Building smoke exhaust intelligent adjusting system and method for multi-air-port cooperative control

Abstract

The invention discloses a building smoke exhaust intelligent adjusting system and method for multi-air-port cooperative control, and the system comprises a data collection module which is used for collecting the smoke exhaust demands corresponding to all air ports and the pressure data of a public air duct in real time, and an AI cloud intelligent control module which is used for dynamically generating a fan frequency adjustment instruction and a valve opening adjustment instruction; the layered control module is used for controlling air outlet in a layered manner according to the independent air ports and the public air duct; the energy-saving optimization module is used for enabling the fan to operate in the optimal efficiency interval through a frequency optimization algorithm and dynamically distributing the smoke exhaust amount of each air outlet based on the smoke exhaust priority; and the fault early warning module is used for monitoring the current and voltage of the fan and the pressure data of the air duct in real time, identifying the abnormal state of the equipment and triggering multi-stage warning and protection actions. According to the invention, the fan frequency adjusting instruction can be dynamically generated to realize accurate regulation and control; air-out is controlled in a layered mode through the independent air openings and the public air channel, the air channel pressure is more stable, the air-out effect of all areas is ensured, and the energy utilization rate is increased.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent smoke exhaust regulation, and in particular to an intelligent smoke exhaust regulation system and method for coordinated control of multiple air outlets in a building. Background Art

[0002] In today's construction industry, smoke exhaust systems are crucial for maintaining indoor air quality and ensuring the comfort and health of personnel. However, existing building smoke exhaust systems have exposed many problems in actual operation. Traditional systems often use extensive control, which makes it difficult to accurately consider the actual smoke exhaust needs of each air outlet. For example, in places such as restaurants, the number of burners used at different times varies greatly, and the demand for smoke exhaust volume varies significantly. However, traditional systems are unable to adjust in a timely and accurate manner, resulting in poor smoke exhaust or energy waste. In terms of public duct pressure control, existing technologies also have shortcomings. When multiple air outlets are working at the same time, the duct pressure is prone to fluctuations, affecting the overall smoke exhaust effect, and it is difficult to take into account the needs of different areas.

[0003] At the same time, the existing system is not very effective in terms of energy-saving optimization. The fan operating efficiency is low and cannot be dynamically adjusted to the optimal efficiency range according to actual working conditions, resulting in a large amount of electricity waste. In addition, the distribution of smoke exhaust volume at each air outlet lacks flexibility, making it difficult to dynamically adjust according to the smoke exhaust priority, resulting in poor smoke exhaust effects in some areas and excessive air volume in other areas. In addition, the existing smoke exhaust system has weak monitoring and early warning capabilities for equipment failures. The fan current, voltage, and duct pressure data cannot be monitored in real time and comprehensively. The lack of anomaly detection models makes it impossible to detect potential equipment anomalies in a timely manner. Failures are often discovered only after they occur and affect normal use. The maintenance cost is high and affects the normal operation of the building. Summary of the Invention

[0004] In order to solve at least one of the technical problems raised above, the present invention provides a building smoke exhaust intelligent adjustment system and method for coordinated control of multiple air outlets.

[0005] In a first aspect, the present invention provides an intelligent smoke exhaust control system for a building for coordinated control of multiple air outlets, the system comprising:

[0006] The data acquisition module is used to collect the exhaust demand data corresponding to each air outlet and the public air duct pressure data in real time. The exhaust demand data includes the number of burners used, valve opening and fan operating parameters corresponding to each air outlet;

[0007] The AI ​​cloud intelligent control module is deployed in the cloud and includes a machine learning model and a PID control algorithm, which is used to dynamically generate fan frequency adjustment instructions and valve opening adjustment instructions based on the smoke exhaust demand data and the public duct pressure data;

[0008] Hierarchical control module, used to control air outlet in different layers according to independent air outlets and public air ducts;

[0009] Energy-saving optimization module, which uses frequency optimization algorithm to make the fan operate in the optimal efficiency range and dynamically allocates the exhaust volume of each air outlet based on the exhaust priority;

[0010] The fault warning module is used to monitor fan vibration, temperature, current, voltage and duct pressure data in real time, identify abnormal equipment status through anomaly detection models, and trigger multi-level alarms and protection actions.

[0011] In a second aspect, the present invention further provides a building smoke exhaust intelligent adjustment method for coordinated control of multiple air outlets, which is applied to the building smoke exhaust intelligent adjustment system for coordinated control of multiple air outlets as described in any one of the first aspects, comprising:

[0012] Real-time collection of smoke exhaust demand data of each air outlet and public air duct pressure data;

[0013] Calculate the required air volume of each air outlet and the target pressure of the public air duct based on the exhaust demand data of each air outlet and the pressure data of the public air duct, and generate the fan frequency and valve opening adjustment instructions;

[0014] The system coordinates the adjustment of independent air outlets and public air ducts according to the adjustment instructions to ensure the balance between local smoke exhaust demand and global pressure. The frequency optimization algorithm is used to make the fan operate in the optimal efficiency range, and the smoke exhaust volume of each air outlet is dynamically allocated based on the smoke exhaust priority.

[0015] Monitor fan current, voltage, and duct pressure data in real time, identify equipment abnormalities through anomaly detection models, and trigger multi-level alarms and protection actions.

