Edge device appliance network and system and method for energy consumption control and conservation

Abstract

A method for reducing energy consumption and / or demand of an edge device appliance is provided that involves receiving regulation data related to a thermostat or other control output for an edge device appliance for a regulated space or other load, processing the regulation data using artificial intelligence (AI) software to create an optimized thermostat or other control signal for the edge device appliance in the regulated space or other load, and controlling the edge device appliance for the regulated space or other load based on the optimized thermostat or other control signal. The method can also include receiving an energy consumption and / or demand modification command that includes modification data, such as a demand reduction command value, or retrieval of activation modification data, such as a demand reduction command value. Power consumption data can be received from sensors to determine real-time energy consumption and demand data for the edge device appliance. An energy consumption and / or demand modification signal can be generated using the AI software and used to control the edge device appliance.

Description

[0001] Cross-reference of related applications

[0002] This application claims the benefit of U.S. Provisional Patent Application No. 63 / 616,863, filed January 2, 2024, pursuant to 35 USC §119(e), the entire contents of which are incorporated herein by reference. Background Technology

[0003] Due to load growth, particularly from building and transport electrification and from data center operations, power grids around the world are currently reaching capacity limits. Furthermore, peak daily electricity load usage often necessitates grid operators to generate more electricity with higher CO2e emissions. Smarter, more bilateral, and real-time approaches are needed to match electricity supply and demand. Three common approaches are currently in use and under development: (1) permanent demand reduction through improved, more energy-efficient electrical equipment design and retrofitting of existing equipment; (2) load shedding, also known as demand response, which involves reducing load based on signals generated by grid operators; and (3) load shifting, shifting from periods of high demand and high cost to periods of low demand and low cost. Additionally, natural gas consumption has increased worldwide as part of total energy consumption for power generation, space heating, and industrial purposes. Therefore, similar to the approaches described in (1)–(3) above, there is a growing need for permanent demand reduction, as well as load shedding and shifting, in natural gas distribution networks.

[0004] Heating, ventilation, air conditioning, and refrigeration (also known as HVACR and HVAC&R) cooling and heating loads are candidates for demand response and load shifting. In demand response scenarios, HVACR units may not operate at full design capacity during demand response events, and in load shifting scenarios, weather and electricity price signals can be used to shift HVACR power demand to times of day when demand is lower or costs are lower. Mechanisms for achieving HVACR load shifting include pre-cooling (e.g., cooling a building or other load overnight before a hot day) and pre-heating (e.g., heating a building overnight before morning peak demand).

[0005] Various thermostatic-based systems have been proposed to achieve HVACR or other load demand response and load transfer. For power demand response, most thermostatic-based systems manipulate a constant setpoint, thus leading to potential discomfort for building occupants. Furthermore, current technologies for load transfer (such as preheating and precooling buildings) also typically rely on setpoint variations that can also affect occupant comfort.

[0006] Given the enormous impact of leaked air conditioning and refrigerant on CO2e and the projected rapid growth in air conditioning installations worldwide over the next few decades, refrigerant monitoring and management are key elements in mitigating climate change, and are currently poorly managed at individual HVACR levels, from residential to large commercial and industrial HVACR units.

[0007] Therefore, it is desirable to provide original and / or retrofittable smart technologies that provide demand reduction, electrical load reduction, and / or load transfer for the electrical appliances of HVACR systems without the loss of comfort associated with other current methods, and with additional beneficial features as described herein. Summary of the Invention

[0008] One feature of this invention is a method for automatically managing and providing energy and demand control, as well as energy savings and demand reduction estimates, to operate HVACRs and other load devices in an improved manner compared to operation using primitive load control (duty-cycled). As an example of other load devices, it enables the management of compressed air storage units.

[0009] Another feature of the invention is to provide an embodiment of an electronic controller that can be used as an add-on (e.g., retrofit) device in HVACR and other load systems, which automatically manages and provides energy and demand control for operating HVACR or other equipment, as well as energy savings and demand reduction estimates, based on AI-based commands.

[0010] Another feature of the invention is to provide a system that includes the indicated controller to automatically manage and provide energy and demand control and savings estimates for HVACR equipment operating in the system.

[0011] Additional features and advantages of the invention will be set forth in part in the description which follows, and in part will be apparent from the description, or may be learned by practice of the invention. The objects and other advantages of the invention will be realized and obtained by means of the elements and combinations particularly pointed out in the specification and the appended claims.

[0012] To achieve these and other advantages, and in accordance with the purposes of this invention, as embodied and summarized herein, the present invention relates to a method for automatically controlling and managing the energy consumption, demand, and operation of at least one load unit powered by electricity from an HVACR or other system to obtain selected levels of energy savings and demand reduction, and other benefits as described herein. The method may include the step of receiving conditioning data relating to the thermostatic output of edge device appliances in a regulated space, the conditioning data including one or more of control inputs, state variable inputs, and sensor inputs. The method may include processing the conditioning data using artificial intelligence (AI) software to create an optimized thermostatic control signal for edge device appliances in the regulated space. The method may include controlling edge device appliances in the regulated space based on the optimized thermostatic control signal. The method may include receiving an energy consumption or demand modification command that includes or activates the retrieval of modification data, the modification data including a demand reduction command value, a time-of-use energy rate for the day, weather information, or a combination thereof. The method may include receiving power consumption data from sensors to determine real-time energy consumption and demand data for edge device appliances in the regulated space. This method may include using AI software to generate energy consumption and / or demand reduction modification signals based on regulation data, modification data, and real-time energy consumption and demand data. The method may also include controlling edge device appliances in the regulated space based on the energy consumption modification signals, wherein the energy consumption and / or demand modification signals include timing modifications to component commands of HVACR or other systems to achieve energy consumption and / or demand reduction.

[0013] Also provided are raw, temperature-controlled, and edge node devices, networks, and systems for performing the methods.

[0014] The present invention also relates to a non-transitory computer-readable storage medium that stores instructions which, when executed by a computer, cause the computer to perform the indicated method.

[0015] It should be understood that the foregoing general description and the following detailed description are merely exemplary and explanatory, and are intended to provide further explanation of the claimed invention.

[0016] Some features of the invention are illustrated in conjunction with the accompanying drawings, which are incorporated in and constitute a part of this application, and together with the description, serve to explain the principles of the invention. Attached Figure Description

[0017] Figure 1A This is a block diagram / schematic representation of an HVACR system including an electronic edge controller, according to an example of the present invention.

[0018] Figure 1B Is included Figure 1AThe diagram shows a process block diagram of a microcontroller in an electronic controller, and schematically illustrates a microprocessor for storing and executing the instructed controller program, as well as for performing data collection functions, generating signals to control one or more load devices, and calculating estimated energy consumption and demand, energy savings, and demand reduction.

[0019] Figure 1C yes Figure 1A A schematic diagram of a primary and secondary relay connected in series, as shown in the electronic controller.

[0020] Figure 2 This is a flowchart depicting a method according to an embodiment of the present invention, wherein the method includes receiving conditioning data, processing the conditioning data using AI software to create an optimized thermostatic control signal, and estimating the baseline energy consumption of the appliance.

[0021] Figure 3 This is a schematic diagram depicting a network including edge node devices according to the present invention.

[0022] Figure 4 This is a circuit diagram showing the series connection of the main relay and the secondary relay in the electronic controller according to the present invention.

[0023] Figure 5 This is a table illustrating operations that can be implemented by the system of the present invention, and exemplifies various operations for controlling a single control channel for HVACR or other electrical appliances, wherein the terms “PDR PACE AI” and “ADR PACE AI” in the exemplary circuit diagrams shown refer to various forms of AI control algorithms described and defined herein.

[0024] Figure 6 This is a schematic diagram of an exemplary embodiment of an AI-enabled thermostat for a 4-channel HVAC unit that includes 2-stage cooling and 2-stage heating.

[0025] Figure 7 This is a flowchart depicting a method according to an embodiment of the present invention, wherein the method includes receiving regulation data for an energy consumption and energy generation system, processing the regulation data using AI software to create optimized control signals, and estimating the system's energy consumption, demand, and supply.

[0026] Figure 8 This is a flowchart depicting a method according to an embodiment of the present invention, wherein the method includes receiving conditioning data for a refrigeration system, processing the conditioning data using AI software to create an optimized isothermal control signal, and estimating the baseline energy consumption of the refrigeration system.

[0027] Figure 9This is a flowchart depicting a method according to an embodiment of the present invention, wherein the method includes receiving conditioning data of the entire HVACR system, processing the conditioning data using AI software to create an optimized isothermal control signal, and estimating the baseline energy consumption of the system.

[0028] Figure 10 This is a process block diagram of the process control logic of an energy control algorithm according to an example of the present invention, which can be used by an electronic controller for automatic energy control and energy saving estimation of an HVACR system.

[0029] Figure 11 This is a process block diagram of process control logic for determining a general hysteresis thermostat signal that can be used for heating or cooling, according to an example of the present invention.

[0030] Figure 12 This is a process block diagram of an example of process control logic according to the present invention, which applies a time delay to the input signal, i.e., using Figure 1B The process determines the hysteresis signal to generate the control signal (“uPace”) used in the process shown in Figure 1.

[0031] Figure 13 This is a process block diagram of the process control logic used for energy calculation (“calculating energy”) in the process shown in Figure 1, according to an example of the present invention.

[0032] Figure 14 Figure 1 is a process block diagram of the process control logic of the "Plant Model" according to an example of the present invention, which models the dynamics of a forced air heating, cooling or refrigeration system under the control of a hysteresis thermostat.

[0033] Figure 15 This is an example of the off (OFF) state according to the invention. Figure 14 A flowchart of the object model process.

[0034] Figure 16 An example of a calculation scheme for parameter A during the off state, according to the present invention, is shown.

[0035] Figure 17 This is an example of the invention during the ON state. Figure 14 A flowchart of the object model process.

[0036] Figure 18 An example of a calculation scheme for calculating the B parameter according to the present invention is shown.

[0037] Figure 19 A diagram illustrating the use of an extended Kalman filter (EKF) to estimate parameters A and B according to an example of the present invention is shown.

[0038] Figure 20 This is a flowchart illustrating the process of using an extended Kalman filter (EKF) as an estimation method according to an example of the present invention.

[0039] Figure 21 This is an electrical connection diagram of a single-stage cooling application using an electronic controller, according to an example of the invention, wherein the configuration is shown as used when a single thermostat is used to control an HVAC cooling unit (e.g., a compressor). Detailed Implementation

[0040] According to the present invention, a method for reducing the energy consumption and demand of an edge device electrical network is provided. The edge device electrical network may be embodied in 1) circuitry and software in the original equipment, including 2) a thermostat, and 3) additional edge devices, and may include one or more electrical devices selected from thermostats for HVACR equipment, humidity control devices for HVACR equipment, control devices for power generation equipment, and similar controls for edge device electrical equipment.

[0041] This method may include receiving conditioning data relating to the thermostat or other control outputs of edge device appliances used for HVACR conditioning of a space or other load. The conditioning data may include one or more of, for example, control inputs, state variable inputs, and sensor inputs. The method may include processing the conditioning data using artificial intelligence (AI) software to create an optimized thermostat control signal for edge device appliances in the conditioned space. The edge device appliances in the conditioned space may be controlled based on the optimized thermostat control signal. Regardless of whether the invention (PACE AI system) is online or offline, where "offline" for HVACR equipment means under OEM thermostat control, the PACE AI system adaptively estimates the heat capacity, thermal resistance, and internal load of the conditioned space. Using these estimates, an object model is run using measured measurement data and the estimated temperature setpoint. The temperature setpoint and conditioned space temperature generated by the object model are applied to a hysteresis temperature controller model to create the estimated thermostat signal. Energy savings are estimated based on the integral difference between the estimated thermostat signal generated by the object model and the measured PACE AI system control signal.

[0042] This method may include receiving energy consumption and / or demand reduction modification commands, for example, from grid operators, emergency broadcast systems, local utilities, emergency agencies, or police agencies. The method may also include a system for generating continuous energy consumption and / or demand modification control inputs for continuous optimization of energy and demand. Energy consumption and / or demand reduction modification commands may include commands for retrieving modification data. Modification data may include demand reduction command values, time-of-use energy rates for the day, current weather information, forecast weather information, or combinations thereof. The energy consumption modification command may activate the retrieval of modification data. The retrieved modification data may include demand reduction command values, time-of-use energy rates for the day, current weather information, forecast weather information, or combinations thereof.

[0043] This method may include receiving power consumption data from sensors to determine real-time energy consumption and demand data for edge device appliances in an HVACR-regulated space or other load. The method may include generating energy consumption and demand modification signals using AI software. These signals may be based on regulation data, modification data, and real-time energy consumption and demand data. The method may also include controlling edge device appliances in an HVACR-regulated space or other load based on the energy consumption and demand modification signals. These signals may include modifications or settings to the timing of ON / OFF commands to reduce the energy consumption of the appliances.

[0044] The method may further include a sensor output that sends an optimized temperature control signal from an edge device electrical component. The sensor output may be an output of the optimized temperature control signal. The sensor output may be continuously sent from the edge device electrical component.

[0045] The method may also include estimating baseline energy consumption data for edge device appliances in the regulated space. This estimation may be based on regulation data. The energy consumption modification signal may be further adjusted based on or according to the estimated baseline energy consumption data.

[0046] This invention also provides an edge device connected to an internet-connected or building automation system that utilizes edge-based and cloud-based artificial intelligence and machine learning (AI / ML) to optimize the energy consumption of appliances. This invention also provides a control system including the edge device. The edge device or appliance may include, for example, temperature control and other control devices for heating, ventilation, air conditioning, and refrigeration (HVACR) equipment, control devices for power generation equipment, control devices for similar appliances, or appliances including control devices. Systems of one or more different appliances can be controlled by the edge device and control system of this invention. AI / ML can be used for permanent and mandatory reduction of electricity and / or natural gas or other fuel demand. The control system can work with one or any number of edge devices and can be integrated with existing control and remote monitoring systems. For example, the control system can be integrated into existing or new building automation systems, existing or new industrial process control systems, existing and new electricity and natural gas transmission and distribution networks, and existing or new solar power appliances.

