculated values ​​that exhibit statistical characteristics and provide information about how the corresponding system or process (e.g., a temperature control process, a flow control process, etc.) performs based on errors from its set points. These processes can be checked by the FDD layer 416 to expose when the performance of the system begins to degrade, and alert the user to repair the fault before it becomes more serious.

[0101] Reference now Figure 5 , according to some embodiments, a block diagram of another building management system (BMS) 500 is shown. The BMS 500 can be used to monitor and control devices of the HVAC system 100, the water side system 200, the air side system 300, the building subsystem 428, and other types of BMS devices (e.g., lighting devices, security devices, etc.) and / or HVAC equipment.

[0102] The BMS 500 provides a system architecture that facilitates automatic device discovery and device model distribution. Device discovery can occur across multiple different communication buses (e.g., system bus 554, zone buses 556-560 and 564, sensor / actuator bus 566, etc.) and across multiple different communication protocols at multiple layers of the BMS 500. In some embodiments, device discovery is accomplished using an active node table that provides status information for devices connected to each communication bus. For example, each communication bus of a new device can be monitored by monitoring the corresponding active node table of the new node. When a new device is detected, the BMS 500 can begin to interact with the new device (e.g., send control signals, use data from the device) without user interaction.

[0103] Some devices in the BMS 500 present themselves to the network using device models. The device model defines device object properties, view definitions, schedules, trends, and associated BACnet value objects (e.g., analog values, binary values, multi-state values, etc.) for integration with other systems. Some devices in the BMS 500 store their own device models. Other devices in the BMS 500 have device models stored externally (e.g., within other devices). For example, the zone coordinator 508 may store a device model for the bypass damper 528. In some embodiments, the zone coordinator 508 automatically creates a device model for the bypass damper 528 or other devices on the zone bus 558. Other zone coordinators may also create device models for devices connected to their zone buses. Device models for devices may be automatically created based on the type of data points exposed by the device on the zone bus, the device type, and / or other device attributes. Several examples of automatic device discovery and device model distribution are discussed in more detail below.

[0104] Still refer to Figure 5 , the BMS 500 is shown to include a system manager 502; a plurality of zone coordinators 506, 508, 510, and 518; and a plurality of zone controllers 524, 530, 532, 536, 548, and 550. The system manager 502 can monitor data points in the BMS 500 and report the monitored variables to various monitoring and / or control applications. The system manager 502 can communicate with a client device 504 (e.g., a user device, a desktop computer, a laptop computer, a mobile device, etc.) via a data communication link 574 (e.g., BACnet IP, Ethernet, wired or wireless communication, etc.). The system manager 502 can provide a user interface to the client device 504 via the data communication link 574. The user interface can allow a user to monitor and / or control the BMS 500 via the client device 504.

[0105] In some embodiments, the system manager 502 is connected to the zone coordinators 506-510 and 518 via a system bus 554. The system manager 502 can be configured to communicate with the zone coordinators 506-510 and 518 using a master-slave token passing (MSTP) protocol or any other communication protocol through the system bus 554. The system bus 554 can also connect the system manager 502 with other devices such as constant capacity (CV) rooftop units (RTUs) 512, input / output modules (IOMs) 514, thermostat controllers 516 (e.g., TEC5000 series thermostat controllers), and network automation engines (NAEs) or third-party controllers 520. The RTU 512 can be configured to communicate directly with the system manager 502 and can be directly connected to the system bus 554. Other RTUs can communicate with the system manager 502 through intermediary devices. For example, a wired input 562 can connect a third-party RTU 542 to a thermostat controller 516, which is connected to the system bus 554.

[0106] The system manager 502 can provide a user interface for any device that has a device model. Devices such as the zone coordinators 506-510 and 518 and the thermostat controller 516 can provide their device models to the system manager 502 via the system bus 554. In some embodiments, the system manager 502 automatically creates device models for connected devices (e.g., IOM 514, third-party controller 520, etc.) that do not have device models. For example, the system manager 502 can create a device model for any device in response to a device tree request. The device model created by the system manager 502 can be stored within the system manager 502. The system manager 502 can then use the device model created by the system manager 502 to provide a user interface for devices that do not have their own device models. In some embodiments, the system manager 502 stores view definitions for each type of device connected via the system bus 554 and uses the stored view definitions to generate a user interface for the device.

[0107] Each zone coordinator 506-510 and 518 can be connected to one or more of the zone controllers 524, 530-532, 536 and 548-550 via a zone bus 556, 558, 560 and 564. The zone coordinators 506-510 and 518 can communicate with the zone controllers 524, 530-532, 536 and 548-550 using the MSTP protocol or any other communication protocol through the zone bus 556-560 and 564. The zone bus 556-560 and 564 can also connect the zone coordinators 506-510 and 518 with other types of devices such as variable air volume (VAV) RTUs 522 and 540, conversion bypass (COBP) RTUs 526 and 552, bypass dampers 528 and 546, and peak controllers 534 and 544.

[0108] The zone coordinators 506-510 and 518 may be configured to monitor and command various zone systems. In some embodiments, each zone coordinator 506-510 and 518 monitors and commands a separate zone system and is connected to the zone systems via a separate zone bus. For example, the zone coordinator 506 may be connected to the VAV RTU 522 and the zone controller 524 via the zone bus 556. The zone coordinator 508 may be connected to the COBP RTU 526, the bypass damper 528, the COBP zone controller 530, and the VAV zone controller 532 via the zone bus 558. The zone coordinator 510 may be connected to the peak controller 534 and the VAV zone controller 536 via the zone bus 560. The zone coordinator 518 may be connected to the peak controller 544, the bypass damper 546, the COBP zone controller 548, and the VAV zone controller 550 via the zone bus 564.

[0109] A single model of the zone coordinators 506-510 and 518 can be configured to handle multiple different types of zone systems (e.g., VAV zone systems, COBP zone systems, etc.). Each zone system may contain an RTU, one or more zone controllers, and / or bypass dampers. For example, the zone coordinators 506 and 510 are shown as Verasys VAV Engines (VVEs) connected to VAV RTUs 522 and 540, respectively. The zone coordinator 506 is directly connected to the VAV RTU 522 via the zone bus 556, while the zone coordinator 510 is connected to the third party VAV RTU 540 via a wired input 568 that provides access to the peak controller 534. The zone coordinators 508 and 518 are shown as Verasys COBP Engines (VCEs) connected to COBP RTUs 526 and 552, respectively. The zone coordinator 508 is directly connected to the COBP RTU 526 via the zone bus 558 , while the zone coordinator 518 is connected to the third party COBP RTU 552 via a wired input 570 provided to the peak controller 544 .

[0110] The zone controllers 524, 530-532, 536, and 548-550 may communicate with various BMS devices (e.g., sensors, actuators, etc.) via a sensor / actuator (SA) bus. For example, the VAV zone controller 536 is shown connected to the networked sensor 538 via the SA bus 566. The zone controller 536 may communicate with the networked sensor 538 using the MSTP protocol or any other communication protocol. Although Figure 5 Only one SA bus 566 is shown, but it should be understood that each zone controller 524, 530-532, 536 and 548-550 can be connected to a different SA bus. Each SA bus can connect the zone controller with various sensors (e.g., temperature sensors, humidity sensors, pressure sensors, light sensors, occupancy sensors, etc.), actuators (e.g., damper actuators, valve actuators, etc.) and / or other types of controllable devices (e.g., refrigerators, heaters, fans, pumps, etc.).

[0111] Each zone controller 524, 530-532, 536 and 548-550 can be configured to monitor and control different building zones. The zone controllers 524, 530-532, 536 and 548-550 can monitor and control various building zones using inputs and outputs provided through their SA bus. For example, the zone controller 536 can use the temperature input (e.g., the measured temperature of the building zone) received from the networked sensor 538 through the SA bus 566 as feedback in the temperature control algorithm. The zone controllers 524, 530-532, 536 and 548-550 can use various types of control algorithms (e.g., state-based algorithms, extreme value search control (ESC) algorithms, proportional-integral (PI) control algorithms, proportional-integral-derivative (PID) control algorithms, model predictive control (MPC) algorithms, feedback control algorithms, etc.) to control variable states or conditions (e.g., temperature, humidity, airflow, lighting, etc.) in or around the building 10.

[0112] Net Zero Energy Facilities

[0113] Reference now Figure 6 , a block diagram of a net zero energy facility 600 is shown, according to some embodiments. The facility 600 can be a building (e.g., building 10, a residential building, a commercial building, a medical facility, a school, etc.), a collection of buildings, a campus, an outdoor facility (e.g., a park, a rail yard, a port, a sports facility, etc.), or other locations in various embodiments. The facility 600 is shown to include a system manager 502, a green energy source 602, and an energy load 604, and the facility 600 is also shown to be connected to a utility grid 606.

[0114] Green energy 602 is shown to include photovoltaic system 608, windmill 610 and geothermal system 612, but can include any type of green (e.g., renewable, sustainable, eco-friendly, etc.) type of energy generation. Green energy source 602 is shown to be located at facility 600. For example, photovoltaic system 608 and / or windmill 610 can be located on the roof. As another example, photovoltaic system 608 can be located adjacent to a building or other structure. Photovoltaic system 608 can include photovoltaic cells configured to convert solar irradiance into electricity. Windmill 610 can include turbines configured to convert wind at facility 600 into electricity. Geothermal system 612 can be configured to convert geothermal energy (e.g., geothermal heat from underground) into electricity and / or directly use geothermal heat to heat buildings or serve other building needs. Depending on the access of facility 600 to various resources and geological features, various types of green energy sources 602 can be included in various embodiments (e.g., hydroelectric power, etc.). In some embodiments, green energy source 602 includes an energy storage device (eg, a battery) that is capable of delaying between the time of energy production / harvest and the time of energy use.

