Multi-source energy supply open-loop optimization central air conditioner control method and system

Abstract

The invention discloses a multi-source energy supply open-loop optimization central air conditioner control method and system independent of return air temperature closed-loop control. Aiming at the problems of control failure and energy consumption waste caused by air return parameter distortion of the semi-open space, an innovative strategy of humidity priority control and temperature interval expansion is provided, specifically, the humidity of a core functional area is used as a main control index, an RH dynamic threshold value (preferably + / -3% + / -10%) is set, the air supply temperature is allowed to fluctuate within a set range (preferably + / -2 DEG C + / -5 DEG C), and the air supply temperature is allowed to fluctuate within a set range (preferably + / -2 DEG C + / -5 DEG C); and calculating a human body comfort interval in real time through a PMV-PPD model. The system integrates a central air conditioner host, a distributed heat pump and photovoltaic / hydrogen energy equipment, adopts a millimeter wave radar and a multispectral sensor to construct an environment sensing network, and combines an edge reinforcement learning engine to realize multi-source energy supply dynamic scheduling. An AI comfort value recommendation system is innovatively introduced, a traditional air conditioner rigid temperature control normal form is broken through, and a full-scene energy-saving solution covering business, hotels and residences is formed.

Description

Technical Field

[0001] The present invention belongs to the technical field of heating, ventilation, air conditioning and building automation, and particularly relates to an optimization method and device for a central air conditioning system based on multi-source energy supply collaboration and humidity priority control, which is applicable to refined environmental regulation in semi-open spaces or scenarios where return air is uncontrollable. Background Art

[0002] Traditional central air conditioning systems rely on closed-loop control of return air temperature and need to adjust the cooling capacity through precise return air temperature feedback. However, in semi-open spaces (such as shopping mall atriums, airport waiting halls), due to air mixing, the return air parameters are distorted, leading to two major problems: one is control failure: the return air temperature cannot reflect the real load, resulting in overcooling / overheating; the other is energy consumption waste: in order to maintain a temperature control accuracy of ±0.5°C, the compressor starts and stops frequently, and the energy efficiency drops by more than 30%.

[0003] In addition, in the hotel guest room scenario in the southern region, the high humidity environment in spring and summer (the outdoor relative humidity is often higher than 85%) and the difference in the airtightness of the guest rooms together make it difficult for traditional air conditioning systems to stably maintain a comfortable environment.

[0004] The existing technical improvement directions focus on increasing the sensor density or strengthening the building envelope, but there are defects such as high renovation costs (≥500 yuan / m²) and damage to the spatial openness. Summary of the Invention

[0005] The present invention discloses an open-loop optimized central air conditioning control method and system for multi-source energy supply that does not rely on closed-loop control of return air temperature. Aiming at the control failure and energy consumption waste problems caused by the distortion of return air parameters in semi-open spaces, an innovative strategy of "humidity priority control + temperature range expansion" is proposed: taking the humidity in the core functional area as the main control index, setting a dynamic threshold for RH (preferably ±3% - ±10%), allowing the supply air temperature to fluctuate within a set range (preferably ±2°C - ±5°C), and calculating the human comfort interval in real time through the PMV-PPD model.

[0006] The system architecture of the open-loop optimized central air conditioning control system for multi-source energy supply of the present invention is as Figure 1 shown, including a hardware system and a software system. The hardware system includes an environmental perception layer and an execution device layer, and the software system includes an edge computing node and a user interaction platform ( Figure 2 ).

[0007] The environmental perception layer is equipped with a millimeter wave radar array and a multi-spectral infrared sensor: 1) Millimeter wave radar array: with a detection radius of 15m and a resolution of 5cm, it can track the personnel distribution and heat source position in real time; 2) Multi-spectral infrared sensor: synchronously monitors the surface temperature (accuracy ±0.2°C) and the material emissivity (distinguishing human / equipment heat generation).

[0008] The execution device layer is equipped with multi-directional air ducts and wireless power supply interfaces: 1) Multi-directional air ducts: Support 0-180° pitch angle adjustment, and stepless speed regulation of wind speed from 0.1 to 5 m / s; 2) Wireless power supply interface: Provide 5W power for the desktop air blower through magnetic resonance technology, with a transmission efficiency ≥ 85%.

[0009] The edge computing node includes an embedded reinforcement learning engine and a digital twin interface; 1) Embedded reinforcement learning engine: Generate control strategies every 10 seconds, and the number of model parameters is compressed to 800KB; 2) Digital twin interface: Import the BIM model to generate a CFD simulation scenario and preview the regulation effect.