[0047] At least one edge device of the present invention may include a microcontroller, two or more control relays, multiple input and output connection options for sensors and other input / output (I / O) variables, edge and cloud artificial intelligence / machine learning (AI / ML) software, and other software including firmware. Each control relay may be a dual-state device with an open and a closed state, or may have more than one operating state or an operational state. The control relays may be mounted in series with existing cooling and heating thermostats or other control lines. Each control relay is actuated by the AI / ML software, which controls the state of each relay (e.g., open or closed). The AI / ML software may also control the duration of any state of each control relay. The number of control relays in a single edge device can be set to any number, limited only by installation space requirements. The edge device may be battery-powered, AC-powered, or a combination thereof. In addition to control relays, the edge device may also have a number of other control output options.

[0048] Edge devices can be used to control one or more electrical appliances. Appliances controlled by edge devices can be, or include, for example, cooling equipment, heating equipment, ventilation equipment, refrigeration equipment, solar or other power generation equipment, compressed air equipment, or combinations thereof. Appliances can be, or include, for example, heaters, furnaces, ventilation fans, air conditioning units, refrigerators, other HVACR equipment, solar generators, thermoelectric generators, geothermal generators, other generators, etc. AI / ML software can automatically and continuously adjust the thermal capacity and thermal resistance parameters of the dynamic model of the appliance (e.g., cooling or heating equipment) residing within the AI / ML software. AI / ML software can utilize various sensors and other input signals, which can include supply air or fluid and return air or fluid temperature, pressure or humidity, amperes, voltage, proxy signals, other performance and control variables, and combinations thereof.

[0049] This invention also provides a system for edge devices that utilizes a cloud-based learning database of curated data. Each edge device can send and receive data in real time from a cloud-based system data integration and analysis engine and / or the cloud, remotely, or a cloud and remote command center. The control system can receive location, weather and weather forecast information, and other information in real time from various internet and database sources. Control, state variables, sensor, and other inputs can be processed by AI / ML software as training data to create optimized outputs through normal control sequences of operations of one or more connected machines. The outputs can represent additional sources of machine learning for a range of applications. Dynamic AI / ML modeling is continuously updated, for example, to generate optimized thermostatic control signals for heating, cooling, and refrigeration (HVACR) equipment and solar power generation equipment.

[0050] The present invention also provides an edge node device for controlling at least one edge appliance in an HVACR system. The edge node device can be configured to send control signals to at least one edge appliance. The at least one edge device may include, for example, a thermostat, a power meter, etc. The edge node device may include a controller and an output configured to communicate with at least one edge appliance. The controller may include a main control relay and a secondary control relay arranged in series with the main control relay. The main control relay may include a first artificial intelligence (AI) command input, wherein commands from an AI cloud and a command center can be input to the main control relay. The secondary control relay may include a second AI command input. Both the first and second AI command inputs can be configured to communicate with the AI ​​cloud and the command center to receive the first AI command signal and the second AI command signal, respectively.

[0051] The main control relay can have both an open and closed state. The secondary control relay can also have both an open and closed state. The output can be configured to communicate with at least one edge appliance and to send control signals from the controller to at least one edge appliance. The main control relay can operate as a two-state relay, i.e., closed or open. When closed, the main control relay can transmit a signal generated by the thermostat as it varies between 24VAC and 0VAC. The 24VAC signal voltage can typically be used for a call to cooling or heating in the corresponding control channel. If the call is met, a 0VAC signal voltage can be sent, thus stopping the control signal. When open, the main control relay switches to its second state, i.e., energizing a continuous 24VAC voltage signal (i.e., a continuous "call" signal).

[0052] The edge node device can be configured such that: when the main control relay is open, a continuous ON command signal is transmitted to the first control relay via the main control relay, and the output of the main control relay is used as an AI command signal. The edge node device can be configured such that: when the main control relay is closed, the output of the main control relay is a thermostat command signal generated by a thermostat, power meter, etc. The edge node device can be configured such that: when the first control relay is closed, the output of the main control relay is sent to the AI ​​command center for processing by one or more AI control algorithms to form a processed signal, which is then transmitted as a control signal to at least one edge device. The edge node device can be configured such that: when the first control relay is open, regardless of the state of the main control relay, an OFF command signal is generated as a control signal.

[0053] The present invention also provides a network comprising edge node devices as described herein and an AI command center. The AI ​​command center can be configured to generate a first AI command signal, a second AI command signal, or both. One or more different AI command centers can be used to generate the first AI command signal, the second AI command signal, or both.

[0054] AI command centers can be located remotely to edge nodes, locally, or in a combination of remote and local processing. AI command centers can communicate with controllers via wired connections. AI command centers can be cloud-based. AI command centers can be located remotely to edge nodes but can communicate with them via cloud-based computing and communications.

[0055] Edge node devices can be configured to send multiple control signals to various edge appliances. Each edge appliance may include, for example, a thermostat, a power meter, a humidifier, etc. The controller may include a main control relay and multiple secondary control relays. Each secondary control relay may be connected in series with a main control relay. Each secondary control relay may include a corresponding second AI command input. Each second AI command input can be configured to communicate with an AI command center to receive a corresponding second AI command signal. Each secondary control relay may have an open state and a closed state.

[0056] The system of this invention can utilize hybrid edge and cloud technologies, whereby some processing occurs at edge node devices while some processing occurs in the cloud, for example, at a remote AI command center. Continuous cellular or other Internet of Things (IoT) cloud communication and data uploads are possible.

[0057] The controller may have multiple corresponding outputs, each configured to communicate with a corresponding edge appliance among multiple edge appliances. The controller may be configured to send corresponding control signals from the controller to the corresponding edge appliance. The edge node device may be configured such that, for each corresponding edge appliance among the multiple edge appliances, one or more of the following rules apply: Rule 1: When the main control relay is open, a continuous ON command signal is transmitted through the main control relay to the corresponding secondary control relay, and the output of the main control relay is used as an AI command signal. Rule 2: When the main control relay is closed, the output of the main control relay is a thermostat or instrument command signal generated by the thermostat or instrument. Rule 3: When the corresponding secondary control relay is closed, the output of the main control relay is sent to the AI ​​command center for processing by one or more AI control algorithms to form a processed signal, which is then transmitted as a control signal to the corresponding edge device. Rule 4: When the corresponding secondary control relay is open, regardless of the state of the main control relay, an OFF command signal is generated as a control signal. The present invention also provides a network comprising edge node devices and an AI command center as described herein. The AI ​​command center can be configured to generate a first AI command signal, a second AI command signal, or both.

[0058] The present invention also provides a system comprising an edge node device as described herein, an edge appliance (such as an HVACR device) as described herein, a current sensor, and an air duct temperature sensor. The current sensor can be configured to send a sensed current signal to a controller, the sensed current signal indicating the current load being applied by the appliance. The air duct temperature sensor can be configured to send a sensed air duct temperature signal to the controller, the sensed air duct temperature signal indicating the air temperature in a duct system through which the HVACR appliance moves air or liquid. The system may include a conduit temperature sensor configured to send a sensed liquid temperature signal to the controller, the sensed liquid temperature signal indicating liquid flow through a conduit connected to the HVACR appliance. The controller can be configured to send data relating to the sensed current signal, the sensed air duct temperature signal, the sensed liquid temperature signal, etc., to an AI command center. The system may also include an AI command center.

[0059] The system can also include a cloud-based learning database of curated data. The AI ​​command center can communicate with and be configured to retrieve data from this cloud-based learning database.

[0060] The control system of this invention utilizes AI system control inputs, state variable inputs, sensor inputs, and other inputs to generate estimates of device and system baselines, such as baselines representing non-optimal operation, including baseline energy consumption and demand. The AI / ML software can accept scalable demand reduction command values ​​from any source (e.g., from a grid operator or a grid demand response service provider). The AI / ML software can process the demand reduction command and use it, along with baseline energy consumption, to generate a control relay signal for the desired percentage demand reduction. Sensor data from current sensors, voltage sensors, or other inputs is converted by the control system of this invention into real-time current, kilowatts, kilowatt-hours, heat, British thermal units (BTU), BTU / hour, and other consumption data. This consumption data can be used to determine the desired percentage demand reduction.

[0061] The control system can generate control commands that can produce continuous and permanent energy consumption reductions for connected edge appliances or systems of appliances. Part of the energy reduction or savings can vary proportionally to estimated time-varying capacitance and resistance values ​​for a given device and load system. Another part of the savings can be based on variable-based algorithms using weather and weather forecast data. Yet another part of the savings can be based on scheduling-oriented optimization and the addition of optimization rules. Optimization rules could be, for example, that if the sensed return air or fluid temperature is within a specific range, such as (>) (<) X, then the HVACR unit will not operate in vapor compression air conditioning or heating mode. In this way, the control system can be used to achieve the purpose of operating the HVAC unit energy saver without incurring the cost of the energy saver.

[0062] The control system and its AI / ML software of this invention can provide building operators or industrial process control operators with a customizable set of variables, in which user-controllable "sliders" can be applied. The slider can provide greater energy savings and / or demand reduction at one end, and more operation as a characteristic of native control architecture operation at the other end. Therefore, an adjustable slider can provide operators with greater assurance of "normal" comfort while ensuring the achievement of performance variables.

[0063] The AI / ML software can use OEM recommendations and a database of sensed data to calculate optimized hourly startup times for HVACR and other equipment. The AI / ML software can provide limits on equipment startup to ensure compliance with equipment manufacturer specifications. It can implement anti-short-cycle and other machine protection mechanisms. The control system of this invention can also use the aforementioned variables to detect and correct anomalies in HVACR and processing unit operation that exceed anti-short-cycle and maximum hourly startup limits.

[0064] The control system can incorporate numerous fail-safe functions, including shutting down all control relays or other control outputs and relinquishing control to normal (non-optimized) device inputs in the event of a power failure at the edge device or, in some cases, at a cloud-based data analytics and command center. The control system can detect and provide alerts about potential tampering with both the control system and its attached devices, for example, by issuing statements or displaying messages such as "connected, not following commands, possibly not wired." The control system can detect, alert on, and log voltage and frequency drops in the mains and microgrids, as well as other power quality issues, which can serve as documentation in the event of equipment damage caused by such problems.

[0065] The control system can utilize the aforementioned inputs to optimize pre-cooling or pre-heating of the building. For training and input / output data, the control system can use the output (PACE_CH) The signal continues to cool beyond the input OEM (OEM_CH). The system receives the signal until a specific return air temperature setpoint is reached. The control system can also improve the performance of HVACR systems that are either too large or too small for their connected cooling or heating loads, and where the native control architecture provides suboptimal energy efficiency.

[0066] The control system can be configured to work with OpenADR 2.0 and OpenADR 3.0 demand reduction and price signal inputs, the latter including time of use (TOU) and forecasted electricity price signals. The control system can use these inputs to dynamically reduce HVACR operations in a powerful manner and can be used as an easy-to-deploy, low-cost OpenADR virtual end node (VEN).

[0067] In addition to the dynamically optimized "intelligent precooling" and "intelligent preheating" as described above, the control system can also use, for example, current sensors and / or direct real-time OEM_CH. Input is used to perform 24 / 7 / 365 enhanced energy consumption and demand reduction, where training routines will involve, for example: TRAINING DATA: PACE_CH ::TEMP MIN (kWh). Similar applications will use dynamically sensed HVACR air conditioning and heating and return air temperatures or supply and return water circulation to heat water, increasing the supply temperature while maintaining the return temperature, for energy consumption and demand reduction at the individual, mesh building, or microgrid level. In doing so, the control system can utilize PACE_CH in machine learning. TEMP It can be used in conjunction with other data for machine learning. For example, an example of this use is provided at blog.research.google / 2023 / 12 / advancements-in-machine-learning-for.html.

[0068] The control system edge device can be integrated into a "virtual inverter" driver for optimized motor operation, which does not require or need the power-side connection of the inverter driver required in current practice.

[0069] There is a significant unmet need for tenant-side solutions to reduce building energy consumption, energy costs, and carbon emissions. Meanwhile, building owners lack a consistent incentive to provide these reductions when tenants bear the energy costs, as this incurs additional expenses. Building owners are even less willing to provide such measures because their revenue is threatened by the continued decline in work-from-home and brick-and-mortar retail business. Control systems can be combined and integrated with other control and monitoring measures, including circuit breaker panel monitoring and energy storage, to create portable building automation systems—for example, systems that office and retail building tenants can install and take with them when they vacate their premises.

[0070] The control system can utilize its detailed input data stream to generate machine alarms and diagnostic reports, including, for example, work tickets, machine repair, preventative and predictive maintenance, and can also remedy certain machine problems. For example, in the case of an HVACR unit that has erroneously started due to an electrical short circuit in one of its compressors, the control system can use its detailed input data stream to generate machine alarms and diagnostic reports related to the erroneous start. The invention also provides a machine alarm and diagnostic reporting service, for example, bundled with a subscription or priced per event. Annual credit carry-over can be provided.

[0071] The control system can be used in conjunction with photovoltaic (PV) solar cell arrays. The control system can utilize real-time utility interval meter or PV solar inverter signals, or other signals, including lumen detection of cloud cover over the PV solar cell array, to dynamically reduce HVAC building energy consumption, thereby maintaining higher reliability of dispatchable net power generation from the PV solar cell array to the grid.

[0072] For electrical appliances or machines that utilize natural gas or other flammable materials, the control system can incorporate easily added intelligent emergency shutdown at the machine unit (either inside or outside the building). Shutoff can provide a significant risk mitigation supplement to such appliances, particularly in response to the risks of fire, earthquakes, tornadoes, or industrial accidents.

[0073] The control system can be provided as an open-source AI product and service package, similar to the example described at www.cbinsights.com / research / open-source-ai-development-market-map / ?utm_source=CB+Insights+Newsletter&utm_campaign=73a0a645f0-newsletter_general_sat_2023_11_04&utm_medium=email&utm_term=0_9dc0513989-73a0a645f0-87700161.

