A multi-pass air conditioning system and a method for operating the same

Abstract

A multi-channel air conditioning system comprises a multi-channel unit, a central controller, a human-computer interaction terminal and an alarm module; the multi-channel unit comprises at least three independent ventilation channels, each of which is provided with a fan, an electric sealing valve at an air inlet side, an electric sealing valve at an air outlet side and a sensor array; the central controller is connected with the fan, the electric sealing valve and the sensor array of each channel, and the central controller dynamically allocates the ventilation channels as a main channel, a standby channel or a dormant channel according to sensor data; when the real-time cleanliness of the main channel is lower than a set threshold or the filter resistance exceeds a final resistance threshold, channel switching is triggered; the human-computer interaction terminal is connected with the central controller and is used for receiving the input of the cleanliness threshold, the final resistance threshold and the channel priority.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of air conditioning systems, and particularly discloses a multi-channel air conditioning system and an operation control method thereof. BACKGROUND

[0002] Although the traditional central air conditioning system (such as a multi-split VRV / VRV system) has realized flexible control of "one drags many", its core still relies on a single refrigerant circulation pipeline and centralized indoor and outdoor unit linkage, but has the following limitations: single-channel design is prone to cause system shutdown due to problems such as filter blockage and fan failure, and lacks a redundant switching mechanism; energy efficiency optimization is insufficient, traditional variable frequency technology can adjust the speed of the compressor, but does not dynamically allocate the load in combination with multi-dimensional parameters such as time-of-use electricity prices and filter life; and device failure relies on manual inspection, and lacks a predictive maintenance strategy based on a health index (such as differential pressure and current harmonic). SUMMARY

[0003] In order to overcome the shortcomings and deficiencies in the prior art, the purpose of the present application is to provide a multi-channel air conditioning system and an operation control method thereof.

[0004] To achieve the above-mentioned purpose, a multi-channel air conditioning system according to the present application comprises a central controller, a human-computer interaction terminal, an alarm module and a multi-channel unit; the multi-channel unit comprises at least three independent ventilation channels, each channel is provided with a fan, an inlet-side electrically sealed valve, an outlet-side electrically sealed valve and a sensor array; the central controller is electrically connected with the fan, the inlet-side electrically sealed valve, the outlet-side electrically sealed valve and the sensor array of each channel, and the central controller dynamically allocates a main channel, a standby channel or a dormant channel according to the sensor array; when the real-time cleanliness of the main channel is lower than a set threshold value or the filter resistance exceeds a final resistance threshold value, channel switching is triggered; the human-computer interaction terminal is connected with the central controller, and is used for receiving the input of the cleanliness threshold value, the final resistance threshold value and the channel priority; the alarm module is electrically connected with the central controller, and when the fan of each channel fails, the electrically sealed valve is not closed and the sensor array data is abnormal, the central controller triggers the alarm module, and the alarm module pushes alarm information to the user through a sound and light alarm.

[0005] The sensor array of each channel comprises a differential pressure gauge arranged at the front and rear ends of the channel filter, which is used for monitoring the real-time resistance value of the channel, a particulate matter sensor arranged in the channel, which is used for detecting the PM2.5 and PM10 concentrations to calculate the cleanliness level, a temperature and humidity sensor arranged in the channel, which is used for collecting the air supply temperature and humidity data, and a fan current detection module, which is a Hall current sensor arranged in the fan power supply line or control cabinet, and is used for real-time acquisition of current signals to identify fan operation abnormalities.

[0006] The central controller comprises a programmable logic controller (PLC) and an AI computing unit. The programmable logic controller (PLC) is used to collect sensor data of each channel and fan operation state in real time, dynamically assign channel roles as a primary channel, a backup channel or a dormant channel, and predict equipment failure based on sensor data and generate maintenance instructions. The AI computing unit operates asynchronously independently of the PLC, trains a filter clogging prediction model based on historical resistance value data, and generates an optimal channel scheduling scheme based on time-of-use electricity price data and channel energy efficiency parameters. The filter clogging prediction model periodically pushes optimization strategies to the programmable logic controller (PLC) after generating an optimal channel scheduling scheme.

[0007] The dynamic allocation logic of the channel role is that the selection condition of the primary channel is the highest health index and the optimal energy efficiency level, the backup channel is sorted according to the preset priority, the priority is dynamically adjusted according to the historical failure rate and the maintenance period, and the dormant channel closes the fan and the electric sealing valve to save energy when the system load is less than 50%.

[0008] The central controller is configured to include multiple switching logics, which are emergency switching when the real-time cleanliness N of the primary channel is less than a set threshold N0, the resistance P is greater than a set threshold P0, or the fan current is abnormal, optimization switching when the health index of the backup channel is higher than that of the primary channel by 10%, and load balancing switching when the load of the primary channel exceeds 80% to activate the dormant channel to share the air volume.

[0009] The fan comprises a rack, a plurality of air supply mechanisms arranged on the rack, and a refrigeration mechanism. The first air outlet of the air supply mechanism is connected with the refrigeration mechanism, and the second air outlet is arranged at the end of the refrigeration mechanism away from the air supply mechanism. The air supply mechanism comprises a shell, a rotating disc arranged in the shell, a plurality of groups of blades arranged on the rotating disc, and a driving member for driving the rotating disc to rotate. The shell side is provided with an air inlet. The driving member drives the rotating disc to rotate to drive the blades to rotate and blow the air flow into the refrigeration mechanism to form cold air. The cold air is blown out through the second air outlet to supply air for the ventilation channel.

[0010] The driving member is a permanent magnet synchronous motor with multi-stage speed regulation function. A temperature sensor is arranged on the evaporator of the refrigeration mechanism. The speed control signal of the driving member is linked with the temperature sensor of the evaporator of the refrigeration mechanism. When the temperature sensor detects that the surface temperature of the evaporator is less than a preset value, the driving member automatically switches to a low speed mode to reduce the frosting risk of the evaporator.

[0011] An ultraviolet germicidal lamp is arranged in each ventilation channel of the multi-channel unit. The ultraviolet germicidal lamp is linked with the electric sealing valve on the air inlet side. When the valve is closed, it is automatically turned on to periodically disinfect the inside of the channel.

[0012] The air conditioning system comprises a multi-channel unit, a central controller, a human-computer interaction terminal and an alarm module. The multi-channel unit is provided with at least three independent ventilation channels, each of which is provided with a fan, a two-way electric sealing valve and a sensor array. The central controller is connected with each actuator and sensor, and performs dynamic allocation and switching control of channel roles. The human-computer interaction terminal provides a parameter setting interface, and the alarm module feeds back the abnormal state of the equipment in real time.