[0115] Energy loads 604 are shown to include HVAC equipment 614, lighting fixtures 616, appliances 618, and computing devices 620. HVAC equipment 614 may include equipment such as the waterside system 200 and the airside system 300 described above. In various embodiments, HVAC equipment 614 may include variable refrigerant flow systems, room air conditioners, window air conditioners, etc. Lighting fixtures 616 are configured to illuminate facility 600 and may include various lamps, bulbs, arrays, etc., including indoor and outdoor lighting. Appliances 618 may include various miscellaneous appliances that consume energy at facility 600 and may change in various scenarios and usage of the facility (e.g., ovens, stoves, microwaves, dishwashers, water heaters, laboratory equipment, medical devices, production line equipment, electric vehicle charging stations, etc.). Computing devices 620 may include personal computing devices (e.g., desktop computers, laptop computers, etc.) and / or servers, network infrastructure, data centers, etc. Various energy loads 604 may be included at facility 600.

[0116] The system manager 502 is configured to reference Figures 7 to 11Operations are performed as described below so that the energy load 604 operates to consume the same amount of energy produced by the green energy source 602 over a time period (e.g., more than a month, more than a quarter, more than a year). For a particular sub-period of the time period, the energy production by the green energy source 602 may be greater or less than the energy consumption by the energy load 604 (e.g., due to fluctuations in the availability of renewable energy, a surge in energy demand, etc.), wherein the grid 606 absorbs excess production and serves excess demand. The system manager 502 operates to balance such sub-periods over the time period so that the energy production by the green energy source 602 and the energy consumption by the energy load 604 are balanced at the end of the time period. When such operations are successfully performed, the facility 600 may be characterized as a net zero energy facility. In some embodiments, the energy consumption by the energy load 604 over the time period is less than the energy production by the green energy source 602, in which case the facility 600 produces more energy than it consumes. Such operation may also enable facility 600 to be characterized as a net zero energy facility (i.e., energy consumption by facility 600 is less than or equal to energy consumption by facility 600 over a given time period). Although the disclosure herein focuses on examples of net energy consumption, the teachings herein may be applicable to additionally or alternatively involve net carbon emissions, i.e., the cumulative difference between carbon emissions and carbon capture (sequestration, acquisition of carbon credits, etc.) over a time period (e.g., by using carbon emissions instead of energy consumption and using carbon capture and / or carbon credit acquisition instead of energy production throughout the examples of the present disclosure).

[0117] Now refer to Figure 7 , a flow chart of a process 700 for achieving net zero energy consumption within a time period is shown, according to some embodiments. The process 700 may be performed by the facility 600, such as by the system manager 502. In some embodiments, the process 700 is performed prior to or at the beginning of the time period in which net zero energy consumption is to be achieved. In other embodiments, the process 700 may be performed partially into the time period, such that a certain amount of energy consumption and / or energy production by the facility 600 has occurred within the time period up to the point in time when the process 700 is performed.

[0118] At step 702, the energy margin of the facility 600 up to the current time (i.e., the difference between energy production and energy consumption by the facility 600) is calculated. The energy margin can be defined as the total amount of energy produced by the facility 600 in the time period up to the current time minus the total amount of energy consumed by the facility 600 in the time period up to the current time. For example, the energy margin can be calculated based on data from an energy meter that measures energy production and consumption at the facility 600. The energy margin can be calculated from a specified starting point (e.g., the beginning of a month, the beginning of a quarter, the beginning of a fiscal year or calendar year, etc.) until the current time when the process 700 is executed. The symbols herein designate the energy margin up to the current time as E0. In some embodiments, the energy consumption of the facility includes the energy consumption of equipment located at the building or directly serving the building (e.g., providing heating or cooling to the facility). In some embodiments, the consumption of the facility includes consumption associated with the transportation of goods and / or personnel into and out of the facility or consumption otherwise associated with the operation of the facility, such as energy consumption by transportation vehicles (e.g., cars, trucks, trains, airplanes, ships, etc.).

[0119] At step 704, the facility 600 predicts energy consumption for a future time period (e.g., an upcoming time period), for example, using a neural network trained to predict future energy consumption. The energy consumption predicted at step 704 can be the total energy consumption for the entire future time period, or the energy consumption can be predicted for a specific time step (e.g., a sub-portion of the future time period discretized at any level of granularity), as a continuous function, etc. Notation herein designates energy consumption as E 消耗 The energy consumption predicted at step 704 may be the energy consumption of facility 600 without load shedding (e.g., energy conservation measures, curtailment, etc.). If load shedding is subsequently employed, actual energy consumption of facility 600 may be less than the amount predicted in step 704, as described below.

[0120] At step 706, the facility 600 predicts energy production for a future time period (e.g., an upcoming time period), for example, using a neural network trained to predict future energy production. The energy production predicted at step 706 can be the total energy production for the entire future time period, or the energy production can be predicted for a specific time step (e.g., a sub-portion of the future time period discretized at any level of granularity), as a continuous function, etc. Notation herein designates energy production as E 生产 In some embodiments, energy production by facility 600 over future time periods is predictable (eg, based on sunlight, weather patterns, etc.) but not controllable.

[0121] At step 708, the amount of energy to be curtailed in the future time period to achieve net zero consumption is calculated. The amount of energy to be curtailed can be calculated as E 削减 =E 消耗 -E 生产 -E0, where E 削减 is the amount by which the energy predicted to be consumed in a future time period exceeds the amount of energy predicted to be produced in a future time period, taking into account the energy margin (i.e. the remaining energy generated until the current time). Therefore, E 削减 Represents the predicted energy consumption E 消耗 The amount of energy that needs to be reduced to achieve net zero energy in that time period. In another statement, the amount of energy that needs to be removed in the future time period from T0 to T1 is the amount that must be reduced To ensure In scenarios where the inequality is satisfied and no load shedding is required (ie, no load shedding is required to achieve a net zero state), process 700 may stop at step 708 .

[0122] At step 710, the energy to be curtailed is allocated (e.g., distributed, decomposed, segmented, categorized, etc.) across the energy loads 604, e.g., based on the flexibility of the different energy loads 604 and the associated loads. For example, in one scenario, the energy consumed by the HVAC equipment 614 may account for a relatively large portion of the total energy load (i.e., the total energy consumption of all energy loads 604) and may be relatively flexible (e.g., depending on the facility's tolerance for mild occupant discomfort) compared to other domains (e.g., data center equipment that is difficult to control, lighting that must be turned on at certain times to allow people to see in the space, etc.). Step 710 works to allocate the total energy to be curtailed across the energy loads 604 so that each energy load (e.g., each building domain) is assigned a specific amount of energy that should be curtailed by that specific energy load (e.g., a first load reduction amount for the HVAC equipment 614, a second load reduction amount for the lighting fixtures 616, a third load reduction amount for the appliances 618, etc.). For example, the actions taken by each of the energy loads 604 to achieve the allocated amount of load reduction may include a change in the operation of the energy load relative to a predicted baseline operation of the energy load.

[0123] At step 712, the operational changes made to achieve the energy curtailed for each energy load 604 are back-calculated. An artificial intelligence model for each energy load 604 (e.g., a model for HVAC equipment 614, a model for lighting fixtures 616, a model for appliances 618, a model for computing and data center equipment 620, etc.) can be used to predict the amount of energy savings that can be achieved through different operational changes. The model can also consider constraints or penalties associated with negative impacts on facility performance and utility (e.g., occupant discomfort, reduced productivity, scheduling inconvenience, etc.), which guide operational changes to still provide acceptable facility performance and utility while achieving energy load reduction. Step 712 outputs a set of operational changes (e.g., set point changes, on / off decisions, schedules, etc.) that can be implemented by controlling the energy loads 604 in accordance with the operational changes. By implementing the operational changes over a time period, the amount of energy curtailed from the predicted baseline energy is such that In this inequality, The value of may decrease relative to the predicted value in step 704 as a result of the operational change performed in step 712.

[0124] Reference now Figure 8 , according to some embodiments, a diagram 800 is shown of the amount of energy that may be involved in an example execution of process 700. In particular, Figure 8 A first region 802 is shown before time T0 and a second region 804 is shown after time T0 and through time T1. Process 700 may be performed at or near time T0 (eg, just before) and continue for a period through time T1.

[0125] Graph 800 shows an actual energy consumption line 806 and an actual energy generation line 808 in a first region 802, which represent actual (e.g., measured) values ​​of energy consumption and production over a time period until time T0 (e.g., a one-day period). As shown, both the energy consumption line 806 and the actual energy generation line 808 follow a curve that increases in the middle of the cycle (e.g., in the middle of the day). As a possibility, graph 800 may represent a scenario in which both photovoltaic energy production and energy demand surge in the middle of the day (e.g., due to increased solar radiation and associated cooling requirements around noon on a clear summer day). In the first region 802, the actual energy consumption line 806 and the actual energy generation line 808 intersect twice. That is, at certain moments, consumption is greater than generation, and at other time points, consumption is less than generation. Although the first region 802 contains a period in which energy consumption is greater than energy generation, if the area below the energy consumption line 806 in the first region 802 is equal to or less than the area below the energy generation line 808 in the first region 802, the first region 802 represents a net zero energy period. The difference between the area under the energy generation line 808 in the first region 802 and the energy consumption line 806 in the first region 802 represents the energy margin (ie, remaining energy generated) during the first time period before time T0.