[0074] The control system can provide automatic detection and correction when the connection is lost, with alarms, as well as enhanced reset and restore connection software.

[0075] To address cybersecurity challenges in both power grid operations and building systems, control systems can provide monitoring and alerting. These control systems can include hardened cloud cellular connectivity beyond standard wired networks and potentially vulnerable firewalls. Through cloud cellular connectivity, control systems can provide monitoring and alerting to improve resilience against adversarial attacks, particularly enhancing the resilience of wired networks against such attacks.

[0076] Control systems can include two or more implementation modes to maximize the adoption of the technology's ability to reduce carbon emissions and costs, particularly in areas where internet connectivity may be problematic. For example, two modes (basic and professional) could be offered for small HVACR units (such as small apartment systems) rather than large commercial units that can easily connect to the internet. In the "basic" mode, the control system can be disconnected or periodically connected, can learn locally, can use open-source and / or proprietary software, can provide savings warranties and other insurtech and PdM services, and can periodically receive over-the-air (OTA) software updates, as well as periodically or continuously receive edge device AI / ML and other updates. In the "professional" mode, the control system can be continuously connected to receive periodic OTAs but can provide greater control and functionality, as well as continuous monitoring and learning, while also providing savings warranties and extended insurtech and PdM services.

[0077] The control system can be integrated with internet-connected thermostats, for example, with an additional wireless edge module at the HVACR controller, and / or with building automation systems. The edge device of the control system can be designed to function as a smart IoT gateway, used in areas such as mobile IoT and logistics, industrial processes, smart agriculture, and water and wastewater management.

[0078] Control systems can monitor and manage refrigerant systems. They can utilize supply and return refrigerant temperature, pressure, and other data to provide refrigerant loss detection and analysis. Refrigerant monitoring and management services can be easily deployed, provided at low cost, and can be subsidized, for example, through AI / ML energy cost savings.

[0079] The control system can include sensors added to HVACR ducts and individual units to provide easily added monitoring and control of building indoor air quality (IAQ). Since building IAQ can also include monitoring and control of viruses and other pathogens, the control system can be integrated with UV sterilization devices placed at the duct level or elsewhere.

[0080] This invention provides automatic control of the on / off operation of heating, cooling, and refrigeration equipment under closed-loop temperature and / or humidity control via a hysteresis thermostat. The invention uses the parameter estimation methods described herein to estimate parameters of a dynamic object model of the hysteresis thermostat, the regulated space, and the cooling, heating, or refrigeration equipment, and then uses this model to create equipment control signals that achieve the desired level of energy savings.

[0081] According to the present invention, energy savings can be achieved by using any one or a combination of a number of methods. Methods that can be used may include: (1) applying preheating or precooling using time-of-use energy costs throughout the day; (2) applying preheating or precooling based on forecasted weather information or data; (3) applying multiple device “on” time values ​​based on demand response command signals; (4) reducing the number of thermostat “on” requests if the estimated device size is too large; (5) modulating thermostat “on” requests to prevent temperature overshoot and undershoot; (6) modulating thermostat “on” requests to reduce “run on” time; and (7) modulating thermostat “off” requests to increase off time. For each of the seven methods (1)-(7), a real-time energy saving estimate is calculated based on the integral difference between the estimated thermostat signal from the object model as described herein and the applied system control signal. The control signal may be from the exemplary PACE AI system described herein.

[0082] The time-of-day energy cost control method utilizes historical temperature setpoint information and "time-of-day energy rate" information transmitted from the cloud to the control system. The control system can be, for example, the exemplary PACE AI system described herein. The energy rate information can be predicted energy rate information, real-time energy rate information, or both. When the "time-of-day energy rate" information is predicted, the PACE AI can, for example, use additional control to pre-cool and pre-heat the regulated space based on known (predicted) energy costs. When the "time-of-day energy rate" information occurs in real time, the system (e.g., PACE AI) can use additional control to change the setpoint temperature of the regulated space to a predetermined or user-specified value.

[0083] The weather control method utilizes historical temperature setpoint information and weather information transmitted from the cloud to PACE AI. The weather information can be forecast, real-time, or both. Regardless of whether the weather information is forecast or real-time, additional control can be provided, for example, through the PACE AI system described herein, to adjust equipment in a timely manner based on forecast and / or real-time weather information to achieve further energy savings.

[0084] Demand response control methods utilize signals from utilities to reduce electricity demand during peak load periods on utility infrastructure. Reduction values ​​can range from 1 (turn off all energy-consuming devices) to 0 (no action required). For example, the algorithm provided in the exemplary PACE AI system described herein can provide control to meet multiple demand reduction values, such as those ranging from 0 (devices off) to 1 (devices on). As an example, if a 25% reduction is specified, the algorithm can adjust sub-relays as described herein to achieve a 25% reduction in existing power consumption.

[0085] The oversized equipment control method utilizes and compares the equipment "on" time and the number of cycles per time unit to establish a dynamic indicator that indicates the equipment is oversized. Based on this indicator, the thermostat "on" request signal can be extracted once every x cycles (x is a variable), thereby generating fewer "on" cycles and an increased "on" duration until the oversized indicator is eliminated.

[0086] Overshoot and undershoot prevention control methods use measurement data and corresponding object models to predict the amount of time that the thermostat "on" signal should be cut off to prevent overshoot or undershoot of the regulated space temperature.

[0087] The shortened-time control method uses both measured values ​​and a corresponding object model to measure the rate of temperature change of the regulated space during equipment operation. When the rate of temperature change reaches a threshold, the thermostat signal is forcibly shut off. Then, when the regulated space temperature rises (for cooling) or falls (for heating), the thermostat controller responds normally.

[0088] The added shutdown time control method utilizes both measured values ​​and a corresponding object model to measure the rate of change of the regulated space temperature when the equipment is not in operation. The thermostat signal is kept off until the regulated space temperature exceeds a threshold. Once this occurs, the thermostat controller responds normally, turning on to perform both cooling and heating.

[0089] The aforementioned seven methods (1)-(7) can be used individually, independently of each other, together, or in any combination thereof.

[0090] This invention provides an electronic controller that can be used to implement indicated methods for estimating parameters of a duty cycle HVACR device and energy savings achievable using adjusted control signals generated according to the invention. The indicated controller can be implemented as part of a retrofittable electronic controller add-on that includes an integrated program capable of automatically and optimally calculating and controlling the duty cycle and cycle duration of heating, cooling, and / or refrigeration equipment controlled to specific thermodynamic and electrical demand levels. The add-on electronic controller can be connected in series in one or more thermostat control signal lines, capable of intercepting the thermostat signal before it reaches the intended load unit of the HVACR system. The electronic controller can apply algorithms to the OEM signal and its behavior to generate an output signal for the load unit that can replace (or allow) the original control signal to provide a selected level of energy savings in the system. The controller can be implemented as a computer program stored in a memory device and executable by a microprocessor embodied in the electronic controller. This program can provide signal processing algorithms. The electronic controller can include signal generation capabilities to output control signals from the electronic controller to the load unit. Electronic controllers can be easily retrofitted into existing HVACR systems or incorporated into new HVACR systems.

[0091] Figure 1AAn HVACR system 11 is illustrated, comprising an electronic controller 18, on which one or more controller programs can reside or be retrieved, and which can be executed to perform signal processing and generation. The electronic controller 18 includes a main relay 30 and secondary relays 50. The electronic controller communicates with a cloud-based command center 70 via, for example, a receiver and a transmitter. The cloud-based command center 70 is configured to run artificial intelligence (AI) / machine learning (ML) software and provide control signals or AI commands to the electronic controller 18 based on received and / or input data. The electronic controller 18 can then control load units 20, edge device electrical appliances, based on the AI ​​commands. Alternatively, the electronic controller 18 can alternatively send different commands to the load units 20 (e.g., commands from a thermostat).

[0092] Electronic controller 18 can be retrofitted in system 11 to provide control of at least one HVACR load unit 20, which provides conditional control in zone 2. Power line 10 passes through utility meter 12 at the structure, where at least one load unit 20 to be controlled is located. Meter 12 measures the use and demand of electrical energy at that location. Load unit 20 can be, for example, an air conditioner, heat pump, furnace, refrigerator, boiler, or other load unit of an HVACR system.

[0093] The operative main power line 10 is typically unconditional and supplies operative power to the load unit 20 via a load control switch 26 (such as a relay), and also typically supplies operative power to other load units and appliances (not shown) in the same configuration. Power line 10 can be, for example, 110 volt AC (VAC) or 220 VAC, or another main power line supplying power to the HVACR system 11 for retrofitting with the electronic controller 18. For the system 11 to be retrofitted, at least one standard thermostat 14 can be connected to the HVACR load unit 20. The thermostat 14 can be connected to power line 10 via line 13. For simplicity, a step-down transformer, such as a 24 volt transformer, that can be used to supply power from power line 10 to the thermostat is not shown in this figure. However, in… Figure 21 The wiring diagram shown illustrates a step-down transformer.

[0094] The electronic controller 18 also uses direct input from a dedicated temperature sensor 22 to operate and function as designed. The temperature sensor 22 can be an external temperature sensor, an internal temperature sensor, or both. The temperature sensor can be located remotely from the thermostat 14. The thermostat 14 can be located inside a building or structure with a space where temperature regulation is required. The electronic controller 18 and the temperature sensor 22 can communicate via a hardwired or wireless communication line 17. The temperature sensor can be a component physically separate from the electronic controller, or alternatively, it can be integrated with the electronic controller, for example, when the electronic controller is located externally, which is convenient.

[0095] Electronic controller 18 is not directly powered from power line 10, nor is it required to be. Electronic controller 18 is powered by thermostat signaling intended for use with multiple load devices. Electronic controller 18 is typically electrically dormant (or inactive) or asleep relative to its signal processing characteristics until it receives / intercepts an ON signal from the thermostat, at which point electronic controller 18 is awakened (activated) to apply a program, as part of an algorithm, such as that described herein, for signal control processing and the generation of control signals to the intended load devices.

[0096] In a typical scenario, the control signal line 15 of the thermostat 14 can transmit a 24-volt AC voltage during periods of thermostatic control, such as when cooling from an air conditioning unit (load unit) or heating from an electric furnace is required. The control signal typically activates the load control switch 26 in the main power line 10 to supply power to the load unit 20. That is, without the electronic controller 18, the control signal line 15 controls the opening or closing of the load unit control switch 26, thereby opening or closing the circuit of the operating power line 10 and controlling the flow of operating power to the load unit 20. The electronic controller 18 is inserted in series and mounted in the thermostat control signal line 15 at some point between the thermostat 14 and the load unit control switch 26. As shown, the thermostat line 15 can be cut and connected to the electronic controller 18 at one cut end. Also as shown, the remaining portion of the cut signal control line (labeled line 24) can be connected to the electronic controller 18 at one end and to the load control switch 26 at the other end.

[0097] The electronic controller 18 can be physically mounted in a metal plate (not shown), such as a standard metal plate housing used with the load unit, for example, near the load unit 20. Preferably, the tap of the electronic controller 18 to the control signal line 15 (24) is as close as practically feasible to the load control switch 26. Typically, the connection can be made within the physical boundaries of the load unit itself. The connection of the electronic controller 18 to the control signal line can be made, for example, within the housing of the compressor unit containing the residential air conditioning unit. For example, the electronic controller 18 can be mounted in a metal plate housing that houses the OEM controller of the compressor of the air conditioning unit, which is mounted on a flat plate or platform near the ground, adjacent to the home or building supported by the unit, or mounted on its roof.

[0098] The electronic controller 18 may include on-board user interface controls 19 and / or may receive control inputs and / or parameter data 23 from a remote input device 21, as will be further understood from the following description. The input device 21 may be “remote” because it is a device physically separate from the electronic controller 18, which may communicate with the controller (e.g., via an attachable / detachable communication line or cable link or a wireless communication link). The remote input device 21 may be an electronic service tool for the controller, a laptop computer, desktop computer, tablet computer, smartphone, or other device. The temperature sensor 22 may be positioned as a separate unit or integrated with the controller (if also externally located), near the compressor of the air conditioning unit or other load unit to be controlled, mounted on a flat plate or platform near the ground, adjacent to and outside the house or building supported by the unit, or mounted on its roof, or located elsewhere near the house or building supported by the unit.

[0099] In operation, the electronic controller 18 receives current via control signal line 15 based on the thermostat control signal used to power the load unit 20, and the electronic controller 18 can be immediately woken up to intercept the thermostat signal and initiate its control program suite before the output control signal is sent from the electronic controller 18 to the load unit switch 26. As shown, the output control signal can be a replacement signal for the OEM signal or the OEM signal itself, depending on the result of the controller algorithm's execution.

[0100] Thermostat 14 is preferably configured or pre-configured to generate only an ON / OFF signal, by which the air conditioner / heat pump compressor, furnace, or other load unit is turned on / off. Preferably, the thermostat 14 used in system 11 is designed to provide ON / OFF control at the load unit to fully turn the load unit on or off. When the thermostat is an ON / OFF control device, it can determine whether the output needs to be turned on, off, or maintained in its current state. ON / OFF control of OEM thermostats typically includes selecting a setpoint, and can be applied or can be selected by the user across the native or default OEM deadband.

[0101] Temperature sensor 22 may be located near the load unit 20 outside the structure, which includes at least one temperature-controlled space. Temperature sensor 22 may be a remote temperature sensor. Temperature sensor 22 may be a sensing module that can be inserted into an external electrical outlet connected to power line 10 via integrated multiple pins and / or may be battery-powered. Temperature sensor 22 may be part of a device that plugs directly into an external outlet, or it may be connected to the outlet as a module via a power cord or power extension cord. External temperature sensors may be included in a battery-powered unit.