[0013] The multi-channel unit refers to a mechanical structure with parallel ventilation paths, for example, three or more independent air ducts can be configured, and each air duct is provided with a complete airflow control assembly. The independent ventilation channels are operated without interference by physical isolation, avoiding the influence of single-point failure on the overall system. The two-way electric sealing valve adopts an electric actuator to drive the valve plate, for example, a butterfly valve structure driven by a stepper motor can be selected to realize the airflow cutting function of the air inlet and outlet. The sensor array includes multiple types of detection elements such as pressure, particulate matter, temperature and humidity, for example, a differential pressure sensor is used to monitor the filter state, and a laser particle counter is used to detect air quality. The central controller includes a data processing unit and a control logic module, for example, an industrial-grade microprocessor is used to run a real-time operating system to perform channel health calculation and role allocation algorithms. The alarm module integrates an audible and visual alarm device, for example, a combination of a buzzer and an LED indicator light is configured to trigger an alarm signal when the fan is blocked, the valve is stuck or the sensor fails.

[0014] When the system is running, first, the cleanliness and resistance threshold parameters are set through the human-computer interface. The channel sensors continuously collect air duct pressure difference, air particle concentration and fan current data, and the central controller calculates the channel health index according to the real-time data. When the cleanliness of the main channel is lower than the set value or the filter resistance exceeds the standard, the controller automatically switches the standby channel to the main state. The electric sealing valve adjusts the opening according to the channel state instruction to ensure that the non-working channel is completely closed. When the fan overflows, the valve moves abnormally or the sensor data exceeds the limit, the controller immediately triggers an audible and visual alarm and records the fault code.

[0015] The differential pressure gauge refers to a device installed before and after the filter for measuring the airflow resistance, which can be realized by a differential pressure sensor, and the filter clogging degree is determined by calculating the pressure difference before and after the filter. The particulate matter sensor refers to a device for detecting the concentration of PM2.5 and PM10 in the air, which can be realized by a laser scattering sensor, and the cleanliness level is calculated by the concentration data to evaluate the air quality. The temperature and humidity sensor refers to a device for collecting the temperature and humidity of the supply air, which can be realized by a capacitive temperature and humidity sensor, and is used for real-time monitoring of the supply air environmental parameters. The Hall current sensor refers to a device for measuring the current signal by magnetic field induction, which can be realized by an open-loop Hall sensor, and the motor fault or load mutation is identified by monitoring the abnormal current waveform of the fan. The traditional system only relies on single differential pressure monitoring or manual inspection to determine the filter state, and cannot obtain multi-dimensional data such as cleanliness, temperature and humidity, and motor running state in real time. The present scheme integrates particulate matter concentration detection, temperature and humidity monitoring, and current anomaly identification functions, establishes a channel comprehensive health evaluation model, and solves the misjudgment or delayed maintenance problem caused by insufficient monitoring dimensions of the traditional system.

[0016] The central controller adopts a dual-core architecture to realize decoupled operation of control and calculation. The programmable logic controller continuously monitors the real-time data of the differential pressure gauge, particulate matter sensor and current detection module of each channel, and updates the channel health state database every 100 milliseconds. When it is detected that the differential pressure of the filter of a certain channel exceeds 80% of the critical value, a maintenance work order is automatically generated and pushed to the operation and maintenance system. The AI computing unit executes offline training tasks asynchronously in the background, calls the time-of-use electricity price API of the power company every morning to obtain the electricity price curve for the next 24 hours, and calculates the optimal operation combination based on the energy efficiency test data of each channel. The trained filter life prediction model performs batch reasoning every 6 hours, marks the channels with a predicted remaining life of less than 72 hours as high-risk status, and generates a channel switching recommendation scheme that avoids the peak electricity consumption period, which is transmitted to the programmable logic controller for strategy update through the OPC UA protocol.

[0017] Compared with the prior art, the traditional central air conditioning system relies on manual experience to set a fixed maintenance period, cannot dynamically adjust the operation and maintenance strategy according to the actual working conditions, and only considers the frequency regulation of the equipment in energy efficiency management without considering the electricity price fluctuation factor. The present scheme deploys a dual-core control architecture, introduces machine learning algorithms while ensuring real-time control accuracy, and realizes multi-objective optimization decision based on device health status and external environmental parameters. Further, the offline training and online reasoning are separated, which effectively avoids the resource occupation problem of complex algorithms on the real-time control system.

[0018] The health index refers to a comprehensive index reflecting the running state of the channel, which can be specifically realized by weighted calculation of differential pressure sensor data, particulate matter concentration, and fan current parameters, and is used for quantitative evaluation of the availability and reliability of the channel. The energy efficiency level refers to the energy efficiency level divided according to the ratio of fan power consumption to refrigerating capacity, which can be specifically realized by monitoring the electric energy consumption and air supply volume data in unit time, and is used for screening the channel with the lowest running cost.

[0019] The preset priority refers to the preset standby channel activation sequence, which can be specifically realized by dynamic generation based on the equipment commissioning time length and historical maintenance record scoring model, and is used for quickly determining the optimal standby channel in emergency switching. The historical failure rate refers to the number of times of running abnormalities of the equipment in the statistical period, which can be specifically realized by monthly statistical analysis of the fault log recorded by the central controller, and is used for evaluating the stability of the channel. The maintenance period refers to the interval length between two preventive maintenances, which can be specifically realized by setting according to the filter replacement record and differential pressure growth trend prediction model, and is used for dynamically adjusting the priority weight of the standby channel. The system load refers to the percentage of the air supply volume of the current main channel in the total designed air volume, which can be specifically realized by comparison and calculation of the air volume sensor data and rated parameters, and is used for judging whether to activate the sleep channel to share the load. The closed fan and electrically operated airtight valve refer to cutting off the power supply and air flow path of the sleep channel, which can be specifically realized by cutting off the motor power supply through the relay control circuit and sending the valve position closing instruction, and is used for reducing standby energy consumption.

[0020] In the channel allocation process, the health index and energy efficiency level of each channel are first calculated, and the channel with the highest comprehensive score is selected as the main channel to undertake the air supply task. The standby channel list is dynamically sorted according to historical maintenance data, for example, the channel that has recently completed filter replacement and has a failure rate lower than 5% is placed at the top. When the total system load is lower than the set threshold, the fan of the channel at the end of the ranking is automatically closed and the valve is sealed, so that the channel enters the zero-power standby state. For example, during the night low-load period, when it is monitored that the load rate of the main channel is continuously lower than 45% for 30 minutes, the central controller will close the electrically operated airtight valve of a sleep channel and stop the operation of its refrigerating unit. Compared with the prior art, the traditional multi-split system needs to be shut down for maintenance when a single channel fails, while the present scheme realizes seamless switching by dynamically selecting a standby channel through the health index. The standby equipment of the traditional system adopts a fixed sorting mechanism, while the present scheme dynamically adjusts the priority by combining the maintenance period and failure rate, so that the equipment utilization rate is improved. The traditional scheme only reduces the fan speed during low load, while the present scheme completely closes the sleep channel equipment, and the actual standby energy consumption can be reduced by more than 92%.