[0016] In a third aspect, the present invention also provides a computer-readable storage medium, in which a computer program is stored. The computer program includes program instructions, and when the program instructions are executed by a processor of an electronic device, the processor executes the method as described in the first aspect above and any possible implementation thereof.

[0017] Compared with the prior art, the present invention has the following beneficial effects:

[0018] The intelligent smoke exhaust adjustment system provided by the present invention obtains the smoke exhaust demand data of each air outlet and the public air duct pressure data in real time through the data acquisition module, providing a detailed basis for subsequent precise control. The AI ​​cloud intelligent control module uses machine learning models and PID control algorithms to dynamically generate fan frequency adjustment instructions and valve opening adjustment instructions based on the above data to achieve precise control, effectively solving the problem of inaccurate control of traditional systems and improving the smoke exhaust effect. The hierarchical control module controls the air outlet according to the independent air outlets and public air ducts, making the air duct pressure more stable and ensuring the air outlet effect in each area. The energy-saving optimization module uses frequency optimization algorithms and energy feedback technology to enable the fan to operate in the optimal efficiency range, and dynamically allocates the smoke exhaust volume of each air outlet based on the smoke exhaust priority, greatly improving energy utilization efficiency and reducing energy consumption. The fault warning module monitors the fan current, voltage and air duct pressure data in real time, uses the anomaly detection model to promptly identify abnormal equipment status, triggers multi-level alarms and protection actions, greatly enhancing system reliability and reducing losses caused by failures.

[0019] It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the background technology, the drawings required for use in the embodiments of the present invention or the background technology will be described below.

[0021] The accompanying drawings herein are incorporated into and constitute a part of the specification. These drawings illustrate embodiments consistent with the present disclosure and, together with the specification, are used to explain the technical solutions of the present disclosure.

[0022] Figure 1 A schematic diagram of the structure of an intelligent smoke exhaust adjustment system for building with coordinated control of multiple air outlets provided in an embodiment of the present invention;

[0023] Figure 2 for Figure 1 A schematic diagram of the structure of the subunits of the AI ​​cloud intelligent control module 20;

[0024] Figure 3 for Figure 1 A schematic diagram of the structure of the subunits of the middle layer control module 30;

[0025] Figure 4 for Figure 1 A schematic diagram of the structure of the subunits of the energy-saving optimization module 40;

[0026] Figure 5 for Figure 1 A schematic structural diagram of the subunits of the fault warning module 50;

[0027] Figure 6A flow chart of an intelligent smoke exhaust adjustment method for building with coordinated control of multiple air vents provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0028] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0029] References herein to "embodiments" mean that a particular feature, structure, or characteristic described in connection with the embodiments may be included in at least one embodiment of the present invention. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor does it constitute a separate or alternative embodiment that is mutually exclusive of other embodiments. It is understood, both explicitly and implicitly, by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0030] See also Figure 1 , Figure 1 This is a structural diagram of a building smoke exhaust intelligent adjustment system for multi-air outlet coordinated control provided by an embodiment of the present invention. Figure 1 As shown, the system includes the following modules:

[0031] The data acquisition module 10 is used to collect the exhaust demand data corresponding to each air outlet and the public air duct pressure data in real time. The exhaust demand data includes the number of burners used, valve opening and fan operating parameters corresponding to each air outlet;

[0032] In terms of data collection, the number of burners in use at each tuyere directly reflects the current demand for exhaust volume in that area. For example, in a restaurant kitchen, the more burners in use, the more exhaust fumes and other waste gases are generated, and the higher the exhaust volume required from the exhaust system. By collecting this data, the system can dynamically adjust according to actual demand, avoiding insufficient exhaust or wasted air volume. The number of valves open reflects the current exhaust status of the tuyere, closing valves corresponding to unused stoves. By collecting this data and real-time pressure demand, the system can promptly adjust the speed of the restaurant's independent fans and the opening of the valves behind the fans to ensure smooth exhaust and enhance extraction efficiency.

[0033] Fan operating parameters, including current, voltage, and speed, provide a direct indicator of the fan's health and performance. Abnormal current or voltage may indicate a circuit fault or load problem. Comparing speed data with actual demand can determine whether the fan is operating within reasonable conditions. By collecting these parameters, the system can promptly identify potential fan failures and optimize its operating status.

[0034] Common duct pressure data is crucial for the airflow balance of the entire smoke exhaust system. Duct pressure fluctuates when multiple vents operate simultaneously, or when the air volume at a particular vent changes suddenly. After collecting this data, the system adjusts fan frequency, valve opening, and other factors to stabilize duct pressure, ensuring that each vent achieves the desired airflow and preventing exhaust congestion in certain areas due to uneven pressure.

[0035] In terms of hardware, high-precision sensors are used. For example, pressure sensors are installed behind individual fans and at key points in the common air duct to collect real-time air duct pressure data. Infrared sensors are deployed at each air outlet to monitor burner usage, and built-in opening sensors in electric valves capture valve opening information. Sensors for vibration, temperature, current, voltage, and speed are installed on the fans to collect fan operating parameters. These sensors transmit data to a data processing unit via an industrial bus or wireless communication module. The data processing unit performs pre-processing on the raw data, including filtering and calibration, to remove noise and improve data accuracy. The processed data is then packaged and uploaded to the cloud-based AI intelligent control module.