[0126] The second region 804 shows a forecast for a time period T0 to T1. In particular, an energy consumption forecast line 810, an energy generation forecast line 812, and a curtailed energy consumption line 814 are shown. The energy consumption forecast line 810 shows the energy predicted to be consumed for the time period T0 to T1 (e.g., output from step 704). The energy generation forecast line 812 shows the energy predicted to be generated for the time period T0 to T1 (e.g., output from step 706). The curtailed energy consumption line 814 represents the amount of energy predicted to be consumed if an operational change is made relative to the baseline forecast represented by the energy consumption forecast line 810 to curtail energy (e.g., as a result of step 712).

[0127] In the graph 800 and in the second region 804, the total amount of energy to be curtailed to achieve net zero energy is the difference between the area under the energy consumption forecast line 810 and the area under the energy generation forecast line 812, and the total amount of energy predicted to be curtailed is the area between the energy consumption line 810 and the curtailed energy consumption line 814. Accordingly, if the area between the energy consumption line 810 and the curtailed energy consumption line 814 (plus the energy margin from the first region 802, if any) is greater than or equal to the difference between the area under the energy consumption forecast line 810 and the area under the energy generation forecast line 812, then the graph 800 shows a scenario in which energy consumption is curtailed to achieve net zero energy consumption at time T1.

[0128] Reference now Fig. 9 , a block diagram of a system manager 502 (or a portion thereof) is shown, according to some embodiments. The system manager 502 may be implemented using circuitry including one or more processors and one or more non-transitory computer-readable media storing program instructions that, when executed by the one or more processors of the system manager 502, cause the one or more processors to perform operations attributed herein to the system manager 502. For example, the system manager 502 may be located at a facility 600, may be implemented as a cloud resource or other software-as-a-service platform, or some combination thereof.

[0129] like Fig. 9 As shown, the system manager 502 provides a cascade control architecture in which a first predictive optimization is performed to determine a net energy trajectory within a first longer time period (e.g., a month, a quarter, a year, etc.), and the net energy trajectory is used as an input to a second predictive optimization, which is performed within a sub-period of the first longer time period (e.g., a week, a day, an hour, etc.) to determine curtailment actions for the energy load 604. The system manager 502 is shown to include a long-term predictor 900, a long-term planner 902, a short-term predictor 904, and a short-term advisor 906. The system manager 502 is shown to be in communication with the energy load 604.

[0130] The long-term predictor 900 is configured to predict the baseline energy consumption and energy production by the facility 600 in a first time period. The first time period may correspond to a period of interest to stakeholders (building managers, owners, business leaders, shareholders, regulators, etc.), during which such stakeholders expect the facility 600 to achieve net zero energy consumption. For example, the first time period may be a year (e.g., a company's fiscal year, calendar year). As another example, the first time period may be a quarter (i.e., a three-month period). As another example, the first time period may be a month. The long-term predictor 900 may use climate data, historical building data (e.g., data from a specific facility 600 of a previous time period, data from similar facilities, etc.), and various modeling techniques to predict the baseline energy consumption and energy production in the first time period. The prediction preferably includes a forecast time series or continuous function of the baseline energy consumption and production across the first time period.

[0131] The long term planner 902 is configured to use the predicted baseline energy consumption and energy production by the facility 600 as inputs to a first predictive optimization. The long term planner 902 may also predict utility rates or other time-varying characteristics related to energy use (e.g., marginal operating carbon emission rates). The long term planner 902 performs a predictive optimization to output a net energy trajectory for a first time period. The net energy trajectory indicates a value of net energy consumption at a certain moment in the time period, such as a time series of net energy values ​​associated with time steps within the time period. The first predictive optimization is preferably constrained so that the net energy trajectory achieves a value of zero at the end of the first time period, while allowing the net energy trajectory to take different values ​​within the time period.

[0132] The first forecast optimization can minimize the forecast cost within the time period. For example, the long-term planner can include an objective function that considers the cost of operating the facility 600 within the time period as a function of the net energy trajectory and the forecast of the baseline energy consumption and production (e.g., the cost of purchasing grid energy or other resources, maintenance costs, carbon emissions or other internal costs of pollution associated with the use of grid energy, etc.). The long-term planner 902 can perform forecast optimization to find the net energy trajectory that minimizes the objective function within the time period. For example, time-shifting energy consumption to better align consumption with production can help reduce the overall value of the objective function. As another example, time-shifting energy consumption away from the peak demand cycle of the utility grid can help reduce costs, marginal emissions, etc. associated with facility operation. As another example, at certain times in the first time period, load reduction can be cheaper, more sustainable, less emitting, etc. than at other times. The long-term planner 902 is configured to process any such considerations in order to output a net energy trajectory for the time period.

[0133] The long-term planner 902 thus outputs a net energy trajectory indicating the value of the net energy consumption at a certain moment in the time period. Fig.10 Additional details of some embodiments are shown. Fig. 9 As shown, the net energy trajectory is provided as an input to the short term advisor 906 .

[0134] The short-term predictor 904 is configured to predict energy production and baseline energy consumption within a sub-period (e.g., a day or a week) of the time period used by the long-term predictor 900 and the long-term planner 902. The short-term predictor 904 can also predict utility rates, marginal operating emission rates, etc. within the sub-period. For example, while the long-term predictor 900 can use general climate data, the short-term predictor 904 operates within a shorter timeline, where weather forecasts (e.g., from a third-party weather service) are relatively reliable and can be used for predictions of energy production and baseline energy consumption. Due to the shorter prediction range, the predictions made by the short-term predictor 904 are generally more accurate (i.e., closer to the actual situation that occurs) than the output of the long-term predictor 900. The cascade architecture of the system manager 502 thus benefits from generating predictions at the beginning of sub-periods of longer time periods by the short-term predictor 904 to facilitate higher quality operations of the short-term consultant 906 described below.

[0135] The short term advisor 906 is configured to use the net energy trajectory from the long term planner 902 and the forecast from the short term forecaster 904 to determine a reduction action to be implemented via the energy load 604 during the sub-period to reduce energy consumption such that the actual net energy tracks the net energy trajectory for the sub-period. For example, the short term advisor 906 may determine that the reduction action predicted by the short term advisor 906 will cause the actual net energy consumption from the beginning of the first time period to the end of the sub-period to be equal to the value of the net energy trajectory at the end of the sub-period. Example graphical illustrations of such objectives, constraints, etc. are provided in Fig.11 , and described with reference thereto.

[0136] The short-term consultant 906 may be configured to determine a reduction action as an output of a predictive optimization performed by the short-term consultant 906. Predictive optimization may minimize the impact of implementing a reduction action. In such examples, the short-term consultant 906 may use predictive models for different types of energy loads (e.g., for different building domains) that predict the impact of reduction actions on operating costs, emissions, pollution, facility productivity, occupant discomfort, etc. For example, the short-term consultant 906 may run a model that associates a reduction in energy consumption by a computing device 620 or appliance 618 with a reduction in productivity (e.g., characterized in financial terms) so that the short-term consultant 906 may assign numerical cost values ​​associated with different reduction options. As another example, the short-term consultant 906 may run a model that provides a numerical penalty value based on occupant discomfort associated with an attempt to reduce energy consumption of HVAC equipment. Thus, the short-term consultant 906 may associate various costs (financial or otherwise) with different reduction options. The short-term advisor 906 can then run an optimization for the different reduction options and use the associated costs of the reduction options, the short-term forecast of baseline energy consumption within the sub-period, the short-term forecast of energy production within the sub-period, an indication of the actual (e.g., measurement-based) net energy amount at the beginning of the sub-period, and the net energy trajectory to determine a set of reduction actions that are predicted to achieve the objectives defined by the net energy trajectory with minimal negative impact on facility performance (e.g., minimizing predicted costs).

[0137] Therefore, short-term advisor 906 outputs a reduction action to be implemented by energy load 604. Fig. 9 As shown, the curtailment action can be communicated from the system manager 502 to the energy load 604, for example in the form of an electronic request, a control signal, etc., which causes the energy load 604 to implement the curtailment action. In some embodiments, further optimization or predictive control processes are performed in a distributed manner at the local controller of the energy load 604 to optimally implement the curtailment action. In some embodiments, the energy load 604 is configured to provide feedback to the short-term advisor 906 if the curtailment action is not feasible, so that the short-term advisor 906 and / or the long-term planner 902 can be rerun (e.g., with additional constraints) to find a feasible solution. The energy load 604 (e.g., HVAC equipment 614, lighting devices 616, appliances 618, and computing devices 620) are thereby operated based on the output of the short-term advisor 906.

[0138] Now refer to Fig.10 , a flow chart of a process 1000 for providing a net zero energy facility is shown, according to some embodiments. The process 1000 may be performed by the facility 600, such as by operation of the system manager 502.

[0139] At step 1002, a baseline energy consumption and energy production of a facility is predicted. The baseline energy consumption β may be predicted for each of a plurality of periods t (eg, for t=1, ..., T). t The energy production γ can also be predicted for each of the multiple periods t t For example, various AI methods trained on historical data can be used to predict β t and γ t In some embodiments, the energy consumption of a facility includes the energy consumption of equipment located at or directly servicing the building (e.g., providing heating or cooling to the facility). In some embodiments, the energy consumption of a facility includes the consumption associated with the transportation of goods and / or people to and from the facility or otherwise associated with the operation of the facility, such as the energy consumption of transportation vehicles (e.g., cars, trucks, trains, airplanes, ships, etc.). In some embodiments, the use Fig.12 The process 1200 shown in FIG. 1002 is performed to perform step 1002, and with reference to the following Fig.12 In some embodiments, forecasts of baseline energy consumption and energy production are provided as or with confidence intervals, forecast ranges, forecast distributions, etc., such that the forecasts are provided with information about the uncertainty in such forecasts.