[0021] The central controller is configured to include multiple switching logics, namely, emergency switching when the real-time cleanliness N of the main channel is less than a set threshold N0, the resistance P is greater than a set threshold P0, or the fan current is abnormal, optimized switching when the health index of the standby channel is 10% higher than that of the main channel, and load balancing switching when the load of the main channel exceeds 80% to activate the dormant channel to share the air volume. The emergency switching refers to starting the standby channel immediately when the cleanliness or resistance of the main channel exceeds the safety range. Specifically, the differential pressure gauge and particulate matter sensor data are collected in real time, and when N < N0 or P > P0 is detected, the programmable logic controller triggers the linkage switching of the electric sealing valve and the fan to avoid system downtime due to failure of the main channel. The optimized switching refers to dynamically adjusting the priority of the main and standby channels based on the health index. Specifically, the AI calculation unit continuously compares the difference in health index between the main and standby channels, and when the health index of the standby channel exceeds the set percentage of the main channel, the channel switching is automatically performed to realize continuous optimization of system energy efficiency. The load balancing switching refers to dynamically distributing the running state of multiple channels according to the air volume demand. Specifically, the load rate of the main channel is monitored, and when it exceeds the set threshold, the dormant channel is awakened and the air valve opening is adjusted, so that multiple channels work together to blow air to reduce the operating pressure of a single channel.

[0022] The air supply mechanism inhales air through the air inlet, and when the driving member drives the rotating disc to rotate at high speed, the centrifugal force generated by the blades accelerates the airflow to blow towards the evaporator area of the refrigeration mechanism. The refrigerant inside the refrigeration mechanism absorbs heat to cool the air, and the processed cold air is delivered to the ventilation channel from the second air outlet. Multiple air supply mechanisms can operate independently or work together, and when a certain air supply mechanism fails, other mechanisms can still maintain the air supply function. The combination of the rotating disc and the blades makes the airflow distribution more uniform and reduces vortex generation.

[0023] The multi-stage speed regulation function of the permanent magnet synchronous motor refers to the motor that can adjust the speed level according to the control signal. Specifically, it can be realized by using a frequency converter in cooperation with a vector control algorithm, and the motor speed is changed by adjusting the input current frequency. The temperature sensor refers to a device installed on the surface of the evaporator for detecting temperature. Specifically, it can be realized by using a PT100 platinum resistance or a semiconductor thermosensitive element to monitor the real-time temperature change of the evaporator surface. The low-speed mode refers to the state in which the motor runs at a set value lower than the rated speed. Specifically, it can be realized by adjusting the output frequency of the frequency converter to 30%-50% of the rated frequency, thereby reducing the air flow rate on the surface of the evaporator. The permanent magnet synchronous motor receives the speed regulation instruction from the central controller through the frequency converter, and the temperature sensor installed on the surface of the evaporator continuously collects temperature data. When the evaporator surface temperature is detected to be lower than the preset threshold value (for example, 2°C), the central controller generates a speed reduction instruction, and the frequency converter reduces the motor speed to a preset low-speed interval. At this time, the air flow rate towards the evaporator is reduced, and the probability of frost formation on the low-temperature evaporator surface due to condensation of moisture in the air is reduced. When the temperature rises to the safe interval, the control system automatically restores the original speed level to ensure the refrigeration efficiency. Compared with the prior art, the fan of the traditional air conditioning system usually uses a fixed-speed motor or only has a simple speed regulation function with temperature control, and cannot dynamically adjust the speed according to the temperature of the evaporator surface. For example, some devices only control the temperature by intermittent start-stop, which causes the evaporator to frequently experience temperature fluctuations, thereby increasing the risk of frosting. The present scheme realizes precise speed linkage control by combining the multi-stage speed regulation characteristics of the permanent magnet synchronous motor with real-time temperature monitoring.

[0024] When the ventilation channel is in a non-air supply state, the air inlet side electric sealing valve is controlled to be in a completely closed state, at this time the ultraviolet sterilization lamp is automatically activated and starts to work. The ultraviolet radiation covers the inside of the channel including the filter surface, the inner wall of the air duct and the blade area, and kills the residual bacteria and viruses. The disinfection period can be set to 1-3 times a day according to the use frequency of the channel, and each time the irradiation lasts for 15-30 minutes. When the system needs to re-enable the channel, the electric sealing valve opening operation will synchronously cut off the power supply of the ultraviolet lamp, so as to ensure that there is no ultraviolet exposure risk when the personnel contact the air supply flow. Compared with the prior art, the traditional central air conditioning system lacks active disinfection measures during the channel closing period, and the long-term idle channel is prone to secondary pollution caused by microorganisms. The present scheme automatically executes the disinfection program during the idle period of the equipment through the intelligent linkage of the valve state and the ultraviolet lamp, which not only ensures the air supply hygiene quality, but also avoids the potential harm of ultraviolet to the operators.

[0025] A method for operating a multi-channel air conditioning system, comprising the following steps:

[0026] S1, system initialization: set the cleanliness threshold N0, terminal resistance threshold P0 and channel priority parameters through human-computer interaction terminal; the central controller starts the self-checking program to verify the communication state of each channel electric sealing valve, fan and sensor array;

[0027] S2, real-time monitoring and data acquisition: differential pressure gauge monitors the real-time resistance (P) of the filter, particulate matter sensor detects PM2.5 / PM10 concentration and calculates cleanliness level (N), temperature and humidity sensor collects air supply temperature and humidity, current detection module monitors the running state of the fan, and synchronously transmits the above data to the central controller for real-time analysis;

[0028] S3, dynamic allocation of channel role: the central controller comprehensively analyzes the sensor array data of each channel, the filter differential pressure is ΔP, the PM2.5 concentration is C1, and the fan current is I; the channel health index H is calculated as H = α·(1-ΔP / P_max)+β·(C0 / C1)+γ·(I_rated / I), wherein α+β+γ=1; according to the descending order of H value, the channel with the highest H value and energy efficiency level ≥E1 is set as the main channel, the next highest is set as the standby channel, and the rest enters the sleep state;

[0029] S4, main channel activation: after the H value sorting is completed, the air inlet side electric sealing valve of the main channel is opened, the fan is started to the preset speed, the refrigeration mechanism is started synchronously, and the evaporator temperature is controlled at 5±2℃;

[0030] S5, standby channel standby: the air inlet side electric sealing valve of the standby channel is kept in a 5% slightly open state, the fan enters the preheating mode, and the refrigeration mechanism maintains the lowest energy consumption standby.

[0031] ΔP is the real-time differential pressure, Pmax is the filter terminal resistance threshold (such as 200 Pa), and 1-Pmax / ΔP:

[0032] The smaller the differential pressure is, the closer the value is to 1, indicating that the filter state is better.