[0036] Through the deployment of multiple types of high-precision sensors and precise data collection and processing, it is possible to obtain real-time, comprehensive and accurate smoke exhaust demand data for each air outlet and public air duct pressure data, providing a reliable data basis for the subsequent precise control of the AI ​​cloud intelligent control module. This changes the situation in which traditional systems have poor control effects due to untimely and inaccurate data collection, making the operation decisions of the smoke exhaust system more scientific and reasonable.

[0037] The AI ​​cloud intelligent control module 20 is deployed in the cloud and includes a machine learning model and a PID control algorithm for dynamically generating fan frequency adjustment instructions and valve opening adjustment instructions based on the smoke exhaust demand data and the public air duct pressure data;

[0038] An AI computing platform is built in the cloud, and a deep learning framework is used to build a machine learning model. During the model training phase, a large amount of historical smoke exhaust demand data, public duct pressure data, and the corresponding optimal fan frequency and valve opening data are collected as training samples. The model parameters are continuously optimized through the back-propagation algorithm so that the model can accurately learn the correlation between the data. The PID control algorithm calculates the deviation between the current state and the target state based on the real-time collected data, and dynamically adjusts the output parameters through the three-step operation of proportion, integration, and differentiation. When the data acquisition module uploads new data, the machine learning model predicts the preliminary fan frequency and valve opening adjustment trend. The PID control algorithm performs fine-tuning on this basis, and finally generates accurate fan frequency adjustment instructions and valve opening adjustment instructions, which are then sent to the corresponding control equipment through the network communication protocol.

[0039] Combining the machine learning model with the PID control algorithm enables dynamic and precise control of fan frequency and valve opening. The machine learning model can mine the potential patterns in complex data and adapt to changes under different working conditions, while the PID control algorithm ensures the stability and rapid responsiveness of the control. Compared with the traditional fixed parameter control method, this module enables the smoke exhaust system to adjust its operating status in real time according to actual needs, significantly improving the control accuracy and adaptability of the system, and effectively solving the problem of the traditional system's extensive control and difficulty in meeting changing needs.

[0040] A hierarchical control module 30 is used to control the air outlet in a hierarchical manner according to the independent air outlets and the public air duct;

[0041] The air duct network of the smoke exhaust system is divided into an independent air outlet control layer and a public air duct control layer. The independent air outlet control layer is equipped with an independent electric regulating valve and a small controller for each air outlet. The small controller receives the valve opening adjustment instruction issued by the AI ​​cloud intelligent control module and controls the electric regulating valve to accurately adjust the air outlet air volume. The public air duct control layer installs high-power fans and pressure regulating valves on the main air duct line, and monitors the public air duct pressure in real time through pressure sensors. When the pressure fluctuates, the fan speed and pressure regulating valve opening are adjusted according to the instructions of the AI ​​cloud intelligent control module to maintain the stability of the public air duct pressure. At the same time, an inter-layer communication protocol is set up to realize data interaction and collaborative work between the independent air outlet control layer and the public air duct control layer to ensure that the air outlet control of the entire system is carried out in an orderly manner.

[0042] The layered control approach makes the smoke exhaust system's airflow control more flexible and efficient. The independent air outlet control layer can meet the personalized airflow requirements of each area, enhancing the user experience; while the public air duct control layer ensures overall system pressure balance, preventing changes in airflow from a single air outlet from affecting other areas. The collaborative operation of these two layers effectively resolves the duct pressure disturbances and uneven airflow across different areas that are common in traditional systems with coordinated control of multiple air outlets, improving the overall performance and stability of the smoke exhaust system.

[0043] The energy-saving optimization module 40 is used to operate the fan in the optimal efficiency range through a frequency optimization algorithm and dynamically allocate the smoke exhaust volume of each air outlet based on the smoke exhaust priority;

[0044] The frequency optimization algorithm uses intelligent optimization algorithms such as genetic algorithms or particle swarm optimization algorithms, combining fan performance curves and real-time operating data to establish a fan efficiency optimization model. During operation, the algorithm continuously searches for the optimal fan frequency, ensuring the fan operates within its optimal efficiency range. Energy regeneration technology integrates an energy regeneration device into the fan drive circuit. When the fan is braking or under light load, the motor generates electrical energy that is converted into AC power and fed back to the grid, achieving energy recovery. Furthermore, a smoke exhaust priority database is established, presetting the smoke exhaust priority for each outlet based on the real-time air volume and pressure data of each individual fan, as well as the functions and usage scenarios of different locations (such as kitchens and conference rooms). The energy-saving optimization module dynamically adjusts the opening of each outlet's electric valve based on instructions from the AI ​​cloud intelligent control module and the smoke exhaust priority, rationally allocating exhaust volume and minimizing energy consumption while meeting exhaust requirements.

[0045] By utilizing frequency optimization algorithms, real-time data collection of each individual fan's operating parameters, and energy feedback technology, this module significantly improves fan efficiency, reduces energy waste, and lowers the operating costs of the smoke exhaust system. Dynamic allocation of exhaust volume based on exhaust priority enables the system to rationally allocate resources, ensuring effective exhaust in key areas while avoiding excess airflow in other areas, further improving energy efficiency. Compared to traditional smoke exhaust systems, this module significantly enhances the system's energy-saving performance, meeting the development needs of green buildings.