[0140] At step 1004, a first predictive optimization is performed to output a net energy trajectory {X t}, the net energy trajectory makes the first time period (X T ≤0) to achieve the predicted cost of net zero energy. The first predictive optimization can solve the problem expressed as follows:

[0141]

[0142] Among them C t is the amount of reduction at time t (expressed as a fraction or percentage in the example shown), φ t (·) is a function indicating the cost of reduction in period t, and is less than or equal to 1 and represents the maximum allowable reduction in period t. That is, φ t (C t ) represents the relative prediction baseline β t The amount of energy consumed will be reduced by C t *β t In some embodiments, φ t(·) Model or predict operating costs or penalties associated with consumption, e.g., representing tangible costs incurred by the building or its occupants for taking corresponding reduction actions (e.g., based on comfort metrics). In various embodiments, the predicted baseline value β t It can be a mean value from predictive modeling, it can be the limits of a confidence interval associated with the forecast (e.g., worst case scenario, highest value within a confidence interval), or it can be implemented as a term representing a range or distribution of values. t It may be an average value of the predictive modeling, it may be a limit of a confidence interval associated with the prediction (e.g., the worst case scenario, the highest value within the confidence interval, the value at a selected percentile within the confidence interval, etc.), or it may be implemented in various embodiments as a term representing a range or distribution of values, such as applied in step 1004. Fig.12 processing.

[0143] In other embodiments, select φ t (·) is a convex function, indicating that small cuts are exponentially easier than large cuts. For example, in some embodiments, φ t (C)=C k , k>1 (e.g., k=2). As another example, in some embodiments φ t (C) = ω t C k , where ω t is the weighting of specific time periods during which cuts are prioritized, e.g., having a lower value when cuts are sometimes more appropriate (e.g., weekends, non-business hours in a commercial office building) and a higher value when cuts are sometimes less appropriate (e.g., noon on weekdays in a commercial office building).

[0144] In some embodiments, the selection of φ t (·) to reflect the desire of some building operators to make curtailment actions sparse (infrequent, rare, etc.), so that for most values ​​of t, C t = 0. In some such embodiments, a non-convex or discontinuous cost function φ may be used t (·), for example φ t (C):=ω t 1 >0 (C)≈ω t (1-exp(-αC)), where 1 >0 (·) represents the indicator function of a positive number. t In embodiments where (·) is non-convex, process 1000 may include reformulating the problem using binary variables to be solved by mixed integer linear programming techniques and / or by discretizing the problem and applying dynamic programming.

[0145] Inequality Constraint∑ t(1-C t )β t ≤∑ t γ t The constrained forecast is optimized to achieve net zero energy consumption, i.e., consumption is greater than production. In other embodiments, particularly in scenarios where net zero consumption may not always be feasible, the problem is formulated with constraints expressed as penalties ψ(X), such as with the objective: Where X = ∑ t (1-C t )β t -∑ t γ t .

[0146] In some embodiments, the optimization problem at step 1004 is reformulated as a discrete-time stochastic control process, such as a Markov decision process (MDP). In such examples, the optimization problem solved in step 1004 can be expressed as: Make X t+1 =X t +(1-C t )β t -γ t ,X T ≤0, and Where X t is the net energy consumption at the beginning of period t (and at the end of period t-1). Constraint X T ≤0 requires that the consumption is non-positive at the end of the optimization range. In such embodiments, the solution to such a problem and the output of step 1004 is the net energy consumption X t The time series of values ​​of t}, t=1,…,T.

[0147] In some embodiments, step 1004 includes performing stochastic optimization. In such embodiments, the problem solved at step 1004 is expressed as:

[0148]

[0149] C t =κ(X t ,ξ t ,ζ t ,t)

[0150] where β t Random walk via multiplication Generate, where the Markov state is the current level ξ t A similar structure can be used with random variables ζ t Energy production of t Such statements are specific to β tand γ t Each of them has a separate random trajectory, so the overall Markov state is (X t ,ξ t ,ζ t ), and the optimal reduction strategy is some function C t =κ(X t ,ξ t ,ζ t , t). Step 1004 performs optimization to find the optimal κ(·), for example using deterministic equivalent model predictive control, dynamic programming, reinforcement learning, or direct policy optimization.

[0151] In some embodiments, energy production is also controllable and decisions related thereto can be made in the optimization of step 1004. For example, wind turbines can be shut down when not needed to reduce equipment wear, and other energy generating equipment can also be controllable to selectively produce energy or not produce energy. In such embodiments, appropriate cost terms and decision variables can be added to the cost function described herein. As another example, energy compensation credits can be purchased and included in the net energy consumption of the building, and the cost function and constraints of this article can be adapted to include appropriate terms associated with the costs and effects of purchasing energy compensation credits. Such embodiments can take into account changes in compensation prices during the year, so that if it is determined that it is necessary to meet a net energy goal (e.g., net zero energy), the optimization can determine to purchase compensation at a certain time of the year (low cost time).

[0152] In some embodiments, energy consumption is measured, modeled, calculated, and / or predicted for individual categories (e.g., HVAC, lighting, plug loads) or subspaces (e.g., buildings on a campus, floors of a building, areas of a facility) that can be individually curtailed. The formula used in step 1004 can be adjusted by adding additional exponents k∈K, where the curtailment actions are expanded to C kt (and using the baseline prediction β kt and the cost function φ kt (·)), such that the net energy trajectory generated in step 1004 is determined on a per-category or per-subspace basis and / or by taking into account differences in reduction costs across these categories or subspaces.

[0153] In some embodiments, step 1004 is adapted to provide a net energy trajectory plan over multiple overlapping windows (e.g., to achieve a net zero state over a one-year period and over a shorter period, over a one-year period starting at different dates, etc.). The formulas herein can be adapted to give a separate state X for each window with the same dynamics as described above by using an additional index w∈W for each window tw and the net energy constraint of the window depends on the end time T of the windoww To provide such functionality, for example In such embodiments, the optimization of step 1004 is solved over all such windows including the current day.

[0154] In some embodiments, for example, according to the following Fig.12 In accordance with the teachings of , step 1004 is performed using a method that takes into account uncertainty in the prediction. For example, the uncertainty modeling of process 1200 can be used to determine a value of a certain confidence interval or percentile (e.g., 90% chance of occurrence, 80% chance of occurrence, etc.), such as selecting a value that is within a selected confidence limit (e.g., a user-selectable uncertainty limit, a user-selected percentile, etc.) as the worst-case occurrence.

[0155] At step 1006, a second predictive optimization is performed to output a set of reduction actions {U kt}, which makes it possible to t+1 )) to minimize the predicted cost of the value of the net energy trajectory. In some embodiments, the optimization problem at step 1006 is expressed as Make C kt =π kt (U kt ) and∑ t (γ t -∑ k (1-C kt )β kt )≥χ, where k indicates the reduction category (building domain, energy load type) (e.g., HVAC, lighting, plug load, etc.), U kt Indicates the reduction actions for category k during period t, C kt is the reduction of category k during period t, φ kt (·) Model the cost of reduction actions for category k during period t, π kt (·) is the reduction model for the amount of reduction that will be caused by the reduction action for category k during period t, β kt is the predicted baseline energy consumption of category k during period t, γ t is the predicted energy production during period t, and χ is the net energy target for period t from step 1004. In some embodiments, χ=X t+1 -X t , i.e., the change in the net energy trajectory from step 1004 associated with period t. In other embodiments, χ=X t+1 -X 实际,t , where X 实际,tis the measured net energy consumption up to time step t (e.g., since t=1). In some embodiments, the cost of the reduction action takes into account multiple objectives (e.g., utility prices, occupant comfort, carbon emissions, pollution, air quality, etc.). In some such embodiments, the predictive optimization of step 1006 (or other control process herein) may include a process for adjusting one or more weights of the objective function to facilitate tracking the net energy trajectory over time, such as described in U.S. patent application Ser. No. 17 / 686,320, filed on March 3, 2022, the entire disclosure of which is incorporated herein by reference.

[0156] Through U kt The indicated curtailment action is the primary decision variable of the optimization problem solved in step 1006 and may indicate various actions depending on the associated category k. For example, for the HVAC category, U kt This can be a zone temperature set point or set point change (e.g., +2°, -5°, etc.). kt It is possible to indicate the lighting level or the part of the room to be illuminated. kt The amount of production reduction C kt Through the function π kt (·) specifies that the function can be a simple (e.g., nonlinear) data-driven model. Step 1006 can be formulated as a deterministic problem, for example because the scope of step 1006 is shorter than step 1004 (i.e., one sub-period t compared to a longer time span t=1,…,T. In some embodiments, step 1006 uses user preference input via a user interface, which specifies which categories of loads the user prefers to reduce (e.g., a ranking or priority list). The user interface can be updated over time as the sub-period passes, for example showing reduction options over time and / or displaying a visualization of actual performance relative to a planned net energy trajectory. In some embodiments, the reduction action can include deploying (e.g., installing, bringing online, starting up, etc.) a new device of equipment (e.g., energy storage equipment, green energy production equipment, high-efficiency HVAC equipment, etc.), where step 1006 determines the size, capacity, type, model, etc. of the new equipment to be deployed (e.g., following the various teachings described in U.S. Provisional Patent Application No. 63 / 246,177 filed on September 20, 2021, the entire disclosure of which is incorporated herein by reference).