[0033] C0 is the cleanliness set threshold (such as the PM concentration corresponding to ISO Class 8), C1 is the particulate matter sensor detection value, the larger the ratio (the lower C1 is), the better the air quality is.

[0034] I_rated is the rated current, I is the real-time current, the closer the current is to the rated value (the ratio is close to 1), the more stable the fan runs; if the current is abnormal (such as I>1.2Irated), it indicates that the motor is overloaded or fails.

[0035] The S3 further includes the following steps

[0036] Trigger emergency switching: switch to standby channel within 100ms when N < N0, P > P0 or I > 1.2I_rated occurs in the main channel;

[0037] Trigger optimization switching: switch within 30s when H of standby channel > H of main channel + 10%;

[0038] Trigger load balancing switching: activate the dormant channel and send air in parallel when the system load of the main channel > 80%.

[0039] The second air outlet of the refrigeration mechanism is provided with a flow guide structure, which is a flow guide plate, the flow guide plate is bent into a first inclined section and a second inclined section, the first inclined section is connected with the second air outlet, and the second inclined section is connected with the first inclined section, and the inclination angle of the second inclined section with respect to the horizontal plane is smaller than the inclination angle of the first inclined section with respect to the horizontal plane;

[0040] The included angle A between the first inclined section and the horizontal plane ranges from 60° to 80°, and the included angle α between the second inclined section and the horizontal plane ranges from 20° to 30°.

[0041] A filter screen is detachably installed at the air inlet of the shell, a magnetic sealing strip is arranged at the edge of the filter screen, the filter screen is connected with the air inlet frame through magnetic adsorption, the filter screen adopts a gradient composite structure, including an outer coarse metal wire mesh layer, a middle layer of electret melt-blown cloth, and an inner layer of antibacterial activated carbon fiber, and the layers are connected in an airtight manner through hot-pressing corrugated joints.

[0042] The beneficial effects of the present application: the multi-channel air conditioning system of the present application realizes efficient and reliable air treatment through distributed independent channel design and intelligent dynamic control. The core principle is that: the sensor array is used to collect the differential pressure, particulate matter concentration, temperature and humidity and motor current data of each channel in real time, and the channel operation state is comprehensively evaluated through the health index model; the central controller based on the dual-core architecture of PLC and AI computing unit respectively executes real-time control and offline optimization tasks, dynamically allocates the roles of main, standby and dormant channels, and realizes fast fault response, continuous optimization of energy efficiency and load balancing through multi-level switching logic; combined with ultraviolet sterilization and evaporator speed linkage control, ensure the hygiene and refrigeration efficiency of the air supply, improve the system redundancy and fault tolerance capability, reduce the risk of downtime; through time-of-use electricity price and energy efficiency parameter optimization scheduling, reduce the operating energy consumption; realize predictive maintenance based on the health status of the equipment, prolong the service life of the equipment; ensure the cleanliness and temperature and humidity stability of the air supply, and significantly reduce the standby power consumption through the dormant channel complete closing technology, and improve the overall energy efficiency ratio. BRIEF DESCRIPTION OF DRAWINGS

[0043] Figure 1 It is a schematic diagram of the whole application;

[0044] Figure 2 Structure diagram of the fan of the present application;

[0045] Figure 3 Structure diagram of the fan mechanism of the present application;

[0046] Figure 4 Structure diagram of the flow guide structure of the present application;

[0047] Figure 5 Exploded view of the filter screen of the present application;

[0048] Figure 6 Process flow diagram of the present application.

[0049] Reference signs include:

[0050] 1, multi-channel unit; 2, central controller; 3, human-computer interaction terminal; 4, alarm module; 5, fan; 6, air inlet side electric sealing valve; 7, air outlet side electric sealing valve; 8, differential pressure gauge; 9, particulate matter sensor; 11, temperature and humidity sensor; 12, current detection module; 13, programmable logic controller (PLC); 14, AI computing unit; 15, rack; 16, air supply mechanism; 17, refrigeration mechanism; 18, first air outlet; 19, second air outlet; 21, shell; 22, rotating disc; 23, blade; 24, driving piece; 25, air inlet; 26, evaporator; 27, temperature sensor; 28, ultraviolet sterilization lamp; 29, flow guide structure; 30, first inclined section; 31, second inclined section; 32, filter screen; 33, coarse-effect metal screen layer; 34, electret melt-blown cloth layer; 35, antibacterial activated carbon fiber layer. DETAILED DESCRIPTION

[0051] In order to facilitate the understanding of those skilled in the art, the present application will be further described below in conjunction with the embodiments and drawings. The content mentioned in the embodiments is not a limitation of the present application.

[0052] Please refer to Figures 1 to 4As shown, a multi-channel air conditioning system of the present application comprises a central controller 2, a human-computer interaction terminal 3, an alarm module 4 and a multi-channel unit 1; the multi-channel unit 1 comprises at least three independent ventilation channels, each channel is provided with a fan 5, an inlet-side electrically operated airtight valve 6, an outlet-side electrically operated airtight valve 7 and a sensor array; the central controller 2 is electrically connected with the fan 5, the inlet-side electrically operated airtight valve 6, the outlet-side electrically operated airtight valve 7 and the sensor array of each channel, the central controller 2 dynamically allocates the channels as a main channel, a standby channel or a dormant channel according to the sensor array; when the real-time cleanliness of the main channel is lower than a set threshold or the filter resistance exceeds a final resistance threshold, the channel switching is triggered; the human-computer interaction terminal 3 is connected with the central controller 2, and is used for receiving the input of the cleanliness threshold, the final resistance threshold and the channel priority; the alarm module 4 is electrically connected with the central controller 2, when the fan 5 of each channel fails, the electrically operated airtight valve is not closed and the sensor array data is abnormal, the central controller 2 triggers the alarm module 4, and the alarm module 4 pushes the alarm information to the user through a sound and light alarm.

[0053] The sensor array of each channel comprises a differential pressure gauge 8 arranged at the front and rear ends of the channel filter, which is used for monitoring the real-time resistance value of the channel, a particulate matter sensor 9 arranged in the channel, which is used for detecting the PM2.5 and PM10 concentrations to calculate the cleanliness level, a temperature and humidity sensor 11 arranged in the channel, which is used for collecting the air supply temperature and humidity data, and a fan 5 current detection module 12, which is a Hall current sensor arranged in the fan 5 power supply line or control cabinet, and is used for collecting the current signal in real time to identify the abnormal operation of the fan 5.