[0046] The fault warning module 50 is used to monitor the fan vibration, temperature, current, voltage and duct pressure data in real time, identify abnormal equipment status through an anomaly detection model, and trigger multi-level alarms and protection actions.

[0047] Edge computing devices are used to collect real-time data on fan vibration, temperature, current, voltage, and duct pressure, and this data is transmitted to an anomaly detection model deployed locally or in the cloud. This anomaly detection model uses deep learning-based methods, such as autoencoders or isolation forest algorithms, to build a characteristic model of the equipment's normal operation by learning from historical normal operation data. When there is a significant deviation between the real-time data and the normal characteristic model, the equipment is deemed to be in an abnormal state or the pipeline is deemed to be clean. A multi-level alarm mechanism is implemented based on the severity of the anomaly. For example, minor anomalies can send early warnings to maintenance personnel via text messages or WeChat. Severe anomalies can immediately trigger equipment shutdown protection, sound an on-site alarm via audio and visual alarms, and call emergency personnel. Furthermore, the abnormal data and alarm information are stored in a database to facilitate subsequent fault analysis and resolution.

[0048] The fault warning module provides real-time, comprehensive monitoring of the operating status of smoke exhaust system equipment. Using advanced anomaly detection models, it promptly and accurately identifies equipment anomalies, significantly improving the timeliness and accuracy of fault detection compared to traditional manual inspections or simple threshold judgments. Multi-level alarm and protection mechanisms effectively prevent equipment failures from escalating, reducing downtime and repair costs caused by equipment failures, ensuring the stable operation of the smoke exhaust system and improving system reliability and safety.

[0049] See also Figure 2 In one embodiment, the AI ​​cloud intelligent control module 20 includes:

[0050] The dynamic coordination logic unit 201 is used to calculate the duct pressure distribution based on the required air volume of each duct and the common duct pressure data through the fluid mechanics simulation model and generate a global pressure balance instruction;

[0051] The feedforward compensation unit 202 is used to preset the load curve based on the historical smoke exhaust peak period data and optimize the fan frequency adjustment response speed;

[0052] The resonance avoiding unit 203 is used to embed the FFT analysis module in the frequency control algorithm to automatically avoid the structural resonance frequency band.

[0053] Dynamic Coordination Logic Unit 201 uses computational fluid dynamics (CFD) principles to build a digital twin model of the air duct system. Finite element analysis discretizes the duct network into nodes and pipe units, inputting the required air volume for each outlet as boundary conditions. A lightweight solver was developed and calibrated using data from common duct pressure sensors. The pressure distribution is calculated iteratively every 500ms. A valve opening adjustment matrix is ​​generated based on the pressure field distribution, and a gradient descent algorithm is used to optimize the action sequence of each control valve to ensure pressure fluctuations are controlled within a ±5Pa range.

[0054] Feedforward compensation unit 202 uses a historical data mining algorithm to extract load characteristics for typical periods, such as weekdays / weekends and peak hours for breakfast, lunch, and dinner, and establishes a multi-dimensional time series prediction model. Preconditioning is initiated 15 minutes before the predicted peak period, dynamically adjusting PID parameters using a fuzzy control algorithm to improve system response speed. Using an online learning mechanism, the prediction model parameters are updated weekly to adapt to load fluctuations caused by seasonal changes and changes in tenant habits.

[0055] Resonance avoidance unit 203 is used for spectral feature extraction. An FFT analysis module is embedded in the fan control loop, sampling the motor current signal every 10 seconds and generating a real-time spectrum using a short-time Fourier transform. A database of resonant frequencies of the fan-duct structure is established, and a taboo search algorithm is used to automatically avoid resonant frequency bands during frequency adjustment. A vibration sensor is installed at the fan base. Detecting abnormal vibration triggers frequency fine-tuning, forming a closed-loop control system of "spectral analysis-frequency avoidance-vibration feedback."

[0056] This embodiment utilizes dynamic calculations based on a fluid dynamics model to significantly improve system pressure control accuracy, effectively resolving pressure fluctuations caused by multiple air outlet coupling. A feedforward compensation mechanism shortens the system's response time to load changes, enabling the establishment of a stable wind pressure field in advance, particularly during peak catering hours. The resonance avoidance function reduces fan vibration amplitude by over 30%, minimizing mechanical wear and significantly lowering operation and maintenance costs. The combination of global pressure optimization and precise control reduces overall system energy consumption compared to traditional control methods, achieving the dual goals of dynamic energy conservation and stable operation.

[0057] See also Figure 3 In one embodiment, the hierarchical control module 30 includes:

[0058] An independent air outlet control unit 301 is used to dynamically adjust the exhaust fan frequency and the electric control valve opening of the corresponding air outlet according to the adjustment instruction;

[0059] The common air duct control unit 302 is used to dynamically adjust the frequency of the total induced draft fan according to the common air duct pressure data.