[0010] The user interaction platform is equipped with a comfort democratic voting function and a carbon emission visualization dashboard; 1) Comfort democratic voting function: Users can vote through the APP to adjust the local temperature and humidity thresholds; 2) Carbon emission visualization dashboard: Real-time display of energy-saving contributions and carbon trading revenue distribution.

[0011] The working process of the multi-source energy supply open-loop optimized central air-conditioning control system includes three links: data collection and fusion, multi-source energy supply scheduling, and dynamic environment regulation.

[0012] Data collection and fusion: The millimeter-wave radar generates a personnel density heat map, and the infrared sensor captures the surface temperature field; Through the Kalman filter algorithm, multi-source data is fused to construct a three-dimensional environmental state matrix.

[0013] Multi-source energy supply scheduling ( Figure 3 ) The central air-conditioning host provides basic cooling capacity, and the local heat pump handles sudden loads; Based on the Nash equilibrium algorithm, the energy supply tasks are allocated to minimize the overall energy consumption.

[0014] Dynamic environment regulation: When the humidity deviates from the set value by ±2%, start the dehumidification / humidification equipment for priority regulation; Temperature regulation is only triggered when it exceeds the PMV-PPD comfort interval to avoid ineffective regulation.

[0015] The core innovation points of the present invention are reflected in the proposed new control method and the multi-source energy supply architecture.

[0016] Proposed new control method. The first is humidity priority control: In the scenario where the return air is uncontrollable, the humidity in the core functional area (±3% RH) is used as the main control index, and precise adjustment is achieved through the linkage of the condensation dehumidifier and the ultrasonic humidifier; The second is the dynamic expansion of the temperature range: Allow the supply air temperature to fluctuate within the range of ±2°C (traditionally required ±0.5°C), and dynamically adjust the acceptable temperature range through the human thermal comfort model (PMV-PPD).

[0017] Multi-source energy supply architecture: First, build a collaborative network of distributed cold and heat sources, integrating equipment such as central air-conditioning hosts, local heat pumps, and desktop air supply units, and allocate energy supply tasks through dynamic game algorithms; second, support the access of clean energy such as photovoltaic and fuel cells to achieve a load transfer of more than 30% of traditional air-conditioning loads. Description of the Drawings

[0018] Figure 1 : System architecture diagram.

[0019] Figure 2 : Schematic diagram of hardware deployment.

[0020] Figure 3 : Multi-source energy supply scheduling flowchart.

[0021] Figure 4 : Working process of the comfort value recommendation system.

[0022] Figure 5 : Schematic diagram of the hotel guest room control system. Detailed Implementation Manner

[0023] Example 1: Hotel guest room renovation. The most typical application scenario of the multi-source energy supply open-loop optimized central air-conditioning control system is the hotel guest room scenario. The following takes the hotel guest room renovation as an example for description.

[0024] In the hotel guest room scenario in the southern region, the high humidity environment (the outdoor relative humidity is often higher than 85%) in spring and summer and the difference in the airtightness of the guest rooms jointly cause it difficult for the traditional air-conditioning system to stably maintain a comfortable environment. In response to this problem, the present invention proposes a collaborative renovation plan for fan coil units based on humidity main control: on the premise of retaining the original central air-conditioning host, the intelligent upgrade of the guest room terminal equipment is carried out. In the installation and deployment stage, first, a microwave humidity sensor (detection accuracy ±1.5%RH, response time ≤ 3 seconds) is embedded in the center of the ceiling, and its signal line is connected in parallel with the original fan coil control box; millimeter-wave radar sensors are deployed at the switch positions on the ceiling and walls of the guest room, and the occupancy status of personnel is transmitted to the edge controller through a low-power Bluetooth module. The renovation of the fan coil unit body includes two core components - the water valve module is replaced with a floating-point regulating valve driven by a brushless motor to replace the traditional solenoid valve, and the opening resolution is increased to 0.1%, and it can complete 0-100% linear regulation within 2 seconds; the fan module is upgraded to a PWM variable-frequency motor, and the stepless switching of three-speed air volume is realized by adjusting the duty cycle, and a flow guide grille is installed inside the air duct to reduce the air flow noise to below 32dB.