[0102] Although for the sake of simplification, Figure 1A A single control line 15 is shown cut from and connected to the electronic controller 18 from a single thermostat 14. However, it should be understood that in a single or dual thermostat configuration, multiple control lines from a single thermostat or a single control line from each of multiple thermostats can be cut and individually connected to the electronic controller 18, such as different corresponding input pins of the electronic controller. In cases where the electronic controller 18 controls more than one load device, the output signal control line can be connected at one end to the electronic controller 18 and at the other end to the load control switch of each load device. For simplicity, in Figure 1 and... Figure 14 The HVACR system 11 shown only includes one load unit 20 under the load control and management of the electronic controller 18 and a single control signal line. It should be understood that the HVACR system 11 may include multiple individual loads under thermostat control, such as multiple compressors, or compressor units and blower units, as well as other similar or different loads, depending on the configuration.

[0103] As noted, the electronic controller of the present invention can be fully connected to the control lines of each sub-load of the device. In other words, the air conditioner can have separate control lines for the compressor unit and the blower unit. The electronic controller can be used to control one or both of these sub-loads. The total power line to all sub-loads of the air conditioning unit is generally not altered in any way by the electronic controller of the present invention. Furthermore, conventional grounding components are not included. Figure 1A The diagram is shown because it is not a particular concern in this invention.

[0104] Figure 1A The electronic controller 18 can be implemented, for example, in a stand-alone configuration or a networked configuration. A stand-alone configuration can be used, for example, in a residential application with a single load unit (e.g., <5 tons HVACR load unit). A networked configuration can be used, for example, as part of a building management system (BMS) to provide HVACR in larger-scale applications, such as residential, commercial, or industrial buildings or equipment with higher energy usage / demand, or as a network of electronic controllers, each of which is attached to a dedicated load unit.

[0105] Figure 1A The electronic controller 18 includes at least one microprocessor operable to receive thermostat input signals, apply an indicated program to the received thermostat signals, and send output signals to the HVACR load unit to be controlled under the command of the microcontroller.

[0106] like Figure 1B As shown, including Figure 1A The microcontroller 183 in the electronic controller 18 may include, for example, a microprocessor for storing and executing instructed controller programs, performing data collection functions, controlling signal generation to the load device(s), and calculating estimated power savings. Figure 1B As shown, the microcontroller 183 may include a microprocessor 1832 and a computer-readable storage medium 1833, shown as in-memory 1835, both integrated on the same chip. The microprocessor 1832 (also referred to as a central processing unit (CPU)) contains arithmetic, logic, and control circuitry necessary to provide computing power to support the controller functions shown herein.

[0107] The memory 1835 of the computer-readable storage medium 1833 may include non-volatile memory, volatile memory, or both. The computer-readable storage medium 1833 may include at least one non-transitory computer-usable storage medium or memory storage device. Non-volatile memory may include, for example, read-only memory (ROM) or other permanent storage devices. Volatile memory may include, for example, random access memory (RAM), buffers, cache memory, network circuitry, or combinations thereof. The computer-readable storage medium 1833 of the microcontroller 183 may include embedded ROM and RAM. Programming and data may be stored in the computer-readable storage medium 1833 including memory 1835. Program memory may be provided, for example, for an energy control algorithm controller program 1838, which includes, for example, an energy control main program 1836, an object model controller program 1837, a delay calculation controller program 1831, and an energy calculation controller program 1839, as well as storing menus, operating instructions, and other programming, parameter values, etc., as shown herein, for controlling the electronic controller 18. These programs may be stored in ROM or other memory.

[0108] The programs, when combined, provide an integrated control program 1838 residing on the electronic controller 18. Data memory (such as flash memory) can be configured with data parameters. The memory can be used to store acquired data related to the operation of the load device to be controlled, such as thermostat command on time and calculated off time. The microprocessor 1832 and memory 1833 can be integrated and supported on a common motherboard 1830, etc., which can be housed in a housing (not shown) having input and output connection terminal pins, multiple communication link / interface connector ports (e.g., small or micro or standard-sized USB ports for receiving USB plugs of appropriate sizes), etc., which will relate to... Figure 21 Further discussion.

[0109] The microcontroller 183 may be, for example, an 8-bit or 16-bit or larger microchip microprocessor, which includes the indicated microprocessor and memory components, and is operable to input and execute the indicated demand regulator controller program and other included programs. Programmable microcontrollers are commercially available and can be input with the control program shown herein to provide the required control. Suitable microcontrollers in this regard include those available from commercial vendors such as Microchip Technology Corporation of Chandler, Arizona, USA. Examples of commercially available microcontrollers in this regard include, for example, Microchip Technology Corporation's PIC16F87X, PIC16F877, PIC16F877A, PIC16F887, dsPIC30F4012, and PIC32MX795F512L-801 / PT; Analog Devices' ADSP family; Jennic's JN family; National Semiconductor's COP8 family; Freescale's 68000 family; Maxim's MAXQ family; Texas Instruments' MSP 430 family; and the 8051 family manufactured by Intel, etc. Other possible devices include FPGAs / ARMs and ASICs. The demand regulator controller program shown in this article can be input into a suitable microcontroller using industrial development tools, such as the MPLABX integrated development environment from Microchip Technology.

[0110] Although electronic controller 18 is Figure 1A The controller is shown as a separate unit connected to the thermostat signal line 15 (24) to the load unit to be controlled, but the microelectronics indicated by the controller can optionally be incorporated and integrated into the thermostat unit or building management system (BMS). Algorithms for the control programs and features of the demand regulator controller program and other indications of the electronic controller can be added to the local thermostat signal control software of the thermostat. Algorithms for the control programs and features of the demand regulator controller program and other indications of the electronic controller can be added to the building management system (BMS) software, where the BMS provides control for one or more load units of the HVACR, thereby eliminating the need for a physically separate electronic controller. In a combined thermostat / electronic controller arrangement, interception of the OEM thermostat signal and processing of it by the controller microelectronics can be performed at the modified thermostat unit without connecting a physically separate microelectronic controller to the thermostat signal line 15 (24) between the thermostat and the load unit to be controlled.

[0111] Figure 1C yes Figure 1A A magnified and more detailed view of the main and secondary relays connected in series as shown in the electronic controller. Figure 1CAn example of a dual-relay embodiment installed at an edge node on a single-channel (single-stage) HVAC unit is illustrated. Such an embodiment could be used, for example, in a small heat pump or a small air conditioning unit. Figure 1C As shown, the main relay 30 includes multiple terminals for wired connection to various inputs and outputs. Terminal 32 is configured for connection to a thermostat command signal, which is input to channel 1 of the main relay 30. The connection between the thermostat command signal transmitter and terminal 32 can be wired or wireless. The thermostat command signal transmitter can be located, for example, at the thermostat itself.

[0112] Terminal 34 is configured for a wired connection to a 24-volt AC output, and the 24-volt output is routed to terminal 36 via wire 38, wherein the main relay 30 receives a constant 24-volt power supply. Wire 38 also provides a constant 24-volt power supply to terminal 52 at the secondary relay 50.

[0113] The main relay 30 also includes a terminal 40 at which AI command signals are received, for example, from a command center. The AI ​​command signals can be transmitted via... Figure 1A The cloud-based command center 70 shown transmits data via cloud computing. The connection between the cloud-based command center 70 and the terminal 40 can be a direct wireless transmission or via a path including a wired connection to an intermediate receiver at the terminal 40 using an intermediate Bluetooth® device.

[0114] Figure 1C Terminals 42 and 44 are used for wired connections to a direct current (DC) circuit. In the arrangement shown, terminal 42 is configured to connect to the negative terminal of the DC source, and terminal 44 is configured to connect to the positive terminal of the DC source.

[0115] The secondary relay 50 includes a terminal 60 at which it receives AI command signals, for example, from a cloud-based command center 70. The AI ​​command signals can be transmitted via... Figure 1A The cloud-based command center 70 shown transmits data via cloud computing. The connection between the cloud-based command center 70 and the terminal 60 can be a direct wireless transmission or via a path including a wired connection to an intermediate receiver at the terminal 60 using an intermediate Bluetooth® device.

[0116] Figure 1C Terminals 62 and 64 are used for wired connections to a direct current (DC) circuit. In the arrangement shown, terminal 62 is configured to connect to the negative terminal of the DC source, and terminal 64 is configured to connect to the positive terminal of the DC source.

[0117] The secondary relay 50 also includes a terminal 54 for outputting control signals to a load unit (e.g., a compressor) on channel 1. Terminal 56 is not used in the illustrated embodiment, but can be used for functions such as outputting signals, outputting data, and sending error messages to the cloud-based command center 70.

[0118] Figure 2 This is a flowchart 200 of a method according to an embodiment of the present invention. Flowchart 200 includes receiving regulation data including at least one of control inputs, state variable inputs, and sensor inputs; processing the regulation data using AI software to create an optimized thermostatic control signal for an appliance in the regulated space; and estimating the baseline power consumption of the appliance 210. The regulation data may be received by an edge device or another thermostatic control device. Control inputs may include temperature control, thermostatic setpoint, humidity control, etc. State variable inputs may include thermal capacity and resistance parameters, the size of the regulated space, the estimated power consumption of the appliance, etc. Sensor inputs may include sensed temperature, sensed humidity, sensed power consumption, etc.

[0119] The edge device and / or the thermostat then controls the appliance based on the optimized thermostat control signal and continuously sends sensor outputs to update the thermostat control signal 220. The sensor outputs may include sensed temperature, sensed humidity, sensed power consumption, etc. By sensing the sensor outputs while controlling the appliance, the edge device and / or the thermostat can continuously update and improve the optimized thermostat control signal.

[0120] Flowchart 200 also includes receiving an energy consumption modification command, wherein the energy consumption modification command includes and / or activates the retrieval of modification data, including a demand reduction command value, time-of-use energy rates for the day, and / or weather information 230. The energy consumption modification command can be sent to an edge device and / or an AI command center. As an example, the energy consumption modification command can be generated by a computing device of a third party, such as an electricity / grid operator. The energy consumption modification command can include a demand reduction command value, which can be a specific percentage of power reduction requested. Alternatively, the energy consumption modification command can be generated by a computing device of an operator of an edge device and / or a thermostat, including a request for power reduction of appliances. The edge device and / or AI command center can retrieve time-of-use energy rates and / or weather information for the day. Time-of-use energy rates for the day can provide data indicating current and predicted peak load times, while weather information includes data on current and predicted weather.

[0121] The edge device and / or AI command center then receive power consumption data from the sensors to determine the real-time energy consumption of appliance 240. The edge device and / or AI command center then use AI software to generate an energy consumption modification signal based on modified data, adjustment data, baseline energy consumption, and real-time energy consumption data 250. If the energy consumption modification signal is generated by the AI ​​command center, the AI ​​command center can wirelessly transmit the energy consumption modification signal to the edge device. The edge device can then control the appliance based on the energy consumption modification signal, which includes a timing modification of the appliance's ON / OFF command to achieve energy consumption modification 260.

[0122] As an option, the edge device can persistently control electrical appliances based on energy consumption modification signals. Alternatively, the invention can control the edge device to resume and control electrical appliances based on optimized thermostatic control signals. For example, the edge device can receive a termination command for energy consumption modification 270 from an operator's computing device or from a third party (such as a remote power / grid operator). Once the termination command is received, the edge device and / or the thermostatic control device control the electrical appliances based on the optimized thermostatic control signal and continuously send sensor outputs to update the optimized thermostatic control signal 220.

[0123] Figure 3 This is a schematic diagram of system 302 according to an embodiment of the present invention. System 302 includes an AI edge device 310, a thermostat 320, and a sensor 330 that feeds data to the AI ​​edge device 310 and the thermostat 320. The thermostat 320 and / or the AI ​​edge device 310 control a system for an HVACR device or device 360, for example, for regulating electrical appliances in a regulated space. Figure 3 As shown, the smart device 340 (e.g., a smartphone) can wirelessly communicate with the thermostat 320 and / or the AI ​​edge device 310 directly (e.g., via Bluetooth® or another short-range wireless connection) or via the Internet 350 or a similar network. Bluetooth® is a registered trademark of BLUETOOTHSIG, Inc., Kirkland, Washington (a company based in Delaware). The smart device 340 receives data from the AI ​​edge device 310 and the thermostat 320 that can be displayed to the end user.

[0124] AI edge device 310 communicates with remote AI command center 370 and grid operator 380. Each communication can be independent of the other and can be, for example, via the Internet 350. In this example, remote AI command center 370 can receive data, generate command signals, and send the command signals to AI edge device 310. Grid operator 380 can send an energy consumption modification command including a demand reduction command value. The demand reduction command value can be a request for a specific percentage reduction in electricity consumption. The demand reduction command value is sent to AI command center 370, and AI command center 370 uses AI software to generate an energy consumption modification signal. AI command center 370 then sends the modification signal to AI edge device 310 via Internet 350, and AI edge device 310 then controls HVACR appliances 360 based on the modification signal. Alternatively, the demand reduction command value can be sent directly from grid operator 380 to AI edge device 310. AI edge device 310 can then generate a modification signal and control HVACR appliances 360 based on the modification signal.

[0125] Figure 4 This is a circuit diagram showing a dual-relay circuit for controlling a single control channel of a single load unit, such as a single HVACR unit. More complex edge node controllers can be provided, for example, by increasing the number of secondary relays.

[0126] like Figure 4 As shown, a signal generated by the thermostat 118 or the building automation system (BAS) controller is input to the circuit, such as a 24VAC signal or 0VAC. This signal may correspond to a temperature setpoint. The signal is input to the main relay 120. The main relay 120 is configured to process Automatic Demand Reduction (ADR) and take into account signals from the AI ​​command center, such as receiving real-time weather data, forecast weather data, price signal operator data for preheating and precooling, demand response to grid operator commands, data or estimates from a virtual power plant, etc. This signal may be modified at the main relay 120 and passed to or through the secondary relay 122. The resulting signal or pass-through signal output from the secondary relay 122 is sent to the control terminals of the load unit (e.g., HVACR unit, BAS, solar generator, other generator, etc.).