[0060] The independent tuyere control unit 301 is equipped with an edge computing node for each tuyere. It receives adjustment commands from the AI ​​cloud intelligent control module 20 via the Modbus RTU protocol and locally executes the PID control algorithm. Multi-parameter fusion control is employed, including fan frequency regulation using space vector pulse width modulation (SVM) with an adjustment accuracy of 0.1 Hz; valve opening control using a stepper motor with an angle control accuracy of ±0.5°; and adaptive filtering using a Kalman filter to process sensor data and eliminate measurement noise caused by mechanical vibration. If communication is interrupted, the node automatically switches to a locally preset control curve, dynamically adjusting based on the number of burners.

[0061] The public air duct control unit 302 divides the public air duct into 3-5 pressure control zones, each of which is equipped with an independent pressure sensor and regulating valve. The unit adopts a master-slave control strategy; the master controller adjusts the frequency of the total induced draft fan based on a fuzzy PID algorithm with a response speed of less than 500ms; the slave controller uses feedforward-feedback compound control for each zone regulating valve to compensate for local pressure fluctuations. The optimal duct pressure is calculated in real time based on the required air volume of each air outlet to form a dynamic pressure setting curve. In this way, the control accuracy of the air volume of the independent air outlet can be improved, the pressure fluctuation of the public air duct can be controlled within ±10Pa, the air outlet adjustment response time is less than 1 second, and the air duct pressure recovery time is less than 3 seconds. The energy consumption of the total induced draft fan is significantly reduced, and the distributed architecture supports single-point fault isolation without affecting the overall operation of the system.

[0062] See also Figure 4 In one embodiment, the energy-saving optimization module 40 includes:

[0063] Frequency optimization unit 401, used to dynamically adjust the reference frequency according to the aging degree of the fan to ensure that the total pressure of the fan does not fall below the safety threshold;

[0064] Equipment aging assessment model: Establish a fan efficiency attenuation curve model, through Multiplied by the current efficiency, is the aging coefficient, For time, The base of the natural logarithm. Using 12 characteristic parameters such as motor winding temperature and bearing vibration frequency, the random forest algorithm is used to assess the degree of aging in real time. The model parameters are automatically calibrated quarterly to adapt to the differences in aging characteristics of fans from different brands.

[0065] Dynamic reference frequency adjustment:

[0066] Safety threshold setting: Full pressure safety margin ≥ 1.2 times the current demand pressure

[0067] Frequency compensation can be calculated based on the aging compensation coefficient and the pressure fluctuation coefficient:

[0068] ;

[0069] Where, is the frequency compensation value, 、 are the aging compensation coefficient and the pressure fluctuation coefficient, is the fan operating time, The difference between the current measured pressure of the public air duct and the target pressure;

[0070] Adjustment step: 0.5Hz / time, adjustment cycle: 5 minutes.

[0071] Energy feedback unit 402, used to realize regenerative energy feedback through the common DC bus system to improve system energy efficiency;

[0072] An IGBT rectifier converts regenerative energy into DC bus voltage. A supercapacitor energy storage module absorbs short-term, high-energy feedback, prioritizing other electrical devices on the same bus. The remaining energy is fed back to the grid via a PWM inverter. The SVPWM algorithm achieves unity power factor grid connection, with real-time monitoring of grid current harmonic content to ensure THD is less than 3%. In the event of a grid fault, the system automatically switches to independent energy storage mode.

[0073] The hysteresis control unit 403 is used to implement a hysteresis control strategy within a set pressure range to reduce frequent fan adjustments. In the normal range, the current frequency is maintained and only the valve opening is adjusted. In the alert range, the frequency is adjusted within ±1Hz and the PID parameters are adaptively adjusted. In the emergency range, the rapid response mechanism is triggered and the frequency adjustment step size is increased to 5Hz.

[0074] This embodiment can improve the overall energy efficiency of the system. The energy regeneration unit can recover 20%-25% of braking energy, reducing the total system energy consumption by approximately 5%. Dynamic frequency compensation ensures that the fan always operates in the high-efficiency range, extending the equipment life. Pressure control enhances stability, with an emergency response time of less than 2 seconds, preventing the risk of oil fume backflow. The aging prediction function provides early warning of potential failures, reducing unexpected downtime losses.

[0075] See also Figure 5 In one embodiment, the fault warning module 50 includes:

[0076] A graded alarm unit 501 is used to trigger different levels of alarms according to the severity of the abnormality;

[0077] Abnormal feature extraction, including current anomaly: wavelet transform is used to extract the high-frequency component of the motor current signal, and the threshold is set to ≥1.2 times the rated current; pressure overrun: three-level pressure thresholds are set; vibration overrun: FFT is used to analyze the bearing vibration spectrum, and an early warning is triggered when the specific frequency amplitude is ≥0.8g;

[0078] Level 1 Alarm (Yellow Warning): Pressure fluctuation is greater than P1 or current abnormality lasts for less than 5 minutes; Notification method: SMS + APP push to on-duty personnel; Response requirement: Handle within 4 hours;

[0079] Level 2 alarm (orange warning): Pressure fluctuation is greater than P2 or current abnormality persists for more than 5 minutes; Notification method: SMS + APP + phone call (Level 3 responsible person); Response requirement: Arrive at the scene within 1 hour.