[0025] The operation logic of the multi-source energy supply open-loop optimized central air-conditioning control system applied in the hotel is carried out according to the principle of humidity priority ( Figure 5): When the microwave sensor detects that the indoor humidity exceeds the set threshold (such as 53% RH), the PID algorithm is immediately triggered to calculate the required dehumidification amount, and the water valve opening is adjusted first. Specifically, the water valve control is dynamically adjusted based on the heat load difference (ΔT=set temperature-measured temperature): Under normal working conditions, every 1°C increase or decrease in ΔT corresponds to a 15% linear adjustment of the water valve opening; when there is a sudden surge in wet load caused by the opening of doors and windows (ΔT>3°C for more than 30 seconds), the system will increase the instantaneous opening of the water valve to 70% and maintain it for 5 minutes, and then gradually reduce the fan speed after the heat load falls back to a safe range. This collaborative logic can avoid mechanical wear and energy consumption spikes caused by frequent start and stop of fans in traditional systems. To adapt to the usage characteristics of the guest rooms, the control system introduces a state perception optimization strategy: the human body condition is monitored through millimeter wave radar. If there is no human activity for more than 30 minutes, it will automatically enter the energy-saving mode, relax the humidity control threshold to ±5% RH and lock the wind speed to low speed; the door and window magnetic sensors detect the opening and closing status in real time. When it is recognized that the window is open, the dehumidification output is automatically increased by 20% to compensate for the infiltration of external humid air.

[0026] The above key transformation links include three calibration operations: using a laser locator to ensure that the microwave sensor is facing the core activity area of ​​the guest room (bed and desk area); using a handheld anemometer to verify the actual coverage of the three-speed air volume (the low-speed gear must ensure that the wind speed at 0.5m from the bed is ≤0.2 m / s); using a humidity generating chamber to perform a step response test on the control system. After the transformation, the system maintains a stable indoor humidity of 50±2.7% RH during the rainy season, reducing the number of fan starts and stops compared to traditional solutions, reducing the average daily air conditioning power consumption of a single room, and eliminating the common over-drying problem of traditional dehumidifiers. The condensation phenomenon in the gaps of guest room doors and windows is basically eliminated, and the linen replacement cycle is extended. These effects reflect the dual advantages of multi-source energy supply open-loop optimization of central air-conditioning control systems in refined humidity control and comprehensive energy efficiency improvement.

[0027] Example 2: Application of semi-open space in waiting hall. Scene characteristics: large span space (length 100m × width 30m × height 15m), intermittent peak flow of people (2,000 people per hour), greenhouse effect caused by glass curtain wall. System configuration: 12 sets of multi-directional air vents are deployed along the columns (spacing 8m), 200 micro air blowers are installed under the waiting seats (wind speed 0.3m / s), and photovoltaic film is arranged on the ceiling (power generation capacity 25W / ㎡). Control strategy: 1) Humidity priority: Set 45% RH as the benchmark and use the dehumidifier to handle the passenger's respiratory humidity load; 2) Temperature flexibility: 26-30℃ fluctuations are allowed, and local strong cooling mode is activated during peak hours; 3) Crowd forecasting: Predict regional loads through ticket inspection data and adjust air supply 5 minutes in advance; 4) Implementation effect: The humidity control accuracy is ±2.8% RH, the energy consumption in the air-conditioning season is reduced, and the passenger complaint rate drops significantly.

[0028] Example 3: Application in the mall corridor. Scenario characteristics: A long and narrow enclosed space (width 3m × length 80m), cold quantity infiltration interference from shops on both sides, and radiant heat from glass display windows. System configuration: Ceiling-embedded linear air supply strips (1 air outlet per meter), installation of infrared barrier sensors at the shop facades, use of a phase change energy storage material wall (cool storage density 120 kJ / kg). Control strategy: 1) Dynamic zoning: Divide the corridor into 5m segment units for independent control; 2) Radiation compensation: Start directional cooling when the surface temperature of the display window is detected to be >35°C; 3) Cooperative energy supply: Use the cold energy released by the energy storage material at night to replace the operation of the air conditioner; Implementation effect: The temperature difference in the corridor ≤1.5°C, the peak power demand is significantly reduced, and the electricity cost of the shop air conditioners drops.