[0054] The central controller 2 comprises a programmable logic controller PLC 13 and an AI computing unit 14, the programmable logic controller PLC 13 is used for collecting the sensor data of each channel and the running state of the fan 5 in real time, dynamically allocating the channel role as a main channel, a standby channel or a dormant channel, predicting the equipment failure based on the sensor data and generating a maintenance instruction; the AI computing unit 14 runs asynchronously independently of the PLC, trains a filter clogging prediction model through historical resistance value data, the filter clogging prediction model generates an energy consumption optimal channel scheduling scheme based on the time-of-use electricity price data and the channel energy efficiency parameters, and periodically pushes the optimization strategy to the programmable logic controller PLC 13.

[0055] The dynamic allocation logic of the channel role is that the selection condition of the main channel is the highest health index and the optimal energy efficiency level, the standby channel is sorted according to the preset priority, the priority sorting is dynamically adjusted according to the historical failure rate and the maintenance period, and the fan 5 and the electrically operated airtight valve of the dormant channel are closed to save energy when the system load is lower than 50%.

[0056] The central controller 2 is configured to include multiple switching logics, namely, emergency switching when the real-time cleanliness N of the main channel is less than a set threshold N0, the resistance P is greater than a set threshold P0, or the current of the fan 5 is abnormal, optimized switching when the health index of the standby channel is 10% higher than that of the main channel, and load balancing switching when the load of the main channel exceeds 80% to activate the dormant channel to share the air volume.

[0057] The fan 5 includes a rack 15, multiple air supply mechanisms 5 arranged on the rack 15, and a refrigeration mechanism 17. The first air outlet 18 of the air supply mechanism 5 is connected with the refrigeration mechanism 17. The refrigeration mechanism 17 is provided with a second air outlet 19 at an end away from the air supply mechanism 5. The air supply mechanism 5 includes a shell 21, a rotating disc 22 arranged in the shell 21, multiple groups of blades 23 arranged on the rotating disc 22, and a driving member 24 for driving the rotating disc 22 to rotate. The shell 21 is provided with an air inlet 25 at a side edge. The driving member 24 drives the rotating disc 22 to rotate to drive the blades 23 to rotate and blow air flow into the refrigeration mechanism 17 to form cold air. The cold air is blown out through the second air outlet 19 to supply air for the ventilation channel.

[0058] The driving member 24 is a permanent magnet synchronous motor with multiple speed regulation functions. The evaporator 26 of the refrigeration mechanism 17 is provided with a temperature sensor 27. The rotating speed control signal of the driving member 24 is linked with the temperature sensor 27 of the evaporator 26 of the refrigeration mechanism 17. When the temperature sensor 27 detects that the surface temperature of the evaporator 26 is less than a preset value, the driving member 24 automatically switches to a low speed mode to reduce the frosting risk of the evaporator 26.

[0059] The multiple-channel unit 1 is provided with an ultraviolet sterilization lamp 28 in each ventilation channel. The ultraviolet sterilization lamp 28 is linked with the air inlet side electric sealing valve 6. When the valve is closed, the ultraviolet sterilization lamp 28 is automatically turned on to periodically disinfect the inside of the channel.

[0060] The air conditioning system includes a channel unit, a central controller 2, a human-computer interaction terminal 3, and an alarm module 4. The multiple-channel unit 1 is provided with at least three independent ventilation channels. Each channel is provided with a fan 5, a two-way electric sealing valve, and a sensor array. The central controller 2 is connected with each actuator and sensor to perform channel role dynamic allocation and switching control. The human-computer interaction terminal 3 provides a parameter setting interface. The alarm module 4 provides real-time feedback of abnormal states of the equipment.

[0061] The multi-channel unit 1 refers to a mechanical structure with parallel ventilation paths, for example, three or more independent air ducts can be configured, each air duct is equipped with a complete air flow control assembly. Independent ventilation channels are physically isolated to operate without interference, avoiding single point failure affecting the overall system. The two-way electric sealing valve uses an electric actuator to drive the valve plate, for example, a stepper motor driven butterfly valve structure can be selected to achieve air flow cutoff function between the air inlet 25 and the air outlet. The sensor array includes multiple types of detection elements such as pressure, particulate matter, temperature and humidity, for example, a differential pressure sensor is used to monitor the filter state, and a laser particle counter is used to detect air quality. The central controller 2 includes a data processing unit and a control logic module, for example, an industrial microprocessor is used to run a real-time operating system to execute channel health calculation and role assignment algorithms. The alarm module 4 integrates audible and visual alarm devices, for example, a combination of a buzzer and an LED indicator is configured to trigger an alarm signal when the fan 5 is detected to be locked, the valve is stuck, or the sensor fails.

[0062] When the system is running, first set the cleanliness and resistance threshold parameters through the human-machine interface. The channel sensors continuously collect air duct pressure difference, air particle concentration and fan 5 current data, and the central controller 2 calculates the channel health index according to the real-time data. When the cleanliness of the main channel is lower than the set value or the filter resistance exceeds the standard, the controller automatically switches the standby channel to the main state. The electric sealing valve adjusts the opening according to the channel state instruction to ensure that the non-working channel is completely closed. When the fan 5 overcurrent, valve action abnormal or sensor data is out of limit is detected, the controller immediately triggers an audible and visual alarm and records the fault code.

[0063] The differential pressure meter 8 refers to a device installed before and after the filter to measure the air flow resistance, which can be implemented by a differential pressure sensor. The degree of filter clogging is determined by calculating the pressure difference between the front and rear ends. The particulate matter sensor 9 refers to a device for detecting the concentration of PM2.5 and PM10 in the air, which can be implemented by a laser scattering sensor. The cleanliness level is calculated based on the concentration data to evaluate the air quality. The temperature and humidity sensor 11 refers to a device for collecting supply air temperature and humidity, which can be implemented by a capacitive temperature and humidity sensor 11 for real-time monitoring of supply air environmental parameters. The Hall current sensor refers to a device that measures current signals through magnetic field induction, which can be implemented by an open-loop Hall sensor to identify motor failure or load sudden change by monitoring the current waveform of the fan 5. The traditional system only relies on single differential pressure monitoring or manual inspection to determine the filter state, and cannot obtain multi-dimensional data such as cleanliness, temperature and humidity, and motor operating state in real time. This scheme integrates particulate matter concentration detection, temperature and humidity monitoring, and current anomaly identification functions to establish a comprehensive channel health evaluation model, solving the misjudgment or delayed maintenance problem caused by insufficient monitoring dimensions in traditional systems.

[0064] The central controller 2 adopts a dual-core architecture to realize the decoupling operation of control and calculation. The programmable logic controller continuously monitors the real-time data of the channel differential pressure meter 8, the particulate matter sensor 9 and the current detection module 12, and updates the channel health state database every 100 milliseconds. When it is detected that the differential pressure of a certain channel filter exceeds 80% of the critical value, a maintenance work order is automatically generated and pushed to the operation and maintenance system. The AI computing unit 14 executes offline training tasks asynchronously in the background. Every morning, it calls the time-of-use electricity price API of the power company to obtain the electricity price curve for the next 24 hours, and calculates the optimal operation combination based on the energy efficiency test data of each channel. The trained filter life prediction model performs batch reasoning every 6 hours, marks channels with a predicted remaining life of less than 72 hours as high-risk status, and generates channel switching suggestions to avoid peak electricity consumption periods. The suggestions are transmitted to the programmable logic controller for strategy update through the OPC UA protocol.