[0080] Level 3 alarm (red alert): Pressure fluctuation is greater than P3 or characteristic frequency of bearing failure is detected; Notification method: Sound and light alarm + full system broadcast + emergency call; Response requirement: Activate emergency plan within 15 minutes.

[0081] The automatic protection unit 502 is used to automatically cut off the power supply and start the backup fan when it detects that the pressure exceeds the limit or the current is abnormal.

[0082] Fast fuse: response time is less than 10ms, cutting off short-circuit current;

[0083] Solid-state relay: contactless switching, life span greater than 10 7 times, the switching time is less than 5ms;

[0084] UPS backup power supply: ensures that the control system continues to work for 30 minutes after the main power failure;

[0085] Pressure overlimit protection: When the air duct pressure is greater than P3 for 2 seconds, the faulty air outlet valve will be closed immediately and the backup fan will be started;

[0086] Abnormal current protection: If the motor current exceeds 1.5 times the rated value for 3 seconds, the power supply of the fan will be cut off and the backup fan will be put into use;

[0087] Standby fan start-up: Soft start is adopted, the start-up time is less than 15 seconds, and the air volume loss during the switching process is ≤20%.

[0088] This embodiment can quickly respond to system failures. The hierarchical alarm mechanism greatly improves the timeliness of fault handling, and automatic protection prevents equipment damage from escalating, thereby reducing fault repair costs.

[0089] See also Figure 6 In one embodiment, the present invention further provides a building smoke exhaust intelligent adjustment method for coordinated control of multiple air outlets, which is applied to any of the above-mentioned building smoke exhaust intelligent adjustment systems for coordinated control of multiple air outlets, wherein the method comprises the following steps:

[0090] S10, real-time collection of smoke exhaust demand data of each air outlet and public air duct pressure data;

[0091] Install at each air outlet: infrared counter to detect the number of burners; electric valve angle sensor; anemometer; arrange in the public air duct: pressure sensor array, temperature sensor;

[0092] S20, calculating the required air volume of each air outlet and the target pressure of the public air duct based on the smoke exhaust demand data of each air outlet and the public air duct pressure data, and generating fan frequency and valve opening adjustment instructions;

[0093] Preferably, the calculation of the required air volume of each air outlet includes:

[0094] ;

[0095] in,

[0096] Q: Merchants’ total demand for air volume;

[0097] : Single burner base air volume;

[0098] w: adjustment coefficient for stove type (frying stove, steamer, clay pot stove, etc.);

[0099] The fan frequency is calculated using the following formula:

[0100] ;

[0101] in,

[0102] F: fan frequency setting;

[0103] : fan aging coefficient;

[0104] t: current ambient temperature;

[0105] Q: Merchants’ total ventilation demand;

[0106] N: number of fans in parallel;

[0107] : Fan manufacturer performance parameter table / function.

[0108] Among them, when collecting the parameters of the above formula, it is necessary to base it on real-time wind pressure monitoring parameters and on-site temperature.

[0109] The target pressure for public air ducts is calculated using a simplified fluid dynamics model and matrix operations. An LSTM network is used to predict load trends for the next 15 minutes, with input data exceeding 5,000 entries. PID control parameter adaptive adjustment: Based on a fuzzy control rule library, each parameter is dynamically optimized according to system response characteristics.

[0110] S30: Execute coordinated adjustment of independent air outlets and public air ducts according to the adjustment instructions to ensure the balance between local smoke exhaust demand and global pressure; use the frequency optimization algorithm to make the fan operate in the optimal efficiency range, and dynamically allocate the smoke exhaust volume of each air outlet based on the smoke exhaust priority;

[0111] Independent vent control: Distributed control based on edge computing nodes, locally executing the PID algorithm. Public duct control: Centralized decision-making in the cloud, with instructions issued via Industrial Ethernet. The coordinated regulation strategy includes constructing a pressure balance matrix: a transfer function matrix is ​​established between vent opening and duct pressure. Feedforward compensation is used to eliminate the effects of coupling between vents, with a compensation rate of ≥90%. Kitchen vents are prioritized over other vents.

[0112] S40: Monitor the fan current, voltage and duct pressure data in real time, identify abnormal equipment status through anomaly detection models, and trigger multi-level alarms and protection actions.

[0113] Efficiency optimization control: Based on the fan's PQ curve, the golden section method is used to search for the highest efficiency point. Dynamic slip compensation: Real-time monitoring of motor speed compensates for slip caused by load changes. A tiered response mechanism is implemented: Level 1 alert: The system automatically records and generates a maintenance work order; Level 2 alert: SMS notification to maintenance personnel, with a response within 4 hours; Level 3 alert: An emergency shutdown is triggered, with the backup system activated within 15 minutes.

