Method and system for transforming central air-conditioning terminal system based on function application microenvironment

Abstract

The invention discloses a functional application microenvironment-based central air-conditioning terminal system transformation method and system, and the core innovation of the method is as follows: 1) dynamic microenvironment division: defining a minimum 4m < 2 > control unit according to the actual use demand, and breaking through the physical boundary limitation; (2) multi-source cold and heat collaboration: 15 types of equipment such as a central air conditioner, an independent air conditioner and an electric heater are compatible, and energy is dynamically distributed according to an energy efficiency-preference-electricity price model; and (3) the central air-conditioning microenvironment controller integrates the adaptive capability of day and garden hardware (TY-HY-A / B valves) and multiple protocols, and realizes the temperature difference control precision of + / -0.5 DEG C. The scheme is adaptive to the energy-saving transformation of existing buildings, especially the energy-saving transformation of central air-conditioning fan coils of hotel guest rooms, the energy-saving transformation of office buildings, the energy-saving transformation of data centers and other application scenes.

Description

Technical Field

[0001] The present invention belongs to the field of intelligent buildings and HVAC control, and specifically relates to a method and system for retrofitting the terminal system of a central air conditioner with a functional application microenvironment as the smallest control unit. Through dynamic space division, multi-source cooling and heating coordination, and federated learning optimization, this solution realizes precise temperature and humidity control at a granularity of 4 square meters, and is suitable for energy-saving retrofitting of existing buildings and new construction projects. Background Art

[0002] The traditional central air conditioning system has the following core problems: 1) Rigid control unit: relying on physical partitions to divide areas, resulting in uneven cooling and heating in open spaces (temperature difference ≥ 3°C); 2) Serious energy efficiency waste: uniform air supply across the whole area leads to excessive energy supply in low-utilization areas (COP drops by 25 - 40%); 3) Poor equipment coordination: the terminal valves, fan coil units and third-party equipment operate independently, lacking a global optimization logic.

[0003] Although existing technologies such as CN217604319U (an adaptive air conditioner fan coil water valve control device and air conditioner) optimize local regulation, they do not solve the problems of multi-source coordination and dynamic microenvironment division. The present invention integrates three core patents applied by the applicant (CN2025105094107, CN2025105175068, CN202520770294X) and proposes the following innovative solutions: 1) Dynamic definition of microenvironment: defining the control unit according to actual usage requirements (rather than physical boundaries); 2) Cross-hardware protocol compatibility: supporting the plug-and-play of 15 types of third-party equipment (air conditioners / electric heaters / fresh air, etc.); 3) Dynamic inert regulation: realizing "energy supply on demand and energy supply according to people" based on millimeter-wave radar and infrared thermal maps. Summary of the Invention

[0004] The system architecture of the present invention is as Figure 1 shown, and is composed of a sensing layer, a cloud platform, an edge control layer and an execution layer.

[0005] The main hardware modules of the sensing layer include millimeter-wave arrays, infrared array sensors, VOC sensors, etc. The millimeter-wave array uses a 60GHz millimeter-wave radar (detection radius 8m, accuracy ±0.1m); the infrared array sensor has a resolution of 640×480 and a refresh rate of 30Hz.

[0006] The cloud platform adopts federated learning optimization: aggregating multi-building data to train a global optimization model (differential privacy encryption).

[0007] The main device of the edge control layer is the central air-conditioning microenvironment controller, and its hardware modules include: 1) Integration of the logic of Tianyuan fan coil controllers (integration of functions of types A / B / C); 2) RS485 bus interface (compatible with Modbus RTU / BACnet / IP protocol conversion); 3) Wireless communication module (Wi-Fi 6 + BLE 5.2 dual-mode, transmission distance ≥ 50m); 4) Floating-point adjustment module (resolution of water valve opening 0.1%); 5) Building control communication interface (supporting OPC UA and KNX protocols).

[0008] Core functions of the central air-conditioning microenvironment controller: 1) Dynamically receive data from millimeter-wave radar and infrared sensors, and generate control instructions according to the results of microenvironment division; 2) Coordinate the coordinated actions of the electric ball valve TY-HY-A, intelligent energy-saving valve TY-HY-B and six-way deflector device; 3) Receive global optimization parameters through the federated learning platform to achieve the optimal energy efficiency across microenvironments.

[0009] The execution layer consists of the hardware for fan coil unit transformation (electric ball valve TY-HY-A, intelligent energy-saving valve TY-HY-B, fan coil controller), six-way deflector air outlet, and third-party devices.