[0157] In some embodiments, find the reduction action U ktThis can be done by walking through a building diagram of a digital twin of the facility. A digital twin can contain digital entities (e.g., data objects, software agents, etc.) that represent real-world entities such as building equipment, systems, spaces, people, time series data, or any other building-related entities. U.S. Patent Application No. 17 / 354,436 filed on June 22, 2021, U.S. Patent Application No. 17 / 134,661 filed on December 28, 2020, U.S. Patent Application No. 17 / 134,664 filed on December 28, 2020, U.S. Patent Application No. 17 / 134,671 filed on December 28, 2020, U.S. Patent Application No. 17 / 134,659 filed on December 28, 2020, U.S. Patent Application No. 17 / 134,973 filed on December 28, 2020, U.S. Patent Application No. 17 / 134,999 filed on December 28, 2020 Several examples of digital twins and frameworks / platforms that can be used to define connections (e.g., causal relationships, functional relationships, spatial relationships, etc.) between digital twins are described in detail in U.S. Patent Application Nos. 17 / 135,023 filed on December 28, 2020, 17 / 134,691 filed on December 28, 2020, 17 / 135,056 filed on December 28, 2020, 17 / 135,009 filed on December 28, 2020, 17 / 504,121 filed on October 18, 2021, and 17 / 737,873 filed on May 5, 2022. The entire disclosure of each of these patent applications is incorporated herein by reference.

[0158] For example, a digital twin building knowledge graph can be queried for twins that meet certain criteria, and actuation can be done at a granular level (e.g., at the level of a specific dimmable lighting fixture, etc.). Using a digital twin approach allows decisions to be made based on user preferences, sustainability considerations, comfort preferences, and can provide highly granular mitigation actions. kt (e.g., associated with a specific device, equipment unit, etc.) The digital twin approach can also serve to easily link the net zero algorithms described herein to different facilities, such as being able to easily install and configure the features described herein, and easily adapt when new devices, equipment, energy loads are added (or removed) from the facility.

[0159] Step 1006 is executed to output the reduction action U kt At step 1008, the reduction action U kt is achieved by reducing the action U during the sub-period ktStep 1006 may include sending a control signal to the energy load 604 so that the energy load 604 operates according to the reduction action U kt Reduction of energy consumption relative to a forecasted baseline. Fig.10 As shown, process 1000 can move from step 1008 to step 1006, so that steps 1006 and 1008 can be repeated at a period t (e.g., every day) until the final time period t=T, thereby achieving net zero energy consumption at the end of the entire optimization time period t=1,...,T. In some embodiments, process 1000 can also return from step 1008 to step 1004, so that the net energy trajectory is updated for the remainder of the time period (e.g., at each time step or other period (e.g., weekly, monthly)), thereby facilitating the achievement of net zero consumption at the end of the entire time period (at time T). Execution of process 1000 thus provides a net zero energy facility.

[0160] Now refer to Fig.10 , according to some embodiments, a set of graphs of net energy versus time are shown, illustrating an example execution of process 1000. Fig.10 In the example of FIG. 1 , step 1004 has been performed to generate a net energy trajectory as shown by line 1100 shown in the figure. The net energy trajectory from step 1004 is used in multiple iterations of step 1006, such as by including it in Fig.10 As shown in multiple figures.

[0161] In the first diagram 1101, step 1004 has been run to generate a reduction action resulting in a net energy as shown by the short term advisory line 1102. The short term advisory line 1102 starts with the actual energy consumption at the beginning of the sub-cycle and achieves (becomes equal to) the net energy trajectory (line 1100) by the end of the sub-cycle. Advantageously, the method is able to correct any previous deviations from the net energy trajectory at each short term advisory stage (e.g., at each instance of step 1006). The first diagram 1101 also shows that the short term advisory line 1102 is allowed to deviate from the net energy trajectory (line 1100) during the sub-cycle, as long as the lines converge at the end of the sub-cycle. Such flexibility can achieve savings and improve the feasibility of tracking the net energy trajectory.

[0162] The second graph 1103 shows another short term advisory line 1104 for a later sub-cycle. As with the first graph 1101, the short term advisory line 1104 for a later sub-cycle is allowed to deviate from the net energy trajectory (line 1100) during the sub-cycle, but reach the value of the net energy trajectory by the end of the sub-cycle, such lines 1100 and 1104 converging at the end of the sub-cycle. Fig.11The fourth figure 1105 shows a similar arrangement, where the short-term advisory line 1106 of the corresponding sub-cycle reaches the value of the net energy trajectory by the end of the corresponding sub-cycle, so that the short-term advisory line 1106 converges with line 1100. Fig.11 This shows Fig. 9 Cascade architecture and Fig.10 How multiple optimizations can be performed in an exemplary scenario to achieve net zero energy consumption by providing a net energy trajectory including a plurality of sub-periods of a time period, generating a set of reduction actions at each of the plurality of sub-periods predicted to achieve the net energy target for the sub-period, and implementing the set of reduction actions. Thus, building managers, owners, and other stakeholders can reliably ensure that their facilities achieve net zero energy consumption during the time period of interest.

[0163] Now refer to Fig.12 , according to some embodiments, a method for predicting baseline consumption β t and / or energy production γ t In some embodiments, process 1200 may be performed to provide Fig.10 Step 1002 of process 1000. In some embodiments, process 1200 may be performed by system manager 502.

[0164] The following description uses x t represents the quantity of interest (e.g., β t or γ t ), μ t represents the average value of the quantity of interest, and e t represents the deviation from the mean (e.g., x t -μ t ) in order to describe the baseline consumption β t or energy production γ t The quantities of interest can be modeled separately (such that x t To represent β t or γ t scalar) or modeled together (such that x t To include β t and γ t to capture the correlation between them.

[0165] In some embodiments, the overall model structure provided in process 1200 is configured to predict x to date over a time period (e.g., year to date). t The distribution of a given value, that is, x t ~f(x0,…,x t-1 ), for example using the average and confidence limits and Such embodiments enable predictions to be conditioned on the data observed to date.

[0166] In step 1202, a mean model is fitted to historical consumption or production data. The mean model can be based on any suitable time function. In some embodiments, the mean model has the following characteristics: , where φ(x) can be any static model of the x variables (e.g., a linear or shallow neural network fitted by least squares) or can be omitted, and where z i (t) is an appropriate basis function (e.g., piecewise constant, piecewise linear, Fourier, or spline) and θ i By linear regression fitting. In some embodiments, the mean model uses a finite basis of spline functions covering the time range. For example, using Ψ it As the value of the i-th basis function at time t, the model can be fitted by the linear least squares method in step 1202 as

[0167]

[0168] Among them, x t are known values ​​from the training dataset (i.e., from historical consumption and / or production data), and α i are the model coefficients (parameters) found by solving the minimization / optimization in step 1202. By using spline functions, step 1202 can ensure that the mean model varies smoothly (as expected for energy consumption and production) and can isolate x by removing those terms from the objective function. t Fitting parameter α when samples are damaged or missing i In some embodiments, the mean model fit in step 1202 may include other variables that are known or can be reasonably predicted (e.g., building occupancy schedule, weather) and may be used as additional model inputs to the fit with associated trainable parameters in step 1202. Step 1202 thus provides a mean model that is configured to predict the mean value μ of the quantity of interest as a function of time over the time range t .

[0169] In step 1204, a deviation model is fitted to the deviations in the historical consumption or production data from the mean model. Step 1204 may include determining the deviation (error) in the historical data. t :=x t -μ t and fit the model based on such deviations. In other embodiments, a multiplicative definition of error / noise is used (e.g., t:=log(x t / μ t )).

[0170] Step 1204 may be provided using a linear autoregressive model such that

[0171]

[0172] For an appropriately chosen model order N (e.g., by examining e t In such embodiments, step 1204 includes fitting the coefficients a, for example, by constrained least squares regression. n ,For example,

[0173]

[0174] where e t The value of x is t -μ t This constraint ensures that the generated autoregressive model is stable under the tolerance δ∈[0,1), thus ensuring that the generated samples remain bounded. The noise variance σ 2 can be used as the optimal value of the objective function, corresponding to the variance of the model fit. Expressed in another way, step 1204 can include solving:

[0175]

[0176] To provide an iterative model The previous error term can be used, that is, the error value: = (e0,…,e t-1 ), these error terms are available at least N steps into the prediction horizon. When starting from the beginning of the prediction horizon, step 1204 may include Initially, once known, step 1204 can use e0 by iterating the model and by randomly sampling ε t Or keep ε t = 0 to obtain the average value to obtain the complete sequence e: = (e0,…,e T-1 ). Alternatively, step 1204 may use a sequence The properties of are taken as joint normal and quantiles or confidence intervals of the resulting trajectory are calculated. In some embodiments, the Yule-Walker equation can be applied to the autoregressive coefficients a in step 1204. n and the noise variance σ 2 To obtain the covariance Σ0 or Σ. Thus, in step 1204, a deviation model is generated to characterize the error from the mean model.