[0114] Preferably, in one embodiment, the method further comprises:

[0115] Adopting a master-slave control architecture, the main induced draft fan acts as the master control unit to broadcast the pressure target value, and each independent air outlet fan acts as a slave control unit to synchronously adjust the frequency;

[0116] Fan pressure calculation: ;

[0117] Friction resistance calculation: ;

[0118] Static resistance calculation: ;

[0119] Overall balance relationship: ;

[0120] in,

[0121] : valve opening;

[0122] : Merchant fan air volume;

[0123] : Merchant fan frequency;

[0124] L: duct length;

[0125] W: duct width;

[0126] H: Duct height;

[0127] : Fan manufacturer's wind pressure table / function;

[0128] : Friction coefficient of air duct;

[0129] : Actual wind pressure of the fan;

[0130] k: pressure / temperature correction coefficient;

[0131] : Local resistance coefficient (elbows, reducers, purifier equipment...);

[0132] : air density;

[0133] Pipeline flow rate;

[0134] : Pipeline friction coefficient;

[0135] The air volume for each store is calculated based on the burner on / off status. The resistance of the entire building is then calculated based on the air volume and duct specifications. This allows for dynamic adjustments to each store's valve opening and the negative pressure balance of the public fan to achieve balanced ventilation in the building. This requires collecting real-time monitoring parameters for the most unfavorable point behind each individual fan, as well as on-site temperature.

[0136] Through the above embodiments, the flow balance of each branch is ensured through dynamic proportional distribution; the changes in pipeline network characteristics can be automatically compensated through adaptive adjustment; and the system can be fault-tolerant due to automatic switching to degraded mode in the event of a single point failure.

[0137] Preferably, the method further comprises:

[0138] When it is detected that the public air duct pressure exceeds the threshold, it automatically switches to full-frequency operation mode and starts the backup fan;

[0139] When the system is disconnected from the network, the basic control logic is maintained through the edge computing nodes to ensure the continuous operation of the smoke exhaust system.

[0140] In actual applications, public air ducts may experience abnormal pressure increases due to a variety of factors. For example, during a fire, a large amount of hot, dense smoke can instantly flood into the duct, dramatically increasing the pressure within it. Alternatively, blockages such as foreign objects or structural damage can cause pressure imbalances within the duct. If the smoke exhaust system remains in normal operation, it may not be able to effectively exhaust the smoke, causing it to accumulate within the building. This poses a serious threat to personnel safety and hinders firefighting and rescue efforts.

[0141] When the pressure in the common air duct is detected to exceed the threshold, the system automatically switches to full-frequency operation mode and starts the backup fan, including:

[0142] Pressure monitoring: High-precision pressure sensors are installed at key locations in public air ducts (such as the duct inlet, middle section, and outlet) to monitor pressure changes in the duct in real time. The pressure sensors transmit the collected data to the system control unit.

[0143] Threshold setting: According to the type and scale of the building and the design parameters of the smoke exhaust system, a reasonable pressure threshold is set in the system control unit in advance.

[0144] Automatic switching logic: The system control unit continuously receives data from the pressure sensor and compares it with a set threshold. If the pressure exceeds the threshold, the control unit immediately issues a command to switch the fan operation mode to full-frequency operation. Simultaneously, a start signal is sent to the control module of the backup fan, prompting it to quickly start and work in conjunction with the primary fan.

[0145] Operation status monitoring: During the switching and startup process, the system monitors the operating status of the main and backup fans in real time, including speed, current, voltage, and other parameters. If any abnormal operation of the equipment is detected, an alarm signal will be issued in a timely manner, and appropriate troubleshooting measures will be taken, such as shutting down the faulty fan and starting the backup equipment.

[0146] After automatically switching to full-frequency operation, the fans can run at full capacity, increasing smoke exhaust speed, rapidly reducing pressure in the common air duct, effectively exhausting dense smoke, and improving the exhaust effect of idiopathic strong smoke. The backup fan is activated as a redundant design. If the main fan fails or cannot meet the smoke exhaust requirements, the backup fan can be put into operation in time, ensuring the continuous and stable operation of the smoke exhaust system and reducing the risk of smoke exhaust failure due to equipment failure.

[0147] Modern smoke exhaust systems typically rely on the internet for remote monitoring, data transmission, and intelligent control. However, network environments are complex and volatile, and system disconnections may occur due to natural disasters (such as lightning strikes and earthquakes), network equipment failures, or human sabotage. Once disconnected, if the smoke exhaust system cannot operate normally, smoke will not be discharged in a timely manner in an emergency such as a fire, seriously affecting personnel safety and firefighting rescue efforts. Therefore, in this embodiment, when the system is disconnected, the edge computing nodes are used to maintain basic control logic to ensure the continuous operation of the smoke exhaust system. Specifically, this includes:

[0148] Edge computing node deployment: Edge computing nodes are installed locally within the smoke exhaust system. These devices provide independent data processing, storage, and logical operations capabilities. These nodes establish wired or wireless connections with various smoke exhaust system devices (such as fans, sensors, and controllers) to enable data exchange.

[0149] Preconfigured basic control logic: The basic control logic of the smoke exhaust system is preconfigured in the edge computing node, including rules such as starting the fan based on smoke sensor data and adjusting the fan speed based on temperature sensor data. These logic rules are based on the building's fire protection design specifications and the operating requirements of the smoke exhaust system.

[0150] Data caching and synchronization: When the network is normal, edge computing nodes receive and store configuration information, historical data, and other information from cloud servers, and regularly synchronize data with the cloud. When the system is disconnected from the network, edge computing nodes operate independently using locally stored data and preconfigured logic, while locally caching new data generated during operation.