[0010] 1. Electric ball valve TY-HY-A, function: Replace the traditional electric two-way valve, achieve a sealing pressure ≥ 10Bar through a highly lubricated PTFE valve seat, and a leakage rate ≤ 0.1%; Control interface: Support 4-20mA analog signal / PWM control, and adapt to the instructions of the central control execution unit.

[0011] Intelligent energy-saving valve TY-HY-B, function: Integrate a temperature sensor and an adaptive algorithm to dynamically adjust the water flow (reduction ≥ 50%); Communication protocol: Built-in Modbus RTU interface (baud rate 9600bps), support direct connection to the building system.

[0012] Fan coil controller. The functions of the original type A / B / C controllers are integrated into the central air-conditioning microenvironment controller, and the following modules are retained: 1) Floating ball valve output module (resolution 0.1%); 2) RS485 bus interface (compatible with BACnet / IP conversion); 3) Adaptive multi-parameter control algorithm (cooling and heating load - flow mapping model).

[0013] Six-way deflector air outlet, air supply angle 0 - 180°, air volume 0.1 - 5m³ / s.

[0014] The software system of the present invention includes a microenvironment dynamic partitioning engine, a multi-source cold and heat coordination mechanism, and a user interaction interface.

[0015] Microenvironment dynamic partitioning engine. Input parameters: personnel density (persons / ㎡), equipment heat load (W / ㎡), space function label (meeting / warehousing); algorithm logic: python def microenvironment_partitioning(grid_cell_list): heat_load_threshold = 120W / ㎡ if function_label == "office" else 300W / ㎡ return [merge_adjacent_cells(cell) for cell in grid_cell_list if cell.heat_load >= heat_load_threshold]

[0016] Dynamic update: Refresh the boundary every 5 minutes based on sensor data, and support a minimum cell size of 4㎡.

[0017] 2---Multi-source cold and heat coordination mechanism. The connection architecture is shown as Figure 8 shown. Multi-protocol adapter: Built-in 15 communication modules, supporting parallel processing of device instructions with different protocols; Protocol conversion logic: Convert non-standard protocols (such as manufacturer-customized ASCII) to a unified JSON format through middleware; Offline compatibility: In the case of network disconnection, local protocols (such as Modbus, KNX) can still maintain basic control functions.

[0018] Compatible devices: 15 types of devices such as central air-conditioning terminals, independent air conditioners, electric heaters, solar heating, etc. (see Table 1 for details); Table 1: Connection methods of third-party devices Device Type Connection Method Protocol / Interface Type Description 1. Split Wall-mounted Air Conditioner Infrared Signal Learning + Wi-Fi Relay Infrared Remote Control + HTTP API Simulate remote control commands through the infrared emission module and achieve remote status feedback through Wi-Fi access 2. Floor-standing Commercial Air Conditioner RS485 Wired Connection Modbus RTU Connect to the main control board of the air conditioner, read the operating parameters (temperature, wind speed) and send control commands 3. Smart Electric Heater Zigbee 3.0 Wireless Zigbee HA 1.2 Access through the Zigbee gateway, support power adjustment (10%-100% in 5% steps) 4. Geothermal Heat Pump Sub-unit CAN Bus CANopen Connect to the heat pump controller, obtain the inlet and outlet water temperatures and compressor status, and adjust the energy supply ratio 5. Solar-assisted Heating Unit Custom Serial Port Custom ASCII Protocol Access through the RS232 / RS485 converter, and a protocol parsing middleware is required 6. Fresh Air Unit BACnet / IP BACnet MS / TP Control the air volume and start / stop of heat recovery by converting the protocol through the BACnet router 7. Smart Radiator Bluetooth Mesh Bluetooth 5.2 Network to control the water temperature and valve opening, support offline scene presetting 8. Air-source Heat Pump Water Heater KNX Bus KNX TP1 Access through the KNX / IP interface, monitor the energy efficiency ratio and optimize the operation period 9. Radiant Floor Heating System DALI Bus DALI-2 Adjust the valves of the manifold and the rotation speed of the circulation pump, support zone temperature control 10. Dehumidifier 433MHz Radio Frequency Custom Radio Frequency Encoding Driver Learn the original factory remote control encoding, support humidity threshold setting and linkage start 11. Smart Humidifier Wi-Fi Direct Connection MQTT over TLS Obtain the device status through the cloud API and directly issue control commands locally (to avoid cloud latency) 12. Variable Refrigerant Flow (VRF) Dedicated Communication Line Manufacturer's Protocol (such as Daikin CNET) Connect through the protocol conversion gateway (such as Mitsubishi MELSEC) to Modbus TCP 13. Air Purifier Infrared + Zigbee Dual-mode Infrared Encoding + Zigbee Infrared controls basic functions, and Zigbee transmits sensor data such as PM2.5 14. Smart Curtain Motor Z-Wave Z-Wave Plus Adjust the opening degree to assist in temperature control (such as automatically closing when the sunlight is strong), and a Z-Wave gateway is required 15. Energy Storage Electric Radiant Floor Heating LoRaWAN LoRaWAN 1.0.3 Remotely set the heating strategy during off-peak electricity hours, support offline operation