[0029] Example 4: Application in the school cafeteria and gymnasium. Scenario characteristics: The cafeteria has a high humidity load (cooking steam + dense population), and the gymnasium is used intermittently with high intensity (CO2 concentration >2000 ppm during competitions). System configuration: Deployment of a condensation dehumidification membrane (dehumidification capacity 4 L / h·m²) on the cafeteria ceiling, integration of air supply holes in the gymnasium stands (PM2.5 filtration efficiency 99%), and arrangement of an air curtain unit in the corridor (wind speed 3 m / s, angle 30° inclined). Control strategy: 1) Steam tracking: Locate the steam source through a VOC sensor and start local exhaust; 2) CO2 linkage: Activate the fresh air compensation mode when the concentration exceeds 1500 ppm; 3) Class schedule synchronization: Preheat / pre-cool the venue 1 hour before class; Implementation effect: The humidity in the cafeteria is maintained at 50 ± 3%, the air age in the gymnasium is shortened to 8 minutes, and the annual operation cost is reduced.

[0030] Example 5: Application in a villa. Scenario characteristics: Independent requirements for multiple rooms (the elderly room needs 27°C / the children's room needs 24°C), presence of special spaces such as a sunroom and a basement. System configuration: Deployment of a magnetic micro air conditioner (cooling capacity 500W) in each room, installation of a ground source heat pump in the courtyard (COP 4.8), and integration of a dehumidification module in the smart wardrobe (humidity setting 40% RH). Control strategy: 1) Personalized settings: Define "reading mode" / "sleep mode" through the mobile APP; 2) Radiation balance: The floor heating and the ceiling cold radiation work together; 3) Photovoltaic priority: Preferentially use the rooftop photovoltaic power supply on sunny days (self-sufficiency rate 72%); Implementation effect: The independent error of room temperature control is ±0.8°C, the annual air-conditioning electricity cost is greatly saved, and there is no mildew problem in the basement.

[0031] Example 6: Application in basement and parking garage. Scenario characteristics: High humidity (RH>85% in the garage during the rainy season), exhaust gas accumulation (safety control of CO concentration). System configuration: Install explosion-proof dehumidifiers (dehumidification capacity of 30L / h) on columns, arrange jet fans (air volume of 3000m³ / h) above parking spaces, and set up VOC sensor arrays in the charging pile area; Control strategy: 1) Humidity threshold: Start dehumidification when RH>75%, and stop when <60%; 2) Exhaust gas dilution: When the CO concentration reaches 25ppm, enhance ventilation by 3 times; 3) Traffic flow perception: Predict load changes through geomagnetic sensors; Implementation effect: The humidity is stabilized at 65±5%, the peak CO concentration is reduced, and the start-stop frequency of equipment is greatly reduced.

Claims

1. A multi-source energy supply open-loop optimization central air conditioning control method, characterized in that The following steps are involved: a) Taking the humidity of the core functional area as the main control indicator, the relative humidity control threshold is set to a specific value within the range of ±1%-30% RH; b) Allow the supply air temperature to fluctuate dynamically within the range of ±2°C to ±10°C, and calculate the acceptable temperature range in the current scenario through the PMV-PPD model; c) Generate recommended parameters for environmental comfort values ​​based on AI algorithms, which are dynamically calculated based on local climate data, space usage habits, and human tolerance models; d) Integrate central air-conditioning units, distributed heat pumps and clean energy equipment to build a multi-source energy supply coordination mechanism.

2. The central air conditioning control method and system according to claim 1, characterized in that The preferred range of the humidity control threshold is ±3%-±10% RH, and the preferred range of the temperature fluctuation range is ±2°C-±5°C.

3. The method for generating recommended parameters of environmental comfort values ​​according to item c of claim 1, comprising: 1) Collect historical environmental data (temperature, humidity, and intensity of human activity) and user feedback scores through IoT terminals; 2) Establish a three-layer neural network model, with input dimensions including: longitude and latitude coordinates, seasonal parameters, spatial function type, and cultural habit index; 3) Output dimensions include: recommended temperature range, humidity threshold, and wind speed upper limit; 4) Dynamically update model weights every quarter to adapt to climate change and evolution of behavioral patterns.

4. The central air conditioning control method and system according to claim 1 are characterized in that: 1) Determine the activity status of personnel through millimeter wave radar; 2) Establish a humidity-PID-valve opening mapping table to prioritize dehumidification capacity; 3) Design a coordinated control sequence of "adjusting the valve first and then adjusting the speed" to avoid frequent starting and stopping of the fan.

5. The water valve control method according to claim 4, characterized in that: 1) The floating valve opening and the heat load difference ΔT are in a piecewise linear relationship: 2) When ΔT≤2℃, opening degree = 15%×ΔT; 3) When ΔT>2℃, opening degree = 30% + 10%×(ΔT-2); 4) The valve dead zone is set to ±0.3℃ to suppress oscillation.

Images