[0177] In some embodiments of process 1200, the model for a particular building, facility, etc., is linearly adjusted based on models of similar buildings for which more data is available using summary statistics (e.g., different buildings from similar climates, simulation runs for a climate, buildings of similar size or with similar equipment, etc.). For example, process 1200 may include using the following formula to determine a new model (associated with the subscript "new") based on a previously fitted model (associated with the subscript "old"): mean plus standard deviation: Given a new mean μ 新 and the new standard deviation σ 新 ; and / or min / max values: Given a new minimum value and the new maximum

[0178] In some embodiments, step 1204 includes breaking down different categories of energy consumption, such as heating, cooling, and other energy consumption. In such embodiments, the overall energy consumption may be t The following linear model is used:

[0179]

[0180] in and are heating degree days and cooling degree days, is the basis function value used in the previous mean model, θ h ,θ c ,θ i is a model coefficient, and the “other” category can be further divided using various other energy consumption subdivisions (e.g., lighting, plug load, equipment or device category, etc.). In some embodiments, step 1204 includes calculating a given Life:

[0181]

[0182] Where T oa (τ) is the outdoor air temperature at time τ, is the heating threshold (e.g., 10°C = 50°F), and is the cooling threshold (e.g., 18.3°C = 65°F), T oa (τ) is the outdoor air temperature (e.g., weather data provided by a weather service). As part of step 1204, for each time period (e.g., each day), the hourly T oa Step 1204 can then use this modeling as part of a mean value model to provide a predicted energy consumption.

[0183] In step 1206, the combination of the mean model and the deviation model is converted to a Gaussian form. In some embodiments, the model is converted to a multivariate Gaussian distribution:

[0184]

[0185] x:=[x1,…,x T ]

[0186] μ:=[μ1,…,μ T ]

[0187]

[0188] Σ ij =Σ ji

[0189] The transformed model in step 1206 may provide useful properties, such as enabling the model to be accurately additively and linearly transformed, providing an exact formula for conditional predictions on data observed so far, and by providing confidence intervals that are accurately computed rather than through sampling.

[0190] For example, in embodiments where the model of energy production or consumption is decomposed (e.g., by type of production or consumption), the conversion model in step 1206 enables the models to be effectively added to provide an overall generation or production model. For example, the properties of the Gaussian model enable step 1206 to include providing an overall generation model as the sum of multiple single-category models, such as:

[0191]

[0192] This allows models to be easily combined as part of process 1200 and step 1206.

[0193] In some embodiments, step 1206 includes using the properties of the Gaussian model to calculate the net energy model where n t is the cumulative net energy, c t is the energy consumption, and g t is energy generation. This model can be completed by fitting and transforming a combination of consumption and generation models as described above, for example according to the properties, where μ i and∑ i are the individual model parameters, and L is the lower diagonal matrix of 1s.

[0194] Referring now to step 1208, the transformed combined model is used to predict a range of consumption or production at future time steps, such as a confidence interval around a mean prediction. For example, when planning for the middle of a time period (e.g., a year), step 1208 is performed by predicting energy consumption and generation conditions based on values ​​observed so far in the current time period (e.g., the current year). In some embodiments, step 1208 may include dividing the time period into two parts (e.g., before the current time and after the current time):

[0195]

[0196] μ:=[μ a ,μ b ]

[0197]

[0198] In this scenario, step 1208 may use:

[0199]

[0200] These transformations can advantageously be performed efficiently (compared to methods that do not convert the model to Gaussian form). In these examples, x a is the data observed so far, and x b |x a Predictions conditional on the data observed to date are given.

[0201] Step 1208 may use this method to use the conversion model from step 1206 to determine the confidence interval (the prediction range around the mean). For example, step 1208 may use a property whereby for a given confidence level α∈[0,1] and

[0202]

[0203] in is the inverse cumulative distribution of the chi-square distribution with n degrees of freedom (equal to the length of x). Therefore, a randomly selected x will lie within the ellipsoid region with probability α. In some embodiments, the rectangular approximation is performed as follows:

[0204]

[0205] Such a confidence interface may be an output of step 1208 and may indicate a forecast range within which values ​​of energy consumption and production are expected to fall within a desired degree of certainty (eg, 90% confidence, 80% confidence, etc.).

[0206] In some embodiments, step 1208 includes using the net energy model (e.g., net energy model) formed in step 1208. ) to predict the net energy for the remainder of a time period (e.g., the remainder of a year). Since the net energy is the accumulation of energy production and consumption at each time step in a time period, the uncertainty in the consumption and production forecasts will also accumulate. Therefore, the confidence interval of the net energy model will widen as a function of future time. The above model form can be advantageously used to provide such information in an efficient manner. The output of step 1208 can be provided to step 1004 of process 1000 and used as described above with reference thereto.

[0207] Reference now Fig.13 , a diagram showing example output of process 1200 is shown, according to some embodiments. Fig.13 A first graph 1300 is shown including daily energy consumed and generated for several days of the year, and a second graph 1302 of net energy over a year, where the data is selected so that the average gives a net zero operation over a year. The first graph 1300 includes a generation line 1304 that plots the predicted average of daily energy generation, and confidence bands 1306 around the generation line 1304 that show a confidence interval (e.g., a range of possible values) for the prediction of daily energy generation. The first graph 1300 also includes a consumption line 1308 that plots the predicted average of daily energy consumption, and confidence bands 1310 around the consumption line 1308 that show a confidence interval (e.g., a range of possible values) for the prediction of daily energy consumption. In the example shown, the confidence bands 1306, 1310 are for Fig.13 Each day of the one-year cycle shown in has a substantially constant size.

[0208] The second graph 1302 shows a net energy trajectory 1312 and a confidence band 1314 around the net energy trajectory 1312. The net energy trajectory 1312 is a graph of the difference between the generation line 1304 and the consumption line 1308. The confidence band 1314 is shown to widen over the one-year period shown, reflecting that the confidence band 1314 is an accumulation of uncertainty at each time step represented by the confidence band 1306 of the generation line 1304 and the confidence band 1310 of the consumption line 1308. The confidence band 1314 shows that if such uncertainty is not taken into account, the energy consumption and / or production plan set at the beginning of the one-year period may result in a significant deviation from the expected net energy target (e.g., zero) at the end of the annual period. Fig.13 It is shown that when providing a control strategy for net energy consumption as in process 1000, it may be advantageous to consider confidence bands 1306, 1310, and / or 1314, such as by or as part of performing repeated adjustments to the net energy plan over a yearly cycle as taught herein.

[0209] Reference now Figures 14 to 16 , according to some embodiments, a series of views in a graphical user interface 1400 are shown. The graphical user interface 1400 may be characterized as a net zero energy dashboard and may be provided as part of a building management system application, user portal, etc., such as Johnson Controls, Inc.'s OpenBlue Enterprise Manager. In some embodiments, the graphical user interface 1400 is generated and / or hosted by the system manager 502 and displayed on one or more client devices 504 ( Figure 5 ) and / or provided by the BMS controller 366, the monitoring and reporting application 422, the enterprise control application 426, or the remote systems and applications 444 and displayed on one or more client devices 448 ( Figure 4 shown).

[0210] Figures 14 to 16 The series of views shown may be displayed sequentially by time period (e.g., year) as the time period progresses. Thus, as shown, Fig.14 The user interface 1400 at a first point in time is shown. Fig.15 shows the user interface 1500 at a second point in time subsequent to the first point in time, and Fig.16 The user interface 1600 is shown at a third time point after the second time point. Figures 14 to 16 A time period of one year is shown in FIG, it is contemplated that any time period may be used (e.g., one year, two years, five years, one month, six months, one week, ten weeks, one day, fourteen days, or any other time period). Although the following description refers to a time period of one year for ease of explanation, it should be understood that the systems and methods described herein are not limited to any particular time period or duration and may be applied to time periods of any duration.

[0211] like Figures 14 to 16As shown, the graphical user interface 1400 includes a summary widget 1402, a consumption budget widget 1404, and a plan widget 1406. The summary widget 1402 shows a year-to-date summary of net energy consumption, for example, including an indication of year-to-date net energy consumption (e.g., as a percentage normalized relative to year-to-date energy generation), year-to-date energy consumption relative to budget (e.g., relative to the plan indicated by the net energy trajectory generated in process 1000), and year-to-date energy generation relative to budget (e.g., relative to the forecast of energy production from step 1002 and / or process 1200). The indication may include text-based information and graphical information, such as a bar chart style illustration showing the value indicated in the summary widget 1402. Such illustrations may be color coded to further identify status related to the plan / forecast (e.g., red indicates consumption is higher than planned, green indicates consumption is lower than planned, red indicates production is lower than planned, and green indicates production is higher than planned).

[0212] The consumption budget widget 1404 displays energy consumption on a daily basis in a first graph 1408 and on a weekly basis in a second graph 1410 (and / or on some other basis in an alternative or additional graph, such as biweekly, bimonthly, quarterly, etc.). In the example shown, the first graph 1408 includes two bar graphs for each day (e.g., each day of the past week), and the second graph 1410 includes two bar graphs for each week (e.g., each week of the past five weeks). In each case, one of such bars represents the planned consumption for the period, and the second of such bars represents the actual (e.g., measured) consumption for the period, such that a comparison between the planned consumption and the actual consumption for the various sub-periods is shown in the consumption budget widget 1404. The first graph 1408 and the second graph 1410 thus show the user the recent performance of the building system in achieving or failing to achieve the planned energy consumption budget (e.g., the budgeted output as part of the net energy trajectory plan in the process 1000). For the summary widget 1402 , elements of the first and second graphs 1408 , 1408 indicating actual consumption may be color-coded based on comparison with the budget, e.g., colored red if consumption exceeds the budget of the corresponding sub-period, and colored green if consumption is less than the budget of the corresponding sub-period.