[0019] Dynamic weight model: P_weight = 0.6 × energy supply efficiency + 0.3 × user preference + 0.1 × real-time electricity price.

[0020] Lazy regulation: When there is an excess of multi-source energy supply, the valve opening of the central air conditioner is preferentially reduced (step 0.1%) to reduce the load on the host.

[0021] User interaction interface: Supports three-mode control of APP / voice / panel, and presets 12 scenario modes.

[0022] The present invention integrates cross-patent technologies: Spatial grid computing (CN2025105094107) provides benchmark cooling and heating loads and carbon emission data; Multi-directional air guiding device (CN2025105175068) realizes microenvironment directional air supply; Federated learning group control (CN202520770294X) optimizes the global energy efficiency strategy.

[0023] Figure 7 It is the renovation implementation process, and the steps are described as follows: 1) Remove the old valve: Unload the traditional electric two-way valve and retain the DN20-DN50 interface; 2) Install the new valve: Connect TY-HY-A / B to the pipe flange (torque 25N·m ± 5%); 3) Network configuration: Connect the valve to the central control unit through the RS485 bus (address code 1-255); 4) Calibration test: Run the flow-opening calibration program (e.g., 0-100% opening corresponds to 0.5-5m³ / h); 5) Model distribution: The cloud federated learning model pushes the optimal control parameters (e.g., Kp = 0.8, Ki = 0.2).

[0024] The device-level innovation of the present invention includes: The central control microenvironment controller realizes the integration of "perception - calculation - execution", replaces the traditional decentralized controller group, and reduces the number of hardware by 60%; The dual-engine architecture with the built-in Tianyuan algorithm and the federated learning model improves the response speed by 3 times compared with the single control strategy.

[0025] The technical integration value of the present invention is reflected in: 1) Through the seamless connection between the TY-HY-A / B valve and the central control unit, cross-domain collaboration of the hydraulic system and airflow control is realized, and the energy efficiency is significantly improved compared with the traditional renovation plan; 2) Retain the core algorithm of the fan coil controller but remove the independent hardware, further reducing the renovation cost; 3) As a standard off-the-shelf component, the TY-HY-A / B valve, the central air-conditioning microenvironment controller of the present invention will retain and adopt its core algorithms (such as calibration process, control parameter mapping); 4) The multi-parameter algorithm of the central air-conditioning microenvironment controller is redeveloped to form a differentiation from the original controller (the response speed is optimized from 2 seconds to 0.5 seconds).

[0026] Technical compatibility breakthrough of the present invention: The first "dynamic microenvironment + multi-source collaboration + inert regulation" trinity control architecture is created, improving energy efficiency compared to traditional systems; The first multi-source cold and heat mixing control architecture is created, breaking through the single energy supply limitation of traditional systems, compatible with 15 categories of third-party devices, covering mainstream devices in the three major fields of HVAC, smart home, and new energy, solving the problem of multi-brand heterogeneous system collaboration; The protocol conversion middleware realizes "one development, multi-device adaptation", greatly reducing the cost of customization and transformation. Description of the Drawings

[0027] Figure 1 : System architecture diagram. The four-layer architecture realizes the perception-decision-execution closed loop. Green is for environmental perception, blue is for edge control, purple is for cloud optimization, and red is for execution devices.

[0028] Figure 2 : Schematic diagram of dynamic microenvironment division. The grid overlays the heat map to generate the microenvironment (yellow is the original plane, green is the final partition).

[0029] Figure 3 : Cross-sectional view of the six-way air outlet structure. The blue magnetic interface connects to the air duct, the purple diversion channel realizes six-way air supply, and the gray hose extends the air supply distance.

[0030] Figure 4 : Flowchart of federated learning model training. Green is for data upload, red is for model download, forming a closed-loop optimization.

[0031] Figure 5 : Installation and configuration flowchart. Yellow is for removing old equipment, blue is for installing new hardware, green is for network configuration, purple is for calibration, and red is for model download.