[0065] Compared with the prior art, the traditional central air conditioning system relies on manual experience to set a fixed maintenance period and cannot dynamically adjust the operation and maintenance strategy according to the actual working conditions. In terms of energy efficiency management, it only considers frequency regulation of equipment and ignores the price fluctuation factor. The present scheme introduces machine learning algorithms by deploying a dual-core control architecture to ensure real-time control accuracy and achieve multi-objective optimization decisions based on device health status and external environmental parameters. Further, the offline training and online reasoning are separated to effectively avoid resource occupation of complex algorithms on real-time control systems.

[0066] The health index refers to a comprehensive indicator reflecting the running state of the channel, which can be calculated by weighting the differential pressure sensor data, particulate matter concentration and fan 5 current parameters, and is used to quantitatively evaluate the availability and reliability of the channel. The energy efficiency level refers to the energy efficiency level divided according to the ratio of fan 5 power consumption to refrigerating capacity, which can be calculated by monitoring the power consumption and air supply data per unit time, and is used to select the channel with the lowest operating cost.

[0067] The preset priority refers to a preset standby channel activation sequence, which can be dynamically generated based on a device commissioning duration, a historical maintenance record scoring model, and is used to quickly determine the optimal standby channel during emergency switching. The historical failure rate refers to the number of times of abnormal operation of the device in a statistical period, which can be achieved by monthly statistical analysis of the fault log recorded by the central controller 2, and is used to evaluate the stability of the channel. The maintenance period refers to the interval between two preventive maintenance, which can be achieved according to the filter replacement record and the pressure difference growth trend prediction model, and is used to dynamically adjust the priority weight of the standby channel. The system load refers to the percentage of the supply air volume of the current main channel in the total design air volume, which can be calculated by comparing the air volume sensor data with the rated parameters, and is used to determine whether to enable the dormant channel to share the load. The closed fan 5 and electric sealing valve refer to cutting off the power supply and air flow path of the dormant channel, which can be achieved by cutting off the motor power supply through the relay control circuit and sending the valve position closing instruction, and is used to reduce standby energy consumption.

[0068] In the channel allocation process, the health index and energy efficiency level of each channel are first calculated, and the channel with the highest comprehensive score is selected as the main channel to undertake the air supply task. The standby channel list is dynamically sorted according to historical maintenance data, for example, the channel that has recently replaced the filter and has a failure rate of less than 5% is placed at the top. When the total system load is lower than the set threshold, the fan 5 of the channel at the end of the ranking is automatically closed and the valve is sealed, so that the channel enters a zero-power standby state. For example, during the night low load period, when it is monitored that the load rate of the main channel is continuously lower than 45% for 30 minutes, the central controller 2 will close the electric sealing valve of a dormant channel and stop the operation of its refrigerating unit. Compared with the prior art, the traditional multi-split system needs to be shut down for maintenance when a single channel fails, while the present scheme realizes seamless switching by dynamically selecting the standby channel through the health index. The standby device of the traditional system uses a fixed sorting mechanism, while the present scheme dynamically adjusts the priority by combining the maintenance period and the failure rate, so that the device utilization rate is improved. The traditional scheme only reduces the fan 5 speed during low load, while the present scheme completely closes the dormant channel device, and the actual standby energy consumption can be reduced by more than 92%.

[0069] The central controller 2 is configured to include multiple switching logics, namely, emergency switching when the real-time cleanliness N of the main channel is less than the set threshold N0, the resistance P is greater than the set threshold P0, or the fan 5 current is abnormal, optimized switching when the standby channel health index is 10% higher than the main channel, and load balancing switching when the load of the main channel exceeds 80% to activate the dormant channel to share the air volume. Emergency switching refers to immediately starting the standby channel when the cleanliness or resistance of the main channel exceeds the safety range. Specifically, the differential pressure gauge 8 and the particulate matter sensor 9 data can be collected in real time, and when N < N0 or P > P0 is detected, the programmable logic controller triggers the linkage switching of the electric sealing valve and the fan 5 to avoid system downtime due to main channel failure. Among them, optimized switching refers to dynamically adjusting the priority of the main and standby channels based on the health index. Specifically, the AI computing unit 14 can continuously compare the difference in health index between the main and standby channels, and when the standby channel health index exceeds the set percentage of the main channel, automatically execute channel switching to achieve continuous optimization of system energy efficiency. Among them, load balancing switching refers to dynamically distributing the running state of multiple channels according to the air volume demand. Specifically, when the load rate of the main channel exceeds the set threshold, the dormant channel is awakened and the air valve opening is adjusted, so that multiple channels work together to blow air to reduce the operating pressure of a single channel.

[0070] The air supply mechanism 5 inhales air through the air inlet 25, and when the driving member 24 drives the rotating disc 22 to rotate at high speed, the blades 23 generate centrifugal force to accelerate the airflow to the evaporator 26 area of the refrigeration mechanism 17. The refrigeration mechanism 17 inside the refrigerant absorbs heat to cool the air, and the processed cold air is delivered to the ventilation channel from the second air outlet 19. Multiple air supply mechanisms 5 can operate independently or work together, and when a certain air supply mechanism 5 fails, other mechanisms can still maintain the air supply function. The combination design of the rotating disc 22 and the blades 23 makes the airflow distribution more uniform and reduces vortex generation.

[0071] The multi-stage speed regulation function of the permanent magnet synchronous motor refers to the motor that can adjust the speed level according to the control signal. Specifically, it can be realized by using a frequency converter in cooperation with a vector control algorithm, and the motor speed is changed by adjusting the input current frequency. The temperature sensor 27 refers to a device installed on the surface of the evaporator 26 for detecting temperature. Specifically, it can be realized by using a PT100 platinum resistance or a semiconductor thermosensitive element to monitor the real-time temperature change of the evaporator 26 surface. The low-speed mode refers to the state in which the motor runs at a set value lower than the rated speed. Specifically, it can be realized by adjusting the output frequency of the frequency converter to 30%-50% of the rated frequency, thereby reducing the air flow speed on the surface of the evaporator 26. The permanent magnet synchronous motor receives the speed regulation instruction from the central controller 2 through the frequency converter, and the temperature sensor 27 installed on the surface of the evaporator 26 continuously collects temperature data. When it is detected that the temperature on the surface of the evaporator 26 is lower than the preset threshold value (for example, 2°C), the central controller 2 generates a speed reduction instruction, and the frequency converter reduces the motor speed to the preset low-speed interval. At this time, the air flow speed blowing towards the evaporator 26 decreases, reducing the probability of frost formation on the low-temperature evaporator 26 surface due to condensation of moisture in the air. When the temperature rises to the safe interval, the control system automatically restores the original speed level to ensure the refrigeration efficiency. Compared with the prior art, the fan 5 of the traditional air conditioning system usually uses a fixed-speed motor or only has a simple speed regulation function with temperature control, and cannot dynamically adjust the speed according to the temperature on the surface of the evaporator 26. For example, some devices only control the temperature by intermittent start-stop, which causes the evaporator 26 to frequently experience temperature fluctuations, thereby increasing the risk of frosting. The present scheme realizes precise speed linkage control by combining real-time temperature monitoring with the multi-stage speed regulation characteristics of the permanent magnet synchronous motor.