[0127] The secondary relay 122 can be configured to provide Persistent Demand Reduction (PDR), for example, by managing AI command center control signals. These AI command center control signals can be sent to AI / ML level edge appliances or devices, such as compressors, burners, refrigerators, solar generators, etc. By using AI command center control signals, systems also including AI / ML level edge device appliances can be configured to achieve in-cycle optimization and persistent demand reduction.

[0128] This can be achieved by changing the main relay and the secondary relay (e.g., respectively in...). Figure 1C The relays 30 and 50 shown in the diagram, as well as in Figure 4 The corresponding states and operation combinations of relays 120 and 122 shown in the figure are used to perform different operations.

[0129] The main relay 120 can be a dual-position relay. The main relay 120 can be an intelligent control input that can be used for autonomous optimization between Permanent Demand Reduction (PDR) and Automatic Demand Reduction (ADR), wherein the input from a user-controlled "slider" selector is used to select whether there are more PDR guarantees or more ADR opportunities (in commanded kW load reduction to obtain additional revenue).

[0130] Figure 5 A table showing the operations that can be implemented by the system of the present invention is shown. Figure 5 Examples of various operations for controlling a single control channel of an HVACR or other electrical appliance are illustrated.

[0131] exist Figure 5 In the present invention, items 1 to 5 are operations that can be performed using the system of the present invention as an edge node controller. The system of the present invention can be added to and / or integrated into existing systems and is configured to intercept existing control signal circuits, enabling comfortable, energy-efficient, and automatically optimized spatial regulation operations. Furthermore, the system of the present invention can be added to and / or integrated into existing systems and is configured to automatically optimize system energy consumption and power generation of the system, including the power generation edge node appliances.

[0132] The electronic controller and system of the present invention can be used in multi-channel systems. Exemplary “control channel” combinations for typical HVACR equipment can include combinations for small, large, and commercial-grade systems. An exemplary channel assignment for small-scale systems (including residential systems containing heat pumps) could be channel 1: cooling signal, channel 2: heating signal, and channel 3: blower / fan signal. For larger heat pumps or gas / electric commercial rooftop units, an exemplary channel assignment could be channel 1: stage 1 cooling, channel 2: stage 2 cooling, channel 3: stage 1 heating, channel 4: stage 2 heating, and channel 5: blower / fan. For even larger commercial systems, such as cooler systems, rack-mounted refrigeration systems, or large A / C units, the setup could be channel 1: stage 1 cooling, channel 2: stage 2 cooling, channel 3: stage 3 cooling, channel 4: stage 4 cooling, and channel 5: chilled water pump or blower / fan.

[0133] In one example, a commercially available PACE AI4 unit that is connected to a commercial gas-heated and electrically cooled roof enclosure can be equipped with features including current sensing and gas duct temperature sensing.

[0134] Figure 6 This is a schematic diagram of an exemplary embodiment of an AI-enabled thermostat 80 for a 4-channel HVAC unit including two-stage cooling and two-stage heating. The thermostat 80 can be configured to operate a larger heat pump unit, a larger air conditioning unit, and a larger gas or electric heating unit. The thermostat 80 may include multiple input terminals that can be connected to corresponding multiple sources and signals. Input terminal 82 can be configured to receive bypass / online commands. Input terminal 84 can be configured to receive signals specifying an application type, such as for controlling the cooling unit, for controlling the heating unit, for controlling the heat pump unit, etc. Input terminal 86 can be configured to receive commands for heating, cooling, or automatic control. The software parameters of each of these three inputs can be updated as an event occurs.

[0135] The thermostat 80 also includes input terminals 90, 92, 94, 96, 98, 100, 102, 104, and 106. Input terminal 90 can be configured to receive a setpoint temperature signal. Input terminal 92 can be configured to receive a zone temperature signal. Input terminal 94 can be configured to receive a date / time signal. Input terminal 96 can be configured to receive a cooling stage 1 command. Input terminal 98 can be configured to receive a cooling stage 2 command. Input terminal 100 can be configured to receive a heating stage 1 command. Input terminal 102 can be configured to receive a heating stage 2 command. Input terminal 104 can be configured to receive a reversing valve command. Input terminal 106 can be configured to receive a reset signal. Each of terminals 90, 92, 94, 96, 98, 100, 102, 104, and 106 can be configured to receive a software update signal, for example, at an update rate of once per second.

[0136] The temperature control device 80 also includes output terminals 110, 112, 114, and 116. Output terminal 110 can be configured to send a Y1 cooling stage 1 command to the two-stage cooling unit. Output terminal 112 can be configured to send a Y2 cooling stage 2 command to the two-stage cooling unit. Output terminal 114 can be configured to send a W1 heating stage 1 command to the two-stage heating unit. Output terminal 116 can be configured to send a W2 heating stage 2 command to the two-stage heating unit. Each of terminals 110, 112, 114, and 116 can be configured to send an updated command signal, for example, at an update rate of once per second.

[0137] Whether the Y1 cooling stage 1 command sent from terminal 110 to the two-stage cooling unit is the same as the cooling stage 1 command received at input terminal 96 depends on the electronic controller's processing of the signal using the main relay and secondary relay, as described above and with reference to... Figure 1C As shown. Whether the Y2 cooling stage 2 command sent from terminal 112 to the two-stage cooling unit is the same as the cooling stage 2 command received at input terminal 98 depends on the electronic controller's processing of the signal using the main relay and secondary relay, as described above and with reference to... Figure 1C As shown. Whether the W1 heating stage 1 command sent from terminal 114 to the two-stage heating unit is the same as the heating stage 1 command received at input terminal 100 depends on the electronic controller's processing of the signal using the main relay and secondary relay, as described above and with reference to... Figure 1C As shown. Whether the W2 heating stage 2 command sent from terminal 116 to the two-stage heating unit is the same as the heating stage 2 command received at input terminal 102 depends on the electronic controller's processing of the signal using the main relay and secondary relay, as described above and with reference to... Figure 1C As shown.

[0138] Figure 7 This is a flowchart depicting a method 121 according to an embodiment of the present invention, wherein the method includes step 123 of receiving regulation data of an energy consumption and energy generation system. Method steps 125, 126, 128, 130, 132, and 134 involve processing the regulation data using AI software to create optimized control signals and estimate the baseline energy consumption and supply of the system. Energy can be generated from, for example, a solar generator operating only as a component of the energy consumption and generation system. Energy generated exceeding the amount required by the energy-consuming appliances operating the system can be channeled back to the grid or battery charging facilities.

[0139] Figure 8 This is a flowchart depicting a method 136 according to an embodiment of the present invention, wherein the method includes step 138 of receiving regulation data of a refrigeration system. Steps 140, 142, 144, 146, 148, and 150 involve processing the regulation data using AI software to create an optimized isothermal control signal and estimate the baseline energy consumption of the refrigeration system. The refrigeration system may include refrigerant line temperature sensors and current sensors, which may send signals to an AI command center to generate one or more command signals, such as a temperature modification signal for changing the temperature of the refrigerated space.

[0140] Figure 9This is a flowchart depicting method 152 according to an embodiment of the present invention, wherein the method includes step 154 ​​of receiving conditioning data for the entire HVACR system. Steps 156, 158, 160, 162, 164 and 166 involve processing the conditioning data using AI software to create an optimized thermostat control signal and estimate baseline energy consumption and / or demand reduction for the entire system.

[0141] exist Figures 10-20 The diagram illustrates the functionality of a controller that can be used in the methods of the present invention, as well as other features of process control logic and calculation schemes for energy control and savings estimation, and can be implemented using microprocessor-executable software of the electronic controller of the present invention. Examples of controllers for automatically controlling the duty cycle of HVACR equipment and systems that can be used in conjunction with the present invention, and methods of using such controllers, are described in the U.S. Patent Nos. US 10,151,506B2 and US 10,782,032B2 of Kolk, the entire contents of each of which are incorporated herein by reference.

[0142] Figure 10 The process control logic of the energy control algorithm, identified by the number 300, is shown. This logic can be applied by the electronic controller of this invention to provide automatic energy control and energy saving estimates for HVACR systems. Figure 10 In this context, the uOEM and uPACE control signals are generated using an estimated object model description used only for hysteresis thermostats. Figure 11 The diagram shows a block diagram of a universal hysteresis thermostat, identified by the number 400, which can be used for heating or cooling. Figure 11 (as well as Figure 12 and Figure 13 In the code, the "merge" block represents if-then-else logic with respect to inputs b, t, and f, and output x. Figure 10 In this context, the uPACE control signal is generated by the "delay" block. This block uses... Figure 12 The logic presented in the code applies a time delay (the second input on the left) to the input signal (the first input on the left). Figure 12 The processing logic within is identified by the number 500. Figure 12In the "Delay" block shown, the "u" signal is the input signal, and the "upv" signal is the delay value of the "u" signal. When the difference between "u" and "upv" is positive, the input signal has transitioned from its 0 state to its 1 state (OFF to ON). When this occurs, the feedback loop calculates the elapsed time, and when this elapsed time > "timeDelay", the output signal "uTimeDelayed" is turned on. To prevent the timer from delaying when the input signal transitions from 1 to 0 (ON to OFF), the "uTimeDelayed" signal is multiplied by the "upv" signal.

[0143] exist Figure 10 In the code, a block titled "Calculate Energy" calculates the signal energy of the uOEM or uPACE signal. Figure 13 The diagram shows the block's block structure, where the processing logic is identified by the number 600. The "tWindow" value is globally set to the moving window width; for illustrative purposes, this is set to 2000 seconds. However, it can be set to any value; obviously, larger values ​​will require more time to calculate. An energy calculation is then performed every tWindow seconds. Figure 13 The S&H block in the process samples and stores the calculated energy every tWindow seconds.

[0144] exist Figure 10 In this context, the "Energy Saving Ratio (E)" signal is the energy difference between the uOEM signal and the uPACE signal (normalized to the uOEM signal energy). When the uPACE signal energy is less than the uOEM signal energy, the difference will be positive and will fall between 0 and 1. The "Energy Saving Ratio Reference (E)" signal is... The signal indicates the desired energy savings sought to be achieved. The "energy saving control law" is based on the "energy saving ratio reference, E". The integral controller operates by subtracting the "actual energy saving ratio, E" signal from the actual energy saving ratio signal. The integral control law reduces the error signal to zero by adjusting the time delay signal "td," regardless of whether the uPACE control model achieves the "reference energy saving ratio, E" signal. "What is the required value? Energy saving percentage reference E" The expected or selected value can be input by the user into the controller, such as via... Figure 1A The discussion refers to the airborne user input interface or remote user input device.

[0145] Figure 14 It shows Figure 10The process control logic of the "Object Model" module shown models the dynamics of a forced air heating, cooling, or refrigeration system under hysteresis (OEM) thermostat control. Figure 14 The process control logic identified by the number 700 in the figure is applied by the electronic controller indicated according to an example of the present invention. Figure 14 The following terms are defined as follows: tOAT = outside air temperature; tSetPoint = thermostat setpoint temperature; u1 = thermostat control signal (ON or OFF) for the compressor or burner; u2 = thermostat control signal (ON or OFF) for the blower; supply air temperature = inlet temperature of the regulated space register; airflow = inlet airflow of the regulated space register (equipment nameplate value); tZone = zone (regulated space) temperature; A = thermal resistivity between the regulated space and outside air (including zone heat capacity); and B = heat transfer coefficient (including zone heat capacity).

[0146] A and B are two parameters describing the operation of forced air-operated heating, cooling, and refrigeration equipment under closed-loop temperature and / or humidity control via a hysteresis thermostat. These parameters vary over time based on load, environmental, and setpoint variations; however, if they are known, energy and power consumption can be accurately calculated. This paper presents two methods for estimating A and B, referred to herein as "Method 1" and "Method 2".

[0147] In both methods, it can be assumed that for cooling applications, the "supply air temperature" is 15 degrees lower than the zone temperature. For heating applications, it can be assumed that the "supply air temperature" is 15 degrees higher than the zone setpoint temperature. This estimate is not critical to the parameter estimation process. Figure 10 When the system is under the control of the on / off thermostat, the blower and compressor (or heating coil) are controlled in either of two states: ON or OFF. In the OFF state, the B value becomes 0 because the airflow has been shut off by the thermostat.

[0148] Method 1: Method 1 uses two sense signals, external air temperature, and the ON / OFF state of the device to estimate two model parameters, A and B. The following information is provided: Application type: heating, cooling, or refrigeration. The following assumptions are made: (i) The thermostat dead zone is + / -1°F, the setpoint for cooling is 72°F, the setpoint for heating is 68°F, and the setpoint for refrigeration is 40°F (these values ​​can change, but are assumed not to be directly measurable). During the OFF state, the object model in Figure 15 Presented within, and identified therein by the number 800. Using assumptions, given information, and sensor information, it is possible to use... Figure 16The calculation scheme shown calculates the A parameter during the OFF state and is identified by the number 900. If only the outside air temperature and OFF time value are known, reasonable values ​​can be selected for the zone temperature and dead zone (based on the application type: cooling, heating, or refrigeration), and the A parameter can be estimated. The tZone(k) value is set to tZone(0) = tSetPoint – db / 2. Similarly, considering the same cooling application in the ON state, the block diagram for ON operation is shown in... Figure 17 Presented in the diagram and identified by the number 1000. Using assumptions, given information, and sensor information, along with previously calculated parameter A (which remained constant during that calculation), parameter B can be calculated, such as... Figure 18 The calculation scheme is shown in the diagram and identified by the number 1100. If the thermostat dead zone, zone temperature, outside air temperature, and ON time values ​​are known, parameter B can be calculated accurately. If only the outside air temperature and ON time values ​​are known, reasonable values ​​can be selected for the zone temperature and dead zone (based on the application type: cooling, heating, or refrigeration), and parameter B can be estimated. Parameters A and B can be dynamically refined when data is collected from the outside air temperature, ON time, OFF time, and (optionally) zone temperature and thermostat dead zone.