[0032] Figure 6 : Prototype diagram of the APP interaction interface. Blue is for the temperature control module, green is for air supply control, and gray is for energy efficiency analysis.

[0033] Figure 7 : Transformation implementation flowchart. Five-step transformation process (demolition → installation → networking → calibration → download).

[0034] Figure 8 : Multi-source cold and heat source collaborative control diagram. Demonstrates the access and control process of multi-source devices. Blue is for the central air conditioner, red is for the wall-mounted air conditioner, yellow is for the electric heater, and green is for the solar device. Embodiment

[0035] Embodiment 1: Transformation of a smart office building. Scenario: The open office area needs to independently control the temperature according to workstations. Solution: 1) Divide a 4㎡ microenvironment and deploy millimeter-wave radar + infrared sensors; 2) The six-way air outlet distributes the air volume according to the heat load ( +30% in the high-heat area); 3) Automatically switch to full-area mixed mode (temperature difference relaxed to ±1.5℃) when no one is around at noon; Effect: Reduced energy consumption and improved user satisfaction.

[0036] Example 2: Group control of hospital ICU wards (supplement). Solution details: 1) Each ward is equipped with a central control micro-environment controller, which is directly connected to the TY-HY-A valve and millimeter-wave radar; 2) The controller is connected to the hospital HIS system through the OPC UA protocol to achieve automatic switching of infection prevention and control modes.

[0037] Example 3: Energy saving in the dining area of ​​a shopping mall. Scenario: The hot pot area needs to dissipate heat quickly, and the beverage area needs to maintain constant temperature and humidity. Solution: 1) The microenvironment of the hot pot area is marked as high heat load (k3=0.5); 2) Directional air supply + negative pressure exhaust coordination (air changes ≥ 15 times / h); 3) The federated model learns the passenger flow pattern and pre-cools / pre-heats 30 minutes in advance; Effect: Overall energy consumption decreased and the complaint rate decreased.

[0038] Example 4: Precision cooling in data centers. Scenario: Local overheating of a cabinet cluster requires dynamic compensation. Solution: 1) Divide the microenvironment by cabinet power density (threshold 3000W / ㎡); 2) The top air outlet is deflected 60 degrees to blow directly to the high-heat area; 3) Water temperature is dynamically adjusted according to IT load (ΔT=1℃ / 10% load); Effect: PUE was optimized from 1.6 to 1.25, saving considerable electricity bills annually.

[0039] Example 5: Five-star hotel banquet hall (supplement). Control logic: 1) The central control micro-environment controller receives wedding banquet schedule data in real time and starts pre-cooling / pre-heating one hour in advance; 2) 12 TY-HY-B energy-saving valves are synchronously controlled through RS485 bus, and the water flow error is ≤±2%.

[0040] Example 6: Multi-source collaboration in a school lecture hall. Scenario: A classroom with 300 people needs to be zoned for temperature control and ensure air quality. Solution: 1) The central air-conditioning outlet is linked with the independent dehumidifier (starts when CO2>800ppm); 2) The top guide device scatters the air (avoids direct blowing), and the wind speed is adjustable from 0.8 to 1.2 m / s; 3) Federated learning optimizes the pre-cooling strategy for class schedules; Effect: Energy consumption during the air - conditioning season is reduced, and students' concentration is improved.

[0041] Example 7: Energy - saving renovation of hotel rooms. Scenario: A chain hotel needs to reduce the air - conditioning energy consumption in its rooms. Solution: 1) Valve replacement: Replace all valves in the building with TY - HY - A electric ball valves + TY - HY - B energy - saving valves; 2) Control integration: Incorporate the functions of the original fan - coil unit controllers into the central control system; 3) Strategy optimization: In unoccupied rooms, the water temperature is automatically raised to 28°C (energy - saving mode) and restored to the set temperature 10 minutes after occupancy; Effect: The daily power consumption per single room is reduced from 8 kWh to 3.5 kWh, and water consumption is reduced by 30%.

[0042] Example 8: Renovation of a hybrid - energy - supply office area. Scenario: In an open - plan office area, some employees use electric heaters, resulting in uneven heating and cooling. Solution: 1) Equipment connection: The six - way deflector air outlets of the central air - conditioner cover the main area; Electric heaters are connected to the controller via Zigbee (maximum power limit 2000W); 2) Control strategy: When an electric heater is turned on, the opening degree of the water valve of the central air - conditioner is reduced by 30%, and the air outlet switches to the background air - supply mode (wind speed 0.5 m / s); During peak electricity - price periods (￥1.2 / kWh), low - energy - efficiency devices (such as electric heaters with an energy efficiency ratio < 2.0) are automatically disabled; Effect: The comprehensive energy - efficiency ratio in the hybrid - energy - supply scenario is increased to 3.8, and the number of illegal electricity - use incidents is reduced.