[0213] The planning widget 1406 is shown to include a cumulative consumption graph 1412, a cumulative generation graph 1414, and a net energy trajectory graph 1416. The cumulative consumption graph 1412 plots the cumulative energy consumption for the entire year cycle from the beginning of the year to the current date until the end of the year, including a graph of the cumulative energy consumption of the annual plan, the actual cumulative energy consumption to the current time point, and the predicted cumulative energy consumption from the current time point to the end of the year. The predicted cumulative energy consumption from the current time point to the end of the year is displayed as a line based on the predicted average, and the shaded area around the line represents a confidence interval. The cumulative energy consumption value at a given time step can be defined as the sum or summary of the energy consumption at each time step within the one-year time cycle until the given time step (and in some embodiments, including the given time step). Therefore, the cumulative consumption graph 1412 shows a cumulative energy consumption value that increases monotonically over the duration of the one-year time cycle due to the energy consumption in each time step being greater than or equal to 0.

[0214] The cumulative generation graph 1414 plots the cumulative energy generation for the entire year cycle from the beginning of the year to the current date to the end of the year, including a graph of the planned cumulative energy production for the year, the actual cumulative energy production up to the current time point, and the predicted cumulative energy production from the current time point to the end of the year. The predicted cumulative energy production from the current time point to the end of the year is shown as a line based on the predicted mean and a shaded area around the line representing a confidence interval. The cumulative energy generation value at a given time step can be defined as the sum or aggregation of the energy generation at each time step within the one-year time period up to the given time step (and in some embodiments including the given time step). Therefore, the cumulative generation graph 1414 shows a cumulative generation value that increases monotonically over the duration of the one-year time period due to the generation in each time step being greater than or equal to 0.

[0215] The net energy trajectory graph 1416 plots the net energy for a one-year cycle. Specifically, the net energy trajectory graph 1416 includes an actual performance line 1418, which shows the actual net energy trajectory for a first sub-period of the one-year cycle (e.g., from the beginning of the year to the current time). The value of the actual performance line 1418 at each time point up to the current time can be calculated as the difference between (i) the value of the actual cumulative energy consumption up to that time point as shown in the cumulative consumption graph 1412 and (ii) the value of the actual cumulative energy production up to that time point as shown in the cumulative production graph 1414. The actual performance line 1418 is shown as having a first endpoint at the beginning of the one-year cycle and a second endpoint at the current time (i.e., at the end of the first sub-period). As time passes and the actual values ​​of energy production and energy consumption are observed or calculated for each time step up to the current time, the duration of the first sub-period increases and the second endpoint of the actual performance line 1418 moves forward in time. The difference or “net” between the actual energy consumption and the actual energy production at each time step within the first sub-period may be plotted in the net energy trajectory graph 1416 as the value of the actual performance line 1418 .

[0216] The net energy trajectory graph 1416 also shows a planned net energy trajectory line 1420, which shows a planned net energy trajectory over a second sub-period of the one-year time period (i.e., from the current time to the end of the one-year time period), such as generated in process 1000. The value of the planned net energy trajectory line 1420 at each time step within the second sub-period can represent the net cumulative energy (i.e., the difference between the cumulative energy consumption and the cumulative energy production) at that time step according to the energy consumption and production plan to achieve a net energy target (e.g., net zero energy) at the end of the one-year time period. The energy consumption and production plan can include a reduction or decrease in energy consumption and / or an increase in energy generation relative to a baseline energy consumption so as to achieve the net energy target at the end of the one-year period.

[0217] The net energy trajectory graph 1416 also shows a predicted net energy line 1422, which represents the predicted net energy value within the second sub-period of the one-year cycle. The value of the predicted net energy line 1422 at each time step within the second sub-period can represent the net cumulative energy (i.e., the difference between the cumulative energy consumption and the cumulative energy production) at that time step based on the predicted future energy consumption and energy production values ​​that may occur within the remaining time of the one-year time cycle. In some embodiments, the predicted net energy line 1422 represents the net energy value expected to occur without the reduction used to reach the net energy trajectory line 1420, and / or the value expected to occur given the performance and conditions to date. The net energy trajectory graph 1416 also includes a confidence region 1424, which shows the confidence interval of the prediction around the net energy line 1422, for example, so that the user can easily see a range of results that may occur within the time period shown. The planning widget 1406 thus provides the user with a simple view of the expected and predicted performance of the system for the rest of the year (or other relevant periods in other embodiments), including confidence intervals, uncertainties, etc. associated with such expectations and predictions.

[0218] over time, Fig.14 The view in the graphical user interface 1400 is shown to change to Fig.15 The summary widget 1402 is updated to show the new year-to-date data, and the consumption budget 1404 widget is updated to show a different set of data for the most recent sub-period. The plan widget 1406 is also updated so that actual data for a longer period can be represented (e.g., via the actual performance line 1418). Fig.14 In contrast, the confidence region 1424 is Fig.15 , which can be smaller, showing that as the time period progresses (i.e., as the remaining prediction horizon becomes shorter), the predictability of net consumption within the time period increases.

[0219] Fig.16 An even later time (i.e., later than Fig.15 The view in the graphical user interface 1400 of the time shown in FIG. 1400 further illustrates the progress of the user interface 1400 over a one-year period. Fig.16 In the example of , the projected net energy trajectory line 1420 is above the confidence region 1424, indicating that the facility is expected to exceed the expected execution plan to achieve zero net energy within the one-year period. Figures 14 to 16 A graphical user interface is shown that can provide a user (e.g., a building manager) with valuable insights into a facility's performance relative to a net energy target (e.g., a net zero target).

[0220] Although the above examples are primarily directed to energy consumption and production to achieve net zero energy over a time period, the teachings herein can be applied to other types of net consumption, such as resource consumption (e.g., water consumption, gas consumption, fuel cell consumption, hydrogen consumption, raw materials, goods) and consumption characterized by the amount of pollution (e.g., carbon emissions, particulate matter emissions, sound pollution, light pollution, etc.). Consumption can be offset by the production of resources (e.g., hydrogen fuel production, rainwater collection) or by other offsetting actions (e.g., capturing or sequestering carbon, filtering pollutants, recycling, etc.). Therefore, the present disclosure includes disclosure of methods for achieving net zero consumption or other target net consumption amounts of various types or combinations of consumption over a time period by implementing the various features described herein.

[0221] Reference now Fig.17 , according to some embodiments, a flow chart of process 1700 is shown. Process 1700 can be used with Fig.10 The process of 1000 or Fig.12 In some embodiments, process 1700 may be performed by system manager 502 (e.g., by one or more processors executing instructions stored on one or more computer-readable media, by a computing system programmed to perform the steps of process 1700, etc.).

[0222] At step 1702, a net energy target for building operations over a time period is determined. The time period may include a first sub-period prior to a current time and a second sub-period from the current time to the end of the time period. The net energy target may be zero (i.e., a net zero target) or may be some other value of net energy. For example, the net energy target may be selected by a user. Building operations may include various operations of building equipment (e.g., heating, ventilation, cooling, humidity control, pressure control, lighting, security, fire detection and suppression, access control, etc.) and other activities at or associated with the building (e.g., laboratory equipment, machines, appliances, devices, etc., operating in the building; vehicles traveling to and from the building; etc.).

[0223] At step 1704, a first predicted range of energy consumption for a plurality of time steps in the second sub-period is generated. For example, the first predicted range of energy consumption may be generated according to the teachings of process 1200. Step 1704 may output a range (e.g., defined by a minimum predicted value and a maximum predicted value) of energy consumption for each time step in the second sub-period (i.e., each time period between the current time and the end of the time period).

[0224] At step 1706, a second forecast range of energy production for a plurality of time steps in the second sub-period is generated. For example, the second forecast range of energy production may be generated according to the teachings of process 1200. Step 1708 may output a range (e.g., defined by a minimum forecast value and a maximum forecast value) of energy production for each time step in the second sub-period (i.e., each time period between the current time and the end of the time period).

[0225] At step 1708, a third forecast range of net energy for a plurality of time steps in the second cycle is generated. The third forecast range is based on the difference between the energy consumption and energy production represented by the first forecast range and the second forecast range. For example, the best case scenario energy production (upper limit on the second range) for a given time step can be compared with the best case scenario energy consumption (lower limit on the first range) to select the best case scenario for the net energy at that time step, while the worst case scenario energy production (lower limit on the second range) can be compared with the worst case scenario energy consumption (upper limit on the second range) to select the worst case scenario for the net energy at that time step. The resulting third forecast range can be correspondingly wider (e.g., to reflect more uncertainty) than the first forecast range and the second forecast range.

[0226] At step 1710, a strategy for building operation is provided based on the third forecast horizon and the net energy target. For example, step 1710 may include providing a strategy for building operation based on the third forecast horizon and the net energy target. Fig.10 Process 1000 generates control decisions. Step 1710 may include performing a stochastic optimization method using a sampling of values ​​from within the third prediction horizon (and / or the first prediction horizon and / or the second prediction horizon), for example, to populate scenarios used in stochastic optimization, as shown in U.S. Patent Application No. 16 / 115,290 filed on August 28, 2018 (U.S. Patent Application Publication No. 2019 / 0079473), the entire disclosure of which is incorporated herein by reference. For example, according to various embodiments, an optimization may be performed on a set of scenarios selected from the third prediction horizon to generate control decisions for building equipment and / or other strategies for building operations (e.g., building scheduling, maintenance scheduling, installation of additional equipment, building occupancy management, etc.).