[0149] Method 2: For this method, three sensor signals are available for estimating parameters A and B: outside air temperature, the ON / OFF state of the device, and the ambient temperature. The following information is provided for this method: Application type: heating, cooling, or refrigeration. This method uses an extended Kalman filter (EKF) to estimate parameters A and B. However, any recursive estimation method can be used (e.g., EKF, least squares, neural networks, fuzzy logic, observers, or other estimation methods). The differential, difference, state, and partial derivative equations that can be used to implement the EKF are... Figure 19 It is shown in the diagram and identified by the number 1200. (Using...) Figure 19 The computer software implementation of the EKF algorithm shown in the equation can be used to estimate the A and B parameters of the method. The initially assumed values ​​of the A and B parameters used in the first iteration of the calculation can be arbitrary pre-selected values.

[0150] exist Figure 20 The diagram illustrates an object model with an EKF according to indicated method 2, identified by the number 1300. As described in the examples included herein, the performance of the EKF can be verified using measurement signal data simulated from this object model. The EKF receives three measurement signals from the plant; tOAT, compressor and blower commands, and tZone, and estimates parameters A and B. The actual values ​​of parameters A and B are embedded in the object model. As shown in the examples included herein, the EKF performance enables rapid and accurate estimation of the values ​​of parameters A and B.

[0151] like Figures 10-20 As discussed, the present invention (1) can achieve energy savings by applying a calculated time delay of dynamic change to an OEM control signal (“uOEM”) to generate adjusted control (“uPACE”), which is applied to the device to be controlled, and (2) estimating the energy savings achieved by replacing OEM control with adjusted control. As indicated, Figure 17 The overall configuration of the energy controller is presented. Since the device can be controlled by a modified control signal (“uPACE” control signal), the indicated object model can be used to estimate what the thermostat control signal would look like if the plant were controlled by a thermostat control signal (“uOEM” control signal). As indicated, this is achieved by sensing the zone temperature and the outside air temperature, and then calculating the energy present in the uOEM signal over a moving-time window. Also as indicated, the same object model, but including a time delay on the control signal, is then used to estimate the uPACE control signal, and the energy present in the uPACE signal can be calculated over the same moving-time window. The energy difference, normalized to the uOEM energy, is calculated and used as a feedback signal in the indicated integral control algorithm, the output of which is the time delay value used to create the uPACE signal. The feedback signal is controlled at a normalized energy-saving setpoint (which can be set to any value between 0 and 1).

[0152] refer to Figure 21 A wiring terminal is shown as an example of the installation configuration of the electronic controller of the present invention. Figure 21 Electrical connection diagram 1400 illustrates an example of a single-stage cooling application using an electronic controller according to the invention. This configuration can be used when a single air conditioning thermostat is used to control an HVAC cooling unit (compressor). The configuration also supports a thermostat with a manual switch to select heating or cooling operation. The compressor can be a compressor suitable for vapor compression cooling / refrigeration systems. The compressor may include an electric motor (not shown) for driving the compressor. The electric motor itself can be a conventional electric motor or other suitable electric motors used for or suitable for driving such a load unit.

[0153] exist Figure 21In the example shown, the electronic controller 1018 provides two independent control channels that can be wired to support different device configurations. Referring to the first pin module 1001, the first channel 1001A includes one of pins 1-4, and the second channel 1001B includes one of its pins 5-6. Output lines to the load unit(s) (e.g., a cooling unit compressor) are shown as extending from one of pins 1-3. Pin 4 can be used for a hard-wired input of a temperature signal transmitted from a temperature sensor 22 located inside or outside the building, where temperature regulation is being performed. As indicated, the temperature sensor 22 may alternatively communicate with the controller 1018 via a wireless connection, and / or may be integrated with the electronic controller, for example, if the controller is also located externally (not shown). Furthermore, the controller provides a separate “dry contact” input channel, which can be used for remote control of the controller, such as via an existing BMS system. Referring to the second pin module 1010, pins 1-2 can be used for this dry contact input module. Communication port 1020 is shown in these figures as a mini-USB port (e.g., a camera-sized USB port), but is not limited thereto. Service tools, computers, smartphones, or other devices (not shown) can be used to import / input parameters, etc., into the electronic controller 1018 by establishing a communication link with the controller via port 1020. The electronic controller 1018 can preload the indicated controller program into its onboard memory during its assembly and before field installation.

[0154] Thermostats (e.g., OEM thermostats) that can be used with the electronic controller of the present invention, such as those having Figure 21The thermostat, as shown in the wiring configuration or another configuration, can be deployed at a point in a building and sense the temperature of the ambient air. If it is higher than a selected comfort setting, it sends a signal to activate the air conditioning unit. As noted, in this invention, an electronic controller intercepts the thermostat signal, which energizes the electronic controller to process the signal according to the controller's programmed algorithm before sending the controller-processed output signal to the load unit. The air conditioning unit typically includes a compressor and condensers and evaporators connected to each other in a closed refrigerant system (not shown). The refrigeration cycle itself is well known (see, for example, U.S. Patent No. 4,094,166, which is incorporated herein by reference in its entirety). Essentially, gaseous refrigerant is delivered from the compressor to the condenser coil, where it releases heat, and then through an expansion valve to the evaporator coil, where it absorbs heat from the circulating air passed by the evaporator fan. When the thermostat senses that the ambient air has been cooled to a selected level, the thermostat enters a shutdown state to turn off the compressor, evaporator fan, and condenser fan until the ambient temperature reaches a level requiring further cooling again. As indicated, when the thermostat stops signaling to the load unit, the electronic controller of the present invention enters a sleep state until the thermostat sends the next power-on signal to the same load unit. As indicated, this signal will be intercepted by the electronic controller, which is powered on to process the signal according to its programmed algorithm before sending the processed output signal to the load unit. As indicated, a dead zone is typically applied to the control temperature setting at the thermostat, and this dead zone can be effectively modified in a controlled manner by the electronic controller to improve demand savings.

[0155] Figure 21 The pin assignments indicated for the first channel 1001A and the second channel 1001B can be applied to similar pin modules for other types of load units in HVACR systems, such as dual-stage cooling units, heating units (e.g., gas, electric, heat pumps), boilers, etc. Other aspects of the electrical connection configurations that can be used in these other types of load units can be easily adapted and implemented where applicable. In these ways, for example, an electronic controller with the indicated demand regulator controller can operate to intercept and process the thermostat's control signal using an algorithm that can automatically generate enhanced control signals to provide energy and save on control. Among other benefits and advantages, existing HVACR systems can, for example, embody this controller as shown herein to improve energy consumption and reduce the energy costs of heating, cooling, and refrigeration equipment.

[0156] The edge device according to the invention can be used in the method of the invention, wherein, by optimizing rules, if the sensed return air or fluid temperature is within a specific range, for example, (>) (<) X, the HVACR unit can be operated to not operate in vapor compression air conditioning mode or heating mode. In this way, the edge device can be used to achieve the purpose of energy-saving operation of the HVAC unit without incurring the cost of the energy-saving device. Further details regarding this use and optimization can be found in U.S. Patent Application Publication No. US 2016 / 0025364 A1 of Mills, Jr. et al., the entire contents of which are incorporated herein by reference.

[0157] The edge device can be configured to provide a user-controllable “slider” that can be used to apply a customizable set of variables for building operators or industrial process control operators. The slider can provide greater energy savings and / or demand reduction at one end and greater operability as a characteristic of the native control architecture at the other end. Therefore, an adjustable slider can provide greater assurance of “normal” operator comfort while ensuring the achievement of performance variables. Further details regarding this configuration and use can be found in U.S. Patent Application Publication No. US 2016 / 0025364 A1 by Mills, Jr. et al.

[0158] For edge devices, for example, by using a database of OEM recommendations and sensed data, AI control algorithms can calculate optimized hourly startup times for HVACR and other equipment. These AI control algorithms can provide limits on device startup to ensure compliance with equipment manufacturer specifications and can implement anti-short-cycle and other machine protections. In addition to the equipment operation examples described herein, AI control algorithms and sensed data can also use the aforementioned variables to detect and correct anomalies in the operation of HVACR and other equipment.

[0159] Edge devices can utilize AI control algorithms that take into account various weather, electricity prices, operations, and other inputs to optimize the pre-cooling or pre-heating of the regulated space.

[0160] Edge devices can leverage AI control algorithms to improve the performance of HVACRs or other equipment that are too large or too small for their connected cooling, heating, or other loads. This improvement is particularly beneficial when the native control architecture provides suboptimal energy efficiency.

[0161] Edge devices can utilize AI control algorithms that can be combined with photovoltaic (PV) solar cell arrays to dynamically reduce energy consumption in HVAC buildings, thereby maintaining a higher level of reliability in dispatchable net power generation from the PV solar cell arrays to the grid.

[0162] Edge devices can be used with machines that utilize natural gas or other flammable materials. AI control algorithms can be incorporated into intelligent emergency shut-off at the machine unit.

[0163] Edge devices can be configured to utilize AI control algorithms applied to monitor and manage the refrigerant system in HVACR equipment. For example, the AI ​​control algorithm can be used to monitor refrigerant leaks using data such as supply and return refrigerant temperatures, refrigerant pressure, and other parameters.

[0164] AI control algorithms, together with sensors, can easily provide additional monitoring and control of indoor air quality (IAQ) in buildings, and can also be integrated with ultraviolet disinfection equipment placed horizontally in ducts or elsewhere.

[0165] The invention will be further illustrated by the following embodiments, which are intended as examples of the invention.

[0166] Example

[0167] Example 1

[0168] The performance of the object model estimating model parameters A and B using the Extended Kalman Method (EKF) in Method 2, as shown, was evaluated as follows. The EKF used in this method utilizes data from... Figure 20 The measured signal data from the simulation of the object model shown will be used for testing. Figure 19 The computer software program for the equations shown is used in this simulation. In the simulation, the EKF receives three measurement signals from the object: tOAT, compressor and blower commands, and tZone, and estimates the A and B parameters. The actual A and B parameter values ​​are embedded in the object model. Two tests are performed: one using constant values ​​of the actual A and B parameters, and the other using constant A and varying B parameters to model changes in the heat load within the region. The EKF performance is judged based on its ability to quickly and accurately estimate the values ​​of these two parameters.

[0169] In Test 1: The simulation model was configured with A and B set to constant values; A = 1e-4 (i.e., 0.0001) and B = 6e-3 (i.e., 0.006). This is in accordance with US Patent No. 10,151,506B2. Figure 15 A and Figure 15 The graph shown in B presents a graph of the EKF estimate and the actual value of parameter A. The entire contents of this patent are incorporated herein by reference. (See US Patent No. 10,151,506B2.) Figure 16 A and Figure 16 The graph shown in B is a graph showing the EKF estimate and the actual value of parameter B.

[0170] In Test 2: The simulation model was configured with a constant parameter A and a time-varying parameter B. Specifically, the simulation model was configured to set A to a constant value of A = 1e-4 (i.e., 0.0001) and B to a time-varying value to model the effects of dynamically changing loads in the regulated airspace. This is described in US Patent No. 10,151,506B2. Figure 17 A and Figure 17 The graph shown in B presents a curve plotting the EKF estimate and the actual value of parameter A. This is based on US Patent No. 10,151,506B2. Figure 18 The graph shown presents a curve of the EKF estimate and the actual value of parameter B.

[0171] As shown in the test results of Test 1 and Test 2, EKF quickly and accurately estimates the values ​​of two parameters A and B.

[0172] Example 2

[0173] Simulations of a single-stage cooling system are performed on a computer model, where a single thermostat controls a compressor, such as... Figure 21 As shown, this computer model is suitable for simulating the operation of an electronic controller, which applies the principles outlined in this paper. Figures 10-15 , Figure 17 and Figure 20 The process control logic shown and Figure 16 , Figure 18 and Figure 19 The calculation scheme is as follows. The developed model is based in part on actual data obtained from the operation of the same equipment in the single-stage refrigeration configuration shown, as well as in the field using the OEM thermostat alone. The simulation model was calibrated to be consistent with the field data.

[0174] Simulation results demonstrating the performance of the energy control algorithm are presented in the following time-history graph. (Referring to US Patent No. 10,151,506B2) Figure 19 In the middle, the "Energy Saving Ratio Reference, E" will be included. "The curve is applied to the controller. It shows the actual energy saved by adjusting the time delay using an integral controller. The graph of the values ​​of these parameters is in US Patent No. 10,151,506B2." Figure 19 The curve starts at 0.1 (10% energy saving) and reaches E at time = 150,000 seconds. The curve increases to 0.2 (20% energy saving). Between 300,000 and 450,000 seconds, E It decreases back to 0.1, and starting from 450,000 seconds, it increases to 0.15 (15% energy saving). (US Patent No. US 10,151,506B2) Figure 19 The 9000-second low-pass filter is used to smooth the E signal for presentation in the time history graph. It is not used in the control algorithm. (US Patent No. US10,151,506B2) Figure 20 The time history graph shown presents the time delay signal calculated by the integrator controller. This is... Figure 17 The "time delay" block is used to create the values ​​for the uPACE control signals. (US Patent No. US 10,151,506B2) Figure 21 The time history graph shown presents the temperature signal of the regulated area of ​​the space controlled by uOEM and uPACE. As shown, temperature changes increase with energy savings.

[0175] This invention includes the following aspects / embodiments / features in any order and / or in any combination: 1. The present invention relates to a method for reducing the energy consumption and / or demand of electrical appliances in edge devices, the method comprising: Receive regulation data relating to the thermostatic output of the electrical appliances at the edge of the regulated space, the regulation data including one or more of control inputs, state variable inputs and sensor inputs; Artificial intelligence (AI) software is used to process conditioning data to create optimized thermostatic control signals for edge device appliances in the regulated space. The electrical appliances at the edge of the regulated space are controlled based on the optimized constant temperature control signal. Receive energy consumption modification commands, which include or activate the retrieval of modification data, including demand reduction command values, time-of-use energy rates during the day, weather information, or combinations thereof; Power consumption data is received from sensors to determine the real-time energy consumption data of edge device electrical appliances in the regulated space; AI software is used to generate energy consumption modification signals, which are based on adjustment data, modification data, and real-time energy consumption data; and The system controls edge devices and electrical appliances in the regulated space based on energy consumption modification signals, wherein the energy consumption modification signals include timing modifications to the ON / OFF commands of the electrical appliances to achieve energy consumption reduction.