[0043] Example 9: Multi - source collaboration in hotel suites. Scenario: In a hotel suite, the central air - conditioner and bathroom electric warm air blower are used simultaneously. Solution: 1) Dynamic priority setting: Mark the bathroom electric warm air blower as having a temporary high priority (weight + 50%); 2) Cross - device linkage: When the electric warm air blower is enabled, the air outlet of the central air - conditioner avoids the bathroom area (the diversion angle is offset by 45°); 3) Energy consumption balance: According to the predicted length of stay (such as > 2 hours), switch to the central air - conditioner for main energy supply 10 minutes in advance; Effect: The heating speed in the bathroom is increased, and the overall energy consumption is reduced.

[0044] Example 10: Renovation of a hotel with mixed protocols. Scenario: A five - star hotel uses Daikin VRF, Siemens KNX curtains, and Xiaomi humidifiers simultaneously. Solution: 1) Protocol conversion: VRF is converted to Modbus TCP through a MELSEC gateway; Curtains are connected through a KNX / IP interface; Humidifiers are directly connected via Wi - Fi MQTT; 2) Collaborative control: When the curtain is closed, the VRF cooling capacity is reduced by 20%, and the humidity setting of the humidifier is increased by 10%; at night, it switches to the ground-source heat pump for main energy supply (COP > 4.0); Effect: The collaborative efficiency of cross-brand devices is effectively improved, and the operation and maintenance costs are reduced.

Claims

1. A method for transforming a central air-conditioning terminal system based on a functional application microenvironment, characterized in that include: a) Divide the building space into 2m×2m grid units and dynamically aggregate them into functional microenvironments; b) Deploy millimeter-wave radar, infrared sensor and central air-conditioning micro-environment controller for each micro-environment; c) Optimize the global control strategy based on the federated learning model, support independent temperature control and global hybrid mode switching, and support multi-source cooling and heating coordination and inertial regulation.

2. The method according to claim 1, characterized in that: The minimum size of the microenvironment is 4 square meters and it is allowed to exist across physical partitions; the dynamic division frequency is 5-15 minutes, depending on the population density and heat load changes.

3. The method according to claim 1, characterized in that: The multi-directional air outlet adopts the six-way adjustment structure of CN2025105175068, and the air supply angle error is ≤±1°; the terminal air volume is dynamically adjusted according to the formula Q=Qbase×(1+0.3×ΔT).

4. The system according to claim 1, characterized in that: The dual millimeter-wave radar array integrated with CN202520770294X has a detection blind area of ​​≤0.5㎡; the edge controller has a built-in carbon emission tracking module and is linked to the BIM building materials database.

5. The system according to claim 1, characterized in that: The central control microenvironment controller integrates the adaptive multi-parameter algorithm of the Tianyuan fan coil controller, including an RS485 bus interface, a floating point adjustment module and a wireless communication module; it supports data interaction with a third-party building system (such as Honeywell EBI) through the BACnet / IP protocol, and the response delay is ≤100ms.

6. The system according to claim 1, characterized in that: The electric ball valve (TY-HY-A) and the intelligent energy-saving valve (TY-HY-B) are connected in parallel to the central control unit via the RS485 bus, and the address codes are allocated in order of physical positions; the valve opening calibration adopts piecewise linear interpolation method, and a calibration point is set for every 5% opening.

7. The system according to claim 1, characterized in that: The central air-conditioning micro-environment controller is deployed in each functional micro-environment, with a maximum control area of ​​≤50㎡; the controller integrates a multi-protocol communication module (infrared / Wi-Fi / Zigbee / Modbus) and supports plug-and-play of third-party devices.

8. The system according to claim 1, characterized in that: The central air-conditioning micro-environment controller supports the access of mixed cold and heat sources, including central air-conditioning terminals, independent air-conditioning and heating equipment; the dynamic allocation strategy calculates the weights based on energy supply efficiency, user preferences and real-time electricity prices, and the control command response time is ≤1 second.

9. The system according to claim 1, characterized in that: The multi-protocol adapter supports 15 communication standards including infrared, Modbus, Zigbee, KNX, and LoRaWAN; the protocol conversion middleware unifies non-standard instructions into JSON format, and the fields include device_id, command, and value.

Images