[0227] At step 1712, building operations are affected and the current time step is advanced as time passes. For example, building operations may be affected by controlling building equipment according to control decisions generated in step 1710. As another example, building operations may be affected by providing a building schedule (e.g., an occupancy schedule), performing maintenance, installing or uninstalling equipment, etc., according to the output of step 1710. As step 1712 is performed, time passes such that the current time is advanced to a subsequent time step.

[0228] In step 1714, the forecast range is updated, for example, to reflect a reduction in uncertainty as time passes. For example, step 1714 may include re-running steps 1704, 1706, and 1708, where the current time step has advanced to a later point in time compared to the initial execution of those steps. Because time has passed, uncertainty associated with what may have occurred during that elapsed time has been eliminated, and actual energy consumption and production from that time may be known (e.g., measured). Therefore, the range generated in step 1714 may be smaller in magnitude (e.g., representing greater certainty or less uncertainty) than the range generated in the initial execution of steps 1704, 1706, and 1708. Over time, the range may be updated according to step 1714 so that the forecast range is reduced as time passes within the time period.

[0229] In step 1716, the strategy for building operation (e.g., as provided in step 1710) is updated based on the updated range and / or the reduction in uncertainty determined in step 1714. Step 1716 may include re-running the stochastic optimization using a set of scenarios selected from the updated range (e.g., a narrower range than the previously performed stochastic optimization), thereby reflecting the reduction in uncertainty over the second sub-period over time (and the second sub-period being shortened as the current time advances). In some embodiments, step 1716 may include taking more stringent curtailment actions (e.g., larger set point adjustments) than in step 1710 because of the reduced uncertainty that such actions will be required to achieve the net energy goal. In other scenarios, step 1716 may include reducing or eliminating curtailment actions included in the strategy provided in step 1710 in response to actual conditions that are closer to the best case scenario in which such actions may no longer be required. Building operation may then be affected according to the updated strategy, such as by operating equipment according to set points or other control decisions included in the updated strategy, or providing maintenance, scheduling, occupancy, other operations, etc. according to the updated strategy. Such a determination in step 1716 may be based on a net energy target, for example, generated in a manner that is expected to cause actual net energy consumption to converge to the net energy target at the end of the time period (e.g., as described above with reference to Fig.10 and 11 described).

[0230] The teachings herein relating to, for example, predicting the range of potential net energy (or net carbon emissions in other embodiments) according to the techniques of process 1200 can thus be used to determine and implement strategies for achieving net energy goals, such as net zero energy (or net zero carbon emissions) building operation over a period of time. Such an achievement is a technical advantage over existing building systems, which typically operate at substantially more energy consumption than production, and provides technical advantages with respect to resource usage, environmental impact, etc.

[0231] Configuration of the Exemplary Embodiment

[0232] Although the accompanying drawings show a specific order of method steps, the order of steps may be different from that depicted. Also, two or more steps may be performed simultaneously or partially simultaneously. Such variations will depend on the selected software and hardware systems and the designer's choice. All such variations are within the scope of the present disclosure. Similarly, software implementations may be accomplished with standard programming techniques with rule-based logic and other logic to accomplish various connection steps, calculation steps, processing steps, comparison steps, and decision steps.

[0233] The construction and arrangement of the system and method shown in each embodiment are illustrative only. Although only several embodiments are described in detail in the present disclosure, many modifications are possible (for example, the size, scale, structure, shape and ratio, parameter value, installation arrangement, use of materials, color, orientation, etc. of various elements). For example, the position of the element may be reversed or otherwise changed, and the property or number or position of the discrete element may be changed or changed. Therefore, all such modifications are intended to be included in the scope of the present disclosure. The order or sequence of any process or method step may be changed or reordered according to an alternative embodiment. Without departing from the scope of the present disclosure, other substitutions, modifications, changes and omissions may also be made in the design, operating conditions and arrangement of the exemplary embodiments.

Claims

1. A method for achieving a net energy target for building operation for a time period, the time period comprising a first sub-period before a current time and a second sub-period from the current time to an end of the time period, the method comprising: generating a first forecast range of energy consumption at a plurality of time steps in the second sub-period; generating a second forecast range of energy production for the plurality of time steps in the second sub-period; generating a third predicted range of net energy amounts for the plurality of time steps in the second sub-period, wherein the net energy amounts are based on a difference between the energy consumption amounts and the energy production amounts; A strategy is provided for the building operation based on the third forecast horizon and the net energy target.

2. The method of claim 1 , comprising providing a graphical user interface comprising a net energy graph, the net energy graph comprising a first line showing actual net energy within the first sub-period, a second line showing planned net energy within the second sub-period, and an area of ​​the third forecast range based on the second sub-period.

3. The method according to claim 1, further comprising: Fitting a mean model to historical energy consumption data; fitting a bias model to the error in the output of the mean model; as well as Converting the combination of the mean model and the deviation model into a Gaussian model; The generating of the first prediction range of energy consumption for the plurality of time steps in the time period is performed using the Gaussian model.

4. The method according to claim 1, further comprising: Fitting a mean model to historical energy production data; fitting a bias model to the error in the output of the mean model; as well as converting the combination of the mean model and the deviation model into a Gaussian model; and Wherein generating the second prediction horizon of energy production for the plurality of time steps in the time period is performed using the Gaussian model. 5 . The method of claim 1 , wherein the first prediction range and the second prediction range are associated with confidence intervals of a prediction model used to predict energy consumption and energy production.

6. The method of claim 1, wherein providing the policy comprises: generating a net energy trajectory including net energy targets for the plurality of time steps based on the first forecast horizon and the second forecast horizon, wherein each net energy target indicates a target difference between cumulative energy consumption and cumulative energy production or an offset from a start of the time period to a corresponding time step in the plurality of time steps; generating, for a given time step, a set of reduction actions predicted to achieve a net consumption target for the given time step; as well as The set of reduction actions is implemented. The method of claim 6 , wherein generating the net energy trajectory comprises performing an optimization constrained by the net energy target.

8. The method of claim 1, wherein providing the strategy comprises controlling building equipment serving a facility, the energy consumption corresponds at least in part to operation of the building equipment, and the energy production corresponds to green energy production at the facility. 9 . The method according to claim 1 , further comprising decomposing the first predicted range of energy consumption into energy consumption categories, wherein the energy consumption categories include heating consumption, cooling consumption, and other consumption.

10. The method of claim 1, wherein the strategy is configured to drive the net energy amount of a final time step in the second sub-period to the net energy target.

11. The method of claim 1, wherein the net energy amount at a given time step in the plurality of time steps is a cumulative difference between the energy consumption amount and the energy production amount during the time period up to the given time step.

12. A system comprising: an energy load operable to consume energy; a green energy source configured to produce energy; as well as A processing circuit, the processing circuit being programmed to: generating a first forecast range of energy consumption of the energy load at a plurality of time steps in a second sub-period; generating a second forecast range of energy production of the green energy source at a plurality of time steps in the second sub-period; generating a third predicted range of net energy amounts for the plurality of time steps in the second sub-period, wherein the net energy amounts are based on a difference between the energy consumption amounts and the energy production amounts; as well as The energy loads are controlled using a control strategy configured to drive one or more of the net energy amounts to a target.

13. The system of claim 12, wherein the processing circuit is further programmed to host a graphical user interface, the graphical user interface comprising a net energy graph, the net energy graph comprising a first line showing actual net energy within a first sub-period, a second line showing planned net energy within the second sub-period, and an area of ​​the third forecast range based on the second sub-period.

14. The system of claim 12, wherein the processing circuit is further programmed to: Fitting a mean model to historical energy consumption data; fitting a bias model to the error in the output of the mean model; and Converting the combination of the mean model and the deviation model into a Gaussian model; The processing circuit is programmed to use the Gaussian model to generate the first predicted range of energy consumption for the plurality of time steps in the time period.

15. The system of claim 12, wherein the processing circuit is programmed to execute the control strategy by: generating a net consumption trajectory including net consumption targets for the plurality of time steps based on the first forecast horizon and the second forecast horizon, wherein each net consumption target indicates a target difference or offset between total consumption and total production from the start of the time period to the corresponding time step; generating, for a given time step of the plurality of time steps, a set of reduction actions predicted to achieve the net consumption target for the sub-period; and The set of curtailment actions are implemented by controlling the energy load.

16. The system of claim 15, wherein the processing circuit is programmed to generate the net consumption trajectory by performing an optimization subject to the objective constraints.

17. The system of claim 15, wherein the processing circuit is programmed to generate the set of curtailment actions based on a breakdown of energy usage types of the energy loads, the energy usage types comprising heating and cooling.

18. One or more non-transitory computer-readable media storing program instructions that, when executed by one or more processors, cause the one or more processors to perform operations comprising: generating a first forecast range of energy consumption or carbon emissions for a plurality of time steps in a time period; generating a second forecast horizon of energy production or carbon capture for a plurality of time steps in a time period; A control strategy is provided based on the first predicted range of energy consumption or carbon emissions and the second predicted range of energy production or carbon capture, the control strategy being configured to drive a cumulative difference between energy production or carbon capture and energy consumption or carbon emissions over the time period to a target.

19. The one or more non-transitory computer-readable media of claim 16, wherein the operations further comprise: Fitting mean models to historical energy consumption or carbon emissions data; fitting a bias model to the error in the output of the mean model; as well as Converting the combination of the mean model and the deviation model into a Gaussian model; Wherein generating the first prediction horizon for the plurality of time steps in the time period is performed using the Gaussian model.

20. The one or more non-transitory computer-readable media of claim 16, wherein providing the control strategy comprises implementing abatement actions based on the target.

Images