[0176] 2. The method according to any of the foregoing or following embodiments / features / aspects further includes a sensor output from the edge device electrical component for updating the optimized thermostatic control signal.

[0177] 3. The method according to any of the foregoing or following embodiments / features / aspects, wherein the sensor output for updating the optimized thermostatic control signal is continuously sent from the edge device electrical components.

[0178] 4. The method according to any of the foregoing or following embodiments / features / aspects further includes estimating baseline energy consumption data of edge device appliances in the regulated space based on regulation data, and the energy consumption modification signal is also based on the estimated baseline energy consumption data.

[0179] 5. The present invention relates to a method for automatically controlling and managing the energy consumption and operation of at least one edge appliance in an HVACR system, the at least one edge appliance including a hysteresis thermostat and configured to operate based on a control signal, the method comprising the following steps: Receive thermostat command signals from the hysteresis thermostat; The controller receives a first AI command signal at its main control relay. The controller includes a main control relay and secondary control relays arranged in series with it. The AI ​​command signal is generated by an artificial intelligence (AI) command center located remotely from the controller. The AI ​​command center includes a processor that runs one or more AI control algorithms to generate the AI ​​command signal. The main control relay has an open state and a closed state, and the secondary control relay also has an open state and a closed state. When the main control relay is in the off state, the continuous ON command signal is transmitted to the secondary control relay through the main control relay, and the output of the main control relay is the AI ​​command signal. When the main control relay is in the closed state, its output is a thermostat command signal. When the secondary control relay is in a closed state, the output of the primary control relay is sent to the AI ​​command center, where it is processed by one or more AI control algorithms to form a processed signal. This processed signal is then transmitted as a control signal to at least one edge device. When the secondary control relay is in the open state, regardless of the state of the primary control relay, an OFF command signal is generated as the control signal; and Based on the states of the main control relay and the secondary control relay, control signals are sent to at least one edge appliance.

[0180] 6. The method according to any of the foregoing or the following embodiments / features / aspects further includes: The external temperature of at least one adjacent edge appliance is sensed to form the sensed external temperature; A temperature signal is generated based on the sensed external temperature; Send the temperature signal to the controller; and (i) Send the temperature signal from the controller to the AI ​​command center for processing by one or more AI control algorithms and to be taken into account when forming the processed signal, or (ii) Adjust the control signal at the controller to form an adjusted control signal and then send the adjusted control signal to at least one edge appliance.

[0181] 7. The method according to any of the foregoing or the following embodiments / features / aspects further includes: The ambient temperature in the space to be heated or cooled by at least one edge appliance is sensed to form the sensed ambient temperature; A temperature signal is generated based on the sensed ambient temperature; Send the temperature signal to the controller; and (i) Send the temperature signal from the controller to the AI ​​command center for processing by one or more AI control algorithms and to be taken into account when forming the processed signal, or (ii) Adjust the control signal at the controller to form an adjusted control signal, and then send the adjusted control signal from the controller to at least one edge appliance.

[0182] 8. The method according to any of the foregoing or the following embodiments / features / aspects further includes: Inputting predicted temperature information at a location outside but adjacent to the space to be heated or cooled by the edge appliance, the input including feeding the predicted temperature information into one or more AI control algorithms; and Based on the input predicted temperature information, the control signal is adjusted in the AI ​​command center to form the adjusted control signal.

[0183] 9. The method according to any of the foregoing or the following embodiments / features / aspects, further comprising: Input current temperature information sensed at a location outside but adjacent to the space to be heated or cooled by the edge appliance; this input includes feeding the sensed current temperature information into one or more AI control algorithms; and Based on the input of the sensed current temperature information, the control signal is adjusted at the AI ​​command center to form an adjusted control signal.

[0184] 10. The method according to any of the foregoing or following embodiments / features / aspects, wherein the controller further includes a display, and the method further includes: Sending information display signals from the AI ​​command center to the controller; and The information obtained from the information display signal is displayed on the monitor.

[0185] 11. The method according to any of the foregoing or subsequent embodiments / features / aspects, further comprising: Send the updated software version from the AI ​​command center to the controller; and Update the software in the controller based on the latest software version.

[0186] 12. The method according to any of the foregoing or the following embodiments / features / aspects further includes: Based on the updated software version, an update signal is sent from the controller to at least one edge appliance; and The software in at least one edge appliance is updated based on the update signal.

[0187] 13. The method according to any of the foregoing or following embodiments / features / aspects, wherein the energy supplying power to at least one edge appliance is derived from the power grid, and the method further comprises: Receive information related to the grid's energy usage at the AI ​​command center; and AI control algorithms are used to process information related to the power grid's energy use, and processed signals are generated based on this information.

[0188] 14. The method according to any of the foregoing or following embodiments / features / aspects, wherein the energy supplying power to at least one edge appliance is derived from the power grid, and the method further comprises: Receive information at the AI ​​command center regarding: (i) time-of-use energy rates for the day and (ii) demand response; and AI control algorithms are used to process information relating to (i) time-of-use energy rates and (ii) demand response, and processed signals are generated based on this information.

[0189] 15. The method according to any of the foregoing or the following embodiments / features / aspects, further comprising: AI control algorithms are used to estimate the heat capacity and thermal resistance of a space that needs to be heated or cooled by at least one edge appliance. AI control algorithms are used to estimate the power consumption required to provide a comfortable environment in spaces that need to be heated or cooled. It is estimated that the reduction in power demand and energy savings will be achieved by the assumed control signals that will provide a comfortable environment; Based on hypothetical control signals, specific control signals are generated using AI control algorithms; and Specific control signals are sent from the AI ​​command center to the controller.

[0190] 16. The method according to any of the foregoing or the following embodiments / features / aspects, further comprising: Intercepting original equipment manufacturer (OEM) thermostat command signals at the controller on the way from the hysteresis thermostat to at least one edge appliance; and Instead of sending the OEM temperature control signal to at least one edge appliance, send the control signal to at least one edge appliance.

[0191] 17. The method according to any of the foregoing or the following embodiments / features / aspects, further comprising: The control signal is modified by adjusting the open and closed states of the secondary control relay, wherein the adjustment causes at least one edge appliance to cyclically open and close, and the cycle is timed to take advantage of the energy stored in the heat capacity of the space to be heated or cooled by at least one edge appliance.

[0192] 18. The present invention also relates to an edge node device for controlling at least one edge appliance in an HVACR system, the edge node device being configured to send control signals to at least one edge appliance, the at least one edge appliance having a thermostat, the edge node device comprising: A controller includes a main control relay and a first control relay connected in series with the main control relay. The main control relay includes a first artificial intelligence (AI) command input, and the first control relay includes a second AI command input. Both the first and second AI command inputs are configured to communicate with an AI command center to receive a first AI command signal and a second AI command signal, respectively. The main control relay has an open state and a closed state, and the first control relay also has an open state and a closed state. The output terminal is configured to communicate with at least one edge appliance and to send control signals from the controller to at least one edge appliance, wherein the edge node device is configured such that: When the main control relay is in the off state, the continuous ON command signal is transmitted to the first control relay through the main control relay, and the output of the main control relay is the AI ​​command signal. When the main control relay is in the closed state, its output is the thermostat command signal generated by the thermostat. When the main control relay is initially closed, its output is sent to the AI ​​command center for processing by one or more AI control algorithms to generate a processed signal. This processed signal is then transmitted as a control signal to at least one edge device. When the first control relay is in the open state, regardless of the state of the main control relay, an OFF command signal is generated as the control signal.

[0193] 19. The present invention also relates to a network comprising any of the edge node devices described in the foregoing or the following embodiments / features / aspects, and an AI command center, wherein the AI ​​command center is configured to generate a first AI command signal.

[0194] 20. A network according to any of the foregoing or following embodiments / features / aspects, wherein the AI ​​command center is located remotely from the edge node device.

[0195] 21. The network according to any of the foregoing or following embodiments / features / aspects, wherein the AI ​​command center is a cloud-based command center.

[0196] 22. A network according to any of the foregoing or following embodiments / features / aspects, wherein the AI ​​command center is located remotely from the edge node device and communicates with the edge node device via cloud-based computing.

[0197] 23. The present invention also relates to an edge node device as described in any of the foregoing or the following embodiments / features / aspects, wherein: The edge node device is configured to send multiple control signals to a plurality of corresponding edge appliances, each of which has a thermostat; The controller includes a main control relay and multiple secondary control relays, including the primary control relay. Each of the secondary control relays is connected in series with the main control relay; Each of the secondary control relays includes a corresponding second AI command input, each of which is configured to communicate with the AI ​​command center to receive a corresponding second AI command signal; Each of the secondary control relays has an open state and a closed state; The controller has multiple corresponding outputs, each configured to communicate with a corresponding one of the multiple edge appliances and to send a corresponding control signal from the controller to the corresponding edge appliance. The edge node device is configured such that, for each corresponding edge appliance, When the main control relay is in the off state, the continuous ON command signal is transmitted to the corresponding secondary control relay through the main control relay, and the output of the main control relay is the AI ​​command signal. When the main control relay is in the closed state, its output is the thermostat command signal generated by the thermostat. When the corresponding secondary control relay is closed, the output of the primary control relay is sent to the AI ​​command center for processing by one or more AI control algorithms to form a processed signal. This processed signal is then transmitted as a control signal to the corresponding edge periodically. When the corresponding secondary control relay is in the open state, regardless of the state of the primary control relay, an OFF command signal is generated as a control signal.

[0198] 24. The present invention also relates to a network comprising any of the edge node devices described in the foregoing or following embodiments / features / aspects, and an AI command center, wherein the AI ​​command center is configured to generate a first AI command signal.

[0199] 25. The present invention also relates to a system comprising any of the foregoing or following embodiments / features / aspects, an edge node device, an HVACR appliance, a current sensor, and a duct temperature sensor, wherein the current sensor is configured to send a sensed current signal to a controller, the duct temperature sensor is configured to send a sensed duct temperature signal to the controller, and the controller is configured to send data relating to the sensed current signal and the sensed duct temperature signal to an AI command center.

[0200] 26. The system according to any of the foregoing or the following embodiments / features / aspects, wherein the system further includes an AI command center.

[0201] 27. The system according to any of the foregoing or the following embodiments / features / aspects further includes a cloud-based learning database of selected data, wherein the AI ​​command center communicates with and is configured to retrieve data from the cloud-based learning database of selected data.

[0202] 28. The present invention also relates to a non-transitory computer-readable storage medium storing instructions, which, when executed by a computer, cause the computer to perform a process comprising: Receive regulation data relating to the thermostatic output of the electrical appliances at the edge of the regulated space, the regulation data including one or more of control inputs, state variable inputs and sensor inputs; Artificial intelligence (AI) software is used to process conditioning data to create optimized thermostatic control signals for edge device appliances in the regulated space. The electrical appliances at the edge of the regulated space are controlled based on the optimized constant temperature control signal. Receive an energy consumption modification command, which includes or activates the retrieval of modification data, including demand reduction command value, time-of-use energy rates during the day, weather information, or a combination thereof; Power consumption data is received from sensors to determine the real-time energy consumption data of edge device electrical appliances in the regulated space; AI software is used to generate energy consumption modification signals, which are based on adjustment data, modification data, and real-time energy consumption data; and The system controls edge devices and electrical appliances in the regulated space based on energy consumption modification signals, wherein the energy consumption modification signals include the timing modification of the ON / OFF commands of the electrical appliances to achieve energy consumption reduction.

[0203] 29. The method according to any of the foregoing or following embodiments / features / aspects, further comprising one or more of the following steps: (1) applying preheating or precooling using time-of-use energy costs during the day; (2) applying preheating or precooling based on forecasted weather information or data; (3) applying multiple device “on” time values ​​based on demand response command signals; (4) reducing the number of thermostat “on” requests if the estimated device size is too large; (5) regulating the thermostat “on” requests to prevent temperature overshoot and undershoot; (6) regulating the thermostat “on” requests to reduce “run” time; and (7) regulating the thermostat “off” requests to increase off time.

[0204] 30. The method according to any of the foregoing or the following embodiments / features / aspects further includes: Receive regulation data relating to the thermostat or other control output of at least one edge device electrical appliance, the regulation data including one or more of control inputs, state variable inputs and sensor inputs; One or more (AI) control algorithms are used to process regulation data to create an optimized thermostatic control signal for at least one edge device appliance in the regulated space; At least one edge device electrical appliance in the regulated space is controlled based on the optimized constant temperature control signal. Adaptively estimate the heat capacity, thermal resistance, and internal load of the regulated space; The object model is run based on the measured data and the estimated temperature setpoint, wherein the estimated temperature setpoint and the regulated space temperature are generated by the object model and applied to the hysteresis temperature controller model to create the estimated thermostat signal; and Energy savings are estimated based on the integral difference between the estimated thermostat signal and the control signal generated by the object model.

[0205] This invention may include any combination of the various features or embodiments described above and / or below as set forth in sentences and / or paragraphs. Any combination of features disclosed herein is considered part of the invention and is not intended to limit the composable features.

[0206] The entire contents of all references cited in this disclosure are incorporated herein by reference. Furthermore, when quantities, concentrations, or other values ​​or parameters are given as a list of ranges, preferred ranges, or upper and lower preferred values, this should be understood as specifically disclosing all ranges formed by any pair of any upper or preferred value and any lower or preferred value, regardless of whether the range is disclosed individually. Where numerical ranges are listed herein, unless otherwise stated, the range is intended to include its endpoints, as well as all integers and fractions within that range. When ranges are defined, it is not intended to limit the scope of the invention to the specific values ​​listed.