[0072] When the ventilation channel is in a non-air supply state, the air inlet side electric sealing valve 6 is controlled to be completely closed, at this time the ultraviolet sterilization lamp 28 is automatically activated and starts to work. The ultraviolet radiation covers the inside of the channel including the filter surface, the air duct inner wall and the blade 23 area, killing the residual bacteria and viruses. The disinfection period can be set to 1-3 times a day according to the channel use frequency, and each time the irradiation lasts for 15-30 minutes. When the system needs to re-enable the channel, the electric sealing valve opening operation will synchronously cut off the power supply of the ultraviolet lamp, to ensure that there is no ultraviolet exposure risk when the personnel contact the air supply flow. Compared with the prior art, the traditional central air conditioning system lacks active disinfection measures during the channel closing period, and the long-term idle channel is prone to secondary pollution caused by microorganisms. The present scheme automatically executes the disinfection program during the idle period of the equipment through the intelligent linkage of the valve state and the ultraviolet lamp, which not only ensures the air supply hygiene quality, but also avoids the potential harm of ultraviolet radiation to the operators.

[0073] A method for operating a multi-channel air conditioning system, comprising the following steps:

[0074] S1, system initialization: set the cleanliness threshold N0, terminal resistance threshold P0 and channel priority parameters through the man-machine interaction terminal 3; the central controller 2 starts the self-checking program to verify the communication state of the electric sealing valve, fan 5 and sensor array of each channel;

[0075] S2, real-time monitoring and data acquisition: the differential pressure gauge 8 monitors the real-time resistance (P) of the filter, the particulate matter sensor 9 detects the PM2.5 / PM10 concentration and calculates the cleanliness level (N), the temperature and humidity sensor 11 collects the air supply temperature and humidity, and the current detection module 12 monitors the running state of the fan 5, and synchronously transmits the above data to the central controller 2 for real-time analysis;

[0076] S3, dynamic allocation of channel role: the central controller 2 comprehensively analyzes the sensor array data of each channel, the filter differential pressure is ΔP, the PM2.5 concentration is C1, and the fan 5 current is I; the channel health index H = α·(1-ΔP / P_max) + β·(C0 / C1) + γ·(I_rated / I) is calculated, wherein α+β+γ=1; according to the descending order of H value, the channel with the highest H value and energy efficiency level ≥E1 is set as the main channel, the next highest one is set as the standby channel, and the rest enters the sleep state;

[0077] S4, main channel activation: after the H value sorting is completed, the air inlet side electric sealing valve 6 of the main channel is opened, the fan 5 is started to the preset speed, the refrigeration mechanism 17 is synchronously started, and the evaporator 26 temperature is controlled at 5±2℃;

[0078] S5, standby channel standby: the air inlet side electric sealing valve 6 of the standby channel keeps a 5% micro-opening state, the fan 5 enters the preheating mode, and the refrigeration mechanism 17 maintains the lowest energy consumption standby.

[0079] ΔP is the real-time differential pressure, Pmax is the terminal resistance threshold of the filter (such as 200 Pa), and 1-Pmax / ΔP:

[0080] The smaller the differential pressure is, the closer the value is to 1, indicating that the filter state is better.

[0081] C0 is the cleanliness set threshold (such as the PM concentration corresponding to ISOClass 8), C1 is the detection value of the particulate matter sensor 9, the larger the ratio (the lower C1 is), the better the air quality is.

[0082] I_rated is the rated current, I is the real-time current, the closer the current is to the rated value (the ratio is close to 1), the more stable the fan 5 runs; if the current is abnormal (such as I>1.2Irated), it indicates that the motor is overloaded or fails.

[0083] The S3 further includes the following steps

[0084] Trigger emergency switching: when the main channel N < N0, P > P0 or I > 1.2I_rated, switch to the standby channel within 100ms;

[0085] Trigger optimization switching: when the standby channel H > main channel H + 10%, switch within 30s;

[0086] Trigger load balancing switching: when the main channel system load > 80%, activate the dormant channel and send air in parallel.

[0087] The second air outlet 19 of the refrigeration mechanism 17 is provided with a flow guide structure 29, which is a flow guide plate. The flow guide plate is bent into a first inclined section 30 and a second inclined section 31. The first inclined section 30 is connected with the second air outlet 19, and the second inclined section 31 is connected with the first inclined section 30. The inclination angle of the second inclined section 31 with respect to the horizontal plane is smaller than the inclination angle of the first inclined section with respect to the horizontal plane.

[0088] The included angle A between the first inclined section 30 and the horizontal plane ranges from 60° to 80°, and the included angle α between the second inclined section 31 and the horizontal plane ranges from 20° to 30°.

[0089] A filter screen 32 is detachably installed at the air inlet 25 of the shell 21. The filter screen 32 is provided with a magnetic sealing strip at the edge. The filter screen 32 is connected with the frame of the air inlet 25 through the magnetic sealing strip. The filter screen 32 adopts a gradient composite structure, including an outer coarse metal screen layer 33, a middle electret melt-blown cloth layer 34, and an inner antibacterial activated carbon fiber layer 35. The layers are connected in airtight manner through hot-pressed corrugated joints.

[0090] The above is only the preferred embodiment of the present application. For those skilled in the art, the specific implementation and application range can be changed according to the idea of the present application. The content of the specification should not be understood as a limitation of the present application.