[0207] Other embodiments of the invention will be apparent to those skilled in the art upon consideration of this specification and the practice of the invention disclosed herein. This specification and the embodiments are intended to be considered exemplary only, and the true scope and spirit of the invention are indicated by the appended claims and their equivalents.

Claims

1. A method for reducing the energy consumption and / or demand of edge device electrical appliances, comprising: Receive adjustment data relating to a constant temperature or other control output for the edge device electrical components of the regulated space or other load, the adjustment data including one or more of control inputs, state variable inputs, and sensor inputs; Artificial intelligence (AI) software is used to process the conditioning data to create optimized thermostatic or other control signals for the edge device electrical components of the regulated space or other load. Based on optimized constant temperature or other control signals, control the electrical components of the edge device for the regulated space or other load; Receive one or more modification commands related to energy consumption and / or demand, the one or more modification commands including or activating the retrieval of modification data, the modification data including demand reduction command values, time-of-use energy rates during the day, weather information or a combination thereof; Power consumption data is received from the sensor to determine real-time energy consumption and / or demand data for the edge device electrical appliances used in the regulated space or other loads; The AI ​​software is used to generate energy consumption and / or demand modification signals, which are based on the adjustment data, the modification data, and the real-time energy consumption and / or demand data; and The edge device electrical appliances for the regulated space or other load are controlled based on the energy consumption and / or demand modification signal, wherein the energy consumption and / or demand modification signal includes a timing modification of an operating command for the electrical appliances to achieve energy consumption and / or demand reduction.

2. The method of claim 1, further comprising sending a sensor output from the edge device electrical component for updating the optimized thermostat or other control signal.

3. The method according to claim 2, wherein, The sensor output used to update the optimized thermostat or other control signals is continuously sent from the edge device electrical components.

4. The method of claim 1, further comprising estimating baseline energy consumption and / or demand data for the edge device appliances for the regulated space or other load based on the regulation data, and the energy consumption and / or demand modification signal further being based on the estimated baseline energy consumption and / or demand data.

5. A method for automatically controlling and managing the energy consumption and / or demand and operation of at least one edge appliance in an HVACR or other system, said at least one edge appliance including a hysteresis thermostat or other control and configured to operate based on a control signal, the method comprising the steps of: Receive OEM temperature control or other control command signals from the hysteresis thermostat or other controller; The controller receives a first AI command signal at its main control relay. The controller includes a main control relay and secondary control relays arranged in series with the main control relay. The AI ​​command signal is generated by an artificial intelligence (AI) cloud command center located remotely from the controller. The AI ​​cloud command center includes a processor that runs one or more AI control algorithms to generate the AI ​​command signal. The main control relay has an open state and a closed state, and the secondary control relay also has an open state and a closed state. When the main control relay is in the off state, a continuous ON command signal is transmitted to the secondary control relay through the main control relay, and the output of the main control relay is the AI ​​command signal. When the main control relay is in the closed state, its output is an OEM temperature control or other control command signal. When the secondary control relay is closed, the output of the primary control relay is sent to the AI ​​cloud command center, where it is processed by one or more AI control algorithms to form a processed signal. This processed signal is then transmitted as the control signal to the at least one edge device. When the secondary control relay is in the open state, regardless of the state of the primary control relay, an OFF command signal is generated as the control signal. as well as Based on the states of the main control relay and the secondary control relay, the control signal is sent to the at least one edge appliance.

6. The method according to claim 5, further comprising: Sensing weather-related or other performance variables in the vicinity of at least one edge appliance to form a sensed temperature or other performance variable input; Generate a temperature or other performance variable input signal based on the sensed temperature or other performance variable input; The temperature or other performance variable input signal is sent to the controller; as well as (i) sending the temperature or other performance variable input signal from the controller to the AI ​​cloud command center for processing by the one or more AI control algorithms and being taken into account when forming the processed signal, or (ii) adjusting the control signal at the controller to form an adjusted control signal, and then sending the adjusted control signal to the at least one edge appliance.

7. The method according to claim 5, further comprising: The ambient temperature in the regulated space to be heated or cooled by the at least one edge appliance is sensed to form the sensed ambient temperature; A temperature signal is generated based on the sensed ambient temperature; The temperature signal is sent to the controller; as well as (i) sending the temperature signal from the controller to the AI ​​cloud command center for processing by the one or more AI control algorithms and being taken into account when forming the processed signal, or (ii) adjusting the control signal at the controller to form an adjusted control signal, and then sending the adjusted control signal from the controller to the at least one edge appliance.

8. The method according to claim 5, further comprising: Inputting weather forecast temperature information at a location outside but adjacent to the regulated space to be heated or cooled by the edge appliance, the input including inputting the weather forecast temperature information into the one or more AI control algorithms; and Based on the input forecast temperature information, the control signal is adjusted at the AI ​​cloud command center to form an adjusted control signal.

9. The method according to claim 5, further comprising: Inputting current temperature information sensed at a location outside but adjacent to the space to be heated or cooled by the edge appliance, the input including inputting the sensed current temperature information into the one or more AI control algorithms; and Based on the input, sensed current temperature information, the control signal is adjusted at the AI ​​command center to form an adjusted control signal.

10. The method according to claim 5, wherein, The controller further includes a display, and the method further includes: Sending information display signals from the AI ​​cloud command center to the controller; and The information obtained from the information display signal is displayed on the display.

11. The method of claim 5, further comprising: The updated software version is sent from the AI ​​cloud command center to the controller; as well as Update the software in the controller based on the updated software version.

12. The method of claim 11, further comprising: Based on the updated software version, an update signal is sent from the controller to at least one edge appliance; as well as Based on the update signal, the software in the at least one edge appliance is updated.

13. The method according to claim 5, wherein, The energy supplying the at least one edge appliance comes from the power grid, and the method further includes: Receive information related to the energy use and demand of the power grid at the AI ​​cloud command center; and The AI ​​control algorithm is used to process information related to the energy use and demand of the power grid, and the AI ​​command signal is formed based on the information related to the energy use and demand of the power grid.

14. The method according to claim 5, wherein, The energy supplying the at least one edge appliance comes from the power grid, and the method further includes: The AI ​​cloud command center receives information relating to: (i) time-of-use energy prices during the day; and (ii) demand response and load shifting, load shedding, and virtual power plant opportunities with respect to the power grid; and The AI ​​control algorithm is used to process the information relating to: (i) time-of-use energy prices during the day, and (ii) demand response and load shifting, load shedding and virtual power plant opportunities of the power grid, and the AI ​​command signal is formed based on the information relating to: (i) time-of-use energy prices during the day, and (ii) demand response and load shifting, load shedding and virtual power plant opportunities of the power grid.

15. The method of claim 5, further comprising: The AI ​​control algorithm is used to estimate the heat capacity and thermal resistance of the regulated space to be heated or cooled by the at least one edge appliance; The AI ​​control algorithm is used to estimate the power consumption required to provide a comfortable environment in a regulated space that needs to be heated or cooled. It is estimated that power demand will be reduced and energy savings will be achieved by providing the hypothetical control signals that provide the comfortable environment; Based on the assumed control signal, a specific control signal is generated using the AI ​​control algorithm; and The specific control signal is sent from the AI ​​command center to the controller.

16. The method of claim 5, further comprising: Intercept original equipment manufacturer (OEM) thermostat or other control command signals from the hysteresis thermostat or other control to the at least one edge appliance at the controller; as well as Instead of sending the OEM thermostat or other control command signal to the at least one edge appliance, the AI ​​command signal is sent to the at least one edge appliance.

17. The method of claim 5, further comprising: The control signal is modified by adjusting the open and closed states of the secondary control relay, wherein the adjustment causes the at least one edge appliance to cycle on and off, and the cycle is timed to take advantage of the energy stored in the heat capacity of the space to be heated or cooled by the at least one edge appliance.

18. The method of claim 5, further comprising one or more of the following steps: (1) applying preheating or precooling using time-of-use energy costs during the day; (2) applying preheating or precooling based on forecasted weather information or data; (3) applying multiple device “on” time values ​​based on demand response command signals; (4) reducing the number of thermostat “on” requests if the estimated device size is too large; (5) regulating thermostat “on” requests to prevent temperature overshoot and undershoot; (6) regulating thermostat “on” requests to reduce “run” time; and (7) regulating thermostat “off” requests to increase off time.

19. The method of claim 5, further comprising: Receive regulation data relating to the temperature control or other control output of the at least one edge device electrical appliance, the regulation data including one or more of control inputs, state variable inputs and sensor inputs; The conditioning data is processed using one or more (AI) control algorithms to create an optimized thermostatic control signal for at least one edge device electrical appliance in the regulated space; The optimized constant temperature control signal controls at least one edge device electrical appliance in the regulated space. The heat capacity, thermal resistance, and internal load of the regulated space are adaptively estimated. The object model is run based on the measured measurement data and the estimated temperature setpoint, wherein the estimated temperature setpoint and the regulated space temperature are generated by the object model and applied to the hysteresis temperature controller model to create the estimated thermostat signal. as well as Energy savings are estimated based on the integral difference between the estimated thermostat signal and the control signal generated by the object model.

20. An edge node device for controlling at least one edge appliance in an HVACR or other system, the edge node device being configured to send control signals to the at least one edge appliance, the at least one edge appliance having a thermostat or other control, the edge node device comprising: The controller includes a main control relay and a first control relay arranged in series with the main control relay. The main control relay includes a first artificial intelligence (AI) command input, and the first control relay includes a second AI command input. Both the first AI command input and the second AI command input are configured to communicate with an AI command center to receive a first AI command signal and a second AI command signal, respectively. The main control relay has an open state and a closed state, and the first control relay also has an open state and a closed state. as well as The output terminal is configured to communicate with the at least one edge appliance and to send control signals from the controller to the at least one edge appliance, wherein the edge node device is configured such that: When the main control relay is in the off state, a continuous ON command signal is transmitted to the first control relay through the main control relay, and the output of the main control relay is the AI ​​command signal. When the main control relay is in the closed state, its output is a temperature control command signal generated by the thermostat. When the first control relay is in a closed state, the output of the main control relay is sent to the AI ​​command center for processing by one or more AI control algorithms to form a processed signal. This processed signal is then transmitted as the control signal to the at least one edge device. When the state of the first control relay is open, regardless of the state of the main control relay, an OFF command signal is generated as the control signal.

21. A network comprising an edge node device according to claim 20, and an AI command center, wherein, The AI ​​command center is configured to generate the first AI command signal.

22. The network according to claim 21, wherein, The AI ​​command center is located away from the edge node device.

23. The network according to claim 21, wherein, The AI ​​command center is a cloud-based command center.

24. The network according to claim 21, wherein, The AI ​​command center is located remotely from the edge node device and communicates with the edge node device via cloud-based computing.

25. The edge node device according to claim 20, wherein: The edge node device is configured to send multiple control signals to a plurality of corresponding edge electrical appliances, each of which has a constant temperature or other control input. The controller includes a main control relay and a plurality of secondary control relays, including the main control relay. Each of the secondary control relays is connected in series with the main control relay. Each of the secondary control relays includes a corresponding second AI command input, and each of the second AI command inputs is configured to communicate with the AI ​​command center to receive a corresponding second AI command signal; Each of the secondary control relays has an open state and a closed state; The control appliance has multiple corresponding output terminals, each of which is configured to communicate with a corresponding one of the multiple edge appliances and is configured to send a corresponding control signal from the controller to the corresponding edge appliance. The edge node device is configured such that, for each corresponding edge appliance among the multiple edge appliances, When the main control relay is in the off state, a continuous ON command signal is transmitted through the main control relay to the corresponding secondary control relay, and the output of the main control relay is the AI ​​command signal. When the main control relay is in the closed state, its output is a temperature control command signal generated by the thermostat. When the corresponding secondary control relay is closed, the output of the primary control relay is sent to the AI ​​command center for processing by one or more AI control algorithms to form a processed signal. This processed signal is then transmitted as the control signal to the corresponding edge device. When the corresponding secondary control relay is in the open state, regardless of the state of the primary control relay, an OFF command signal is generated as the control signal.

26. A network comprising an edge node device according to claim 20, and an AI command center, wherein, The AI ​​command center is configured to generate the first AI command signal.

27. A system comprising the edge node device, HVACR appliances, current sensor, and duct temperature sensor according to claim 20, wherein, The current sensor is configured to send a sensed current signal to the controller, the duct temperature sensor is configured to send a sensed duct temperature signal to the controller, and the controller is configured to send data related to the sensed current signal and the sensed duct temperature signal to the AI ​​cloud command center.

28. The system of claim 27 further includes an AI command center.

29. The system of claim 28 further includes a cloud-based learning database of selected data, wherein, The AI ​​command center communicates with and is configured to retrieve data from the cloud-based learning database of Selected Data.

30. A non-transitory computer-readable storage medium storing instructions, said instructions causing the computer to perform a process when executed by the computer, said process comprising: Receive regulation data relating to the thermostat or other control output of an edge device appliance in the regulated space or other load, the regulation data including one or more of control inputs, state variable inputs and sensor inputs; The regulation data is processed using artificial intelligence (AI) software to create optimized thermostatic or other control signals for edge device electrical appliances in the regulated space or other load. The edge device electrical appliances in the regulated space or other load are controlled based on the optimized constant temperature or other control signals. Receive energy consumption and / or demand modification commands, the energy consumption and / or demand modification commands including or activating the retrieval of modification data, the modification data including demand reduction command values, time-of-use energy rates during the day, weather information or combinations thereof; Power consumption data is received from sensors to determine the real-time energy consumption and / or demand data of edge device electrical appliances in the regulated space; The AI ​​software is used to generate energy consumption and / or demand modification signals, which are based on the adjustment data, the modification data, and the real-time energy consumption and demand data. as well as The edge device appliances in the regulated space or other load are controlled based on the energy consumption and / or demand modification signal, wherein the energy consumption and / or demand modification signal includes the timing modification of the ON / OFF command of the appliance to achieve energy consumption and / or demand reduction.

Images