Claims

1. A multi-pass air conditioning system, characterized by: The application relates to a multi-channel air supply system, which comprises a central controller (2), a man-machine interaction terminal (3), an alarm module (4) and a multi-channel air supply unit (1); the multi-channel air supply unit (1) comprises at least three independent ventilation channels, each of which is provided with a fan (5), an air inlet side electric sealing valve (6), an air outlet side electric sealing valve (7) and a sensor array; the central controller (2) is electrically connected with the fan (5), the air inlet side electric sealing valve (6), the air outlet side electric sealing valve (7) and the sensor array of each channel, and the central controller (2) dynamically allocates a main channel, a standby channel and a dormant channel according to the data of the sensor array; When the real-time cleanliness of the main channel is lower than a set threshold value or the filter resistance exceeds a final resistance threshold value, channel switching is triggered; The man-machine interaction terminal (3) is electrically connected with the central controller (2) and is used for receiving the input of the cleanliness threshold value, the final resistance threshold value and the channel priority; the alarm module (4) is connected with the central controller (2), and when the fan (5) of each channel is faulty, the electric sealing valve is not closed and the sensor array data is abnormal, the central controller (2) triggers the alarm module (4), and the alarm module (4) pushes alarm information to the user through a sound and light alarm; The central controller (2) comprises a programmable logic controller (PLC) (13) and an AI calculation unit (14), the programmable logic controller (PLC) (13) is used for collecting the sensor data and the running state of the fan (5) of each channel in real time, dynamically allocating the channel roles as the main channel, the standby channel or the dormant channel according to the collected data, predicting the equipment fault based on the sensor data and generating a maintenance instruction; the AI calculation unit (14) is independently operated asynchronously with the programmable logic controller (PLC) (13), a filter clogging prediction model is trained through historical resistance value data, the filter clogging prediction model generates an energy consumption optimal channel scheduling scheme by predicting the remaining life of the filter of each channel, based on the time-of-use electricity price data and the channel energy efficiency parameters, and then periodically pushes the optimization strategy to the programmable logic controller (PLC) (13); The dynamic allocation logic of the channel roles is that the selection condition of the main channel is the highest health index and the optimal energy efficiency level, the standby channel is sorted according to a preset priority, the priority is dynamically adjusted according to the historical failure rate and the maintenance period, and the dormant channel is closed to save energy when the system load of the main channel is lower than 50%. The central controller (2) comprehensively analyzes the sensor array data of each channel, calculates the health index according to the filter differential pressure, the PM2.5 concentration and the fan (5) current.

2. A multi-pass air conditioning system as set forth in claim 1, characterized in that: The sensor array of each ventilation channel includes a differential pressure gauge (8) arranged at the front and back ends of the channel filter for monitoring the real-time resistance value of the channel, a particulate matter sensor (9) arranged inside the channel for detecting PM2.5 and PM10 concentrations to calculate the cleanliness level, a temperature and humidity sensor (11) arranged inside the channel for collecting air supply temperature and humidity data, and a current detection module (12), which is a Hall current sensor arranged in the power supply circuit or control cabinet of the fan (5) to collect real-time current signals for identifying abnormal operation of the fan (5).

3. The multi-pass air conditioning system of claim 1, wherein: The central controller (2) is configured to include multiple switching logics, namely, emergency switching when the real-time cleanliness N of the main channel is less than the set threshold N0, the resistance P is greater than the set threshold P0, or the current of the fan (5) is abnormal, optimized switching when the health index of the standby channel is 10% higher than that of the main channel, and load balancing switching when the load of the main channel exceeds 80% to activate the dormant channel to share the air volume.

4. The multi-pass air conditioning system of claim 1, wherein: The fan (5) includes a rack (15), multiple air supply mechanisms arranged on the rack (15), and a refrigeration mechanism (17), the first air outlet (18) of the air supply mechanism is connected with the refrigeration mechanism (17), the refrigeration mechanism (17) is provided with a second air outlet (19) away from the air supply mechanism, the air supply mechanism includes a shell (21), a rotating disc (22) arranged in the shell (21), multiple groups of blades (23) arranged on the rotating disc (22), and a driving member (24) for driving the rotating disc (22) to rotate, the shell (21) is provided with an air inlet (25) on the side, and the driving member (24) drives the rotating disc (22) to rotate to drive the blades (23) to rotate and blow the air flow into the refrigeration mechanism (17) to form cold air, and the cold air is blown out through the second air outlet (19) to supply air and refrigeration for the ventilation channel.

5. A multi-pass air conditioning system as set forth in claim 4, characterized in that: The driving member (24) is a permanent magnet synchronous motor with multiple speed regulation functions, the evaporator (26) of the refrigeration mechanism (17) is provided with a temperature sensor (27), and the speed control signal of the driving member (24) is linked with the temperature sensor (27) of the evaporator (26) of the refrigeration mechanism (17); when the temperature sensor (27) detects that the surface temperature of the evaporator (26) is less than a preset value, the driving member (24) automatically switches to a low-speed mode to reduce the frosting risk of the evaporator (26).

6. The multi-pass air conditioning system of claim 1, wherein: An ultraviolet sterilization lamp (28) is arranged in each ventilation channel of the multi-channel unit (1), and the ultraviolet sterilization lamp (28) is linked with the air inlet side electric sealing valve (6) and automatically opens when the valve is closed to periodically disinfect the inside of the channel.

7. The operation control method of a multi-pass air conditioning system according to any one of claims 1 to 6, characterized by, The method comprises the following steps: S1, system initialization: set the cleanliness threshold N0, final resistance threshold P0, and priority parameters of each channel through the man-machine interaction terminal (3); the central controller (2) starts the self-checking program to verify the communication states of the electric sealing valve, fan (5), and sensor array of each channel; S2, real-time monitoring and data collection: differential pressure gauge (8) monitors real-time resistance (P) of filter, particulate matter sensor (9) detects PM2.5, PM10 concentration and calculates cleanliness level, temperature and humidity sensor (11) collects air supply temperature and humidity, current detection module monitors fan (5) running state, and synchronously transmits above data to central controller (2) for real-time analysis; S3, dynamic allocation of channel role: central controller (2) comprehensively analyzes sensor array data of each channel, filter differential pressure is ΔP, PM2.5 concentration is C1, fan (5) current is I; P_max is filter final resistance threshold, C0 is cleanliness setting threshold, I_rated is rated current; H=α·(1-ΔP / P_max)+β·(C0 / C1)+γ·(I_rated / I), wherein α+β+γ=1; according to descending order of H value, channel with highest H value and energy efficiency level≥E1 is set as main channel, the second highest is set as standby channel, and the rest enter sleep state; S4, main channel activation: open air inlet side electric sealing valve (6) of main channel, start fan (5) to pre-set rotating speed, simultaneously start refrigeration mechanism (17), and control temperature of evaporator (26) at 5±2℃; S5, standby channel standby: air inlet side electric sealing valve (6) of standby channel keeps 5% micro-opening state, fan (5) enters preheating mode, and refrigeration mechanism (17) maintains minimum energy consumption standby.

8. The method of operating a multi-pass air conditioning system of claim 7 wherein: The S3 further comprises the following steps (a), trigger emergency switching: when the main channel appears N (b), trigger optimization switching: when standby channel H> main channel H+10%, switch within 30s; (c), trigger load balancing switching: when main channel system load>80%, activate sleep channel and run air supply.

Images