[0151] Network recovery processing: When the network is restored, the edge computing node automatically uploads the data cached during the outage to the cloud server, integrating and updating it with the cloud data to ensure the integrity and consistency of the system data. At the same time, it receives the latest configuration information and control instructions from the cloud server and resumes collaborative working mode with the cloud.

[0152] This allows edge computing nodes to operate independently even when disconnected from the network, maintaining the basic control logic of the smoke exhaust system and ensuring the proper functioning of essential functions such as fan start / stop and speed regulation. This prevents the smoke exhaust system from failing due to network outages, ensuring system continuity in emergency situations. During network outages, the smoke exhaust system can continue to operate according to pre-set rules or switch to power frequency, ensuring overall emergency response capabilities. This reduces over-reliance on the network, lowers the risk of system failure due to network issues, enhances the stability and reliability of the smoke exhaust system, and improves the system's autonomous operation capabilities.

[0153] In summary, the intelligent smoke exhaust control system and method provided by the present invention utilizes a data acquisition module to acquire real-time smoke exhaust demand data for each air outlet and common duct pressure data, providing a detailed basis for subsequent precise control. The AI ​​cloud intelligent control module utilizes a machine learning model and PID control algorithm to dynamically generate fan frequency adjustment commands and valve opening adjustment commands based on this data, achieving precise control and effectively resolving the issue of inaccurate control in traditional systems and improving smoke exhaust efficiency. The hierarchical control module controls air output based on individual air outlets and common ducts, stabilizing duct pressure and ensuring effective airflow in each area. The energy-saving optimization module utilizes a frequency optimization algorithm and energy feedback technology to ensure fan operation within the optimal efficiency range and dynamically allocates exhaust volume to each air outlet based on exhaust priority, significantly improving energy efficiency and reducing energy consumption. The fault warning module monitors fan current, voltage, and duct pressure data in real time, employs anomaly detection models to promptly identify equipment abnormalities, and triggers multi-level alarms and protection actions, significantly enhancing system reliability and reducing losses caused by failures.

[0154] An embodiment of the present invention further provides a computer-readable storage medium, in which a computer program is stored. The computer program includes program instructions, and when the program instructions are executed by a processor of an electronic device, the processor executes a method as described in any one of the possible implementation methods described above.

[0155] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present invention.

Claims

1. An intelligent smoke exhaust control system for buildings with coordinated control of multiple air outlets, characterized in that: The system comprises: The data acquisition module is used to collect the exhaust demand data corresponding to each air outlet and the public air duct pressure data in real time. The exhaust demand data includes the number of burners used, valve opening and fan operating parameters corresponding to each air outlet; The AI ​​cloud intelligent control module is deployed in the cloud and includes a machine learning model and a PID control algorithm, which is used to dynamically generate fan frequency adjustment instructions and valve opening adjustment instructions based on the smoke exhaust demand data and the public duct pressure data; Hierarchical control module, used to control air outlet in different layers according to independent air outlets and public air ducts; Energy-saving optimization module, which uses frequency optimization algorithm to make the fan operate in the optimal efficiency range and dynamically allocates the exhaust volume of each air outlet based on the exhaust priority; The fault warning module is used to monitor fan vibration, temperature, current, voltage and duct pressure data in real time, identify abnormal equipment status through anomaly detection models, and trigger multi-level alarms and protection actions.

2. The intelligent smoke exhaust adjustment system for building with multi-air outlet coordinated control according to claim 1 is characterized in that: The AI ​​cloud intelligent control module includes: Dynamic coordination logic unit, used to calculate the duct pressure distribution through fluid mechanics simulation model based on the required air volume of each air outlet and the public duct pressure data, and generate global pressure balance instructions; Feedforward compensation unit, used to preset load curve based on historical smoke exhaust peak period data and optimize fan frequency adjustment response speed; The resonance avoidance unit is used to embed the FFT analysis module in the frequency control algorithm to automatically avoid the structural resonance frequency band.

3. The intelligent smoke exhaust adjustment system for building with multi-air outlet coordinated control according to claim 1 is characterized in that: The hierarchical control module includes: An independent air outlet control unit, used to dynamically adjust the exhaust fan frequency and the electric control valve opening of the corresponding air outlet according to the adjustment instruction; The common air duct control unit is used to dynamically adjust the frequency of the total induced draft fan according to the common air duct pressure data.

4. The intelligent smoke exhaust adjustment system for building with multi-air outlet coordinated control according to claim 1 is characterized in that: The energy-saving optimization module includes: Frequency optimization unit, used to dynamically adjust the reference frequency according to the aging degree of the fan to ensure that the full pressure of the fan does not fall below the safety threshold; Energy feedback unit, used to realize regenerative energy feedback through the common DC bus system to improve system energy efficiency; The hysteresis control unit is used to adopt the hysteresis control strategy within the set pressure range to reduce the frequent adjustment of the fan.

5. The intelligent smoke exhaust adjustment system for building with multi-air outlet coordinated control according to claim 1 is characterized in that: The fault warning module includes: A graded alarm unit is used to trigger different levels of alarms according to the severity of the abnormality; The automatic protection unit is used to automatically cut off the power supply and start the backup fan when it detects excessive pressure or abnormal current.

6. A method for intelligent smoke exhaust adjustment in a building for coordinated control of multiple air outlets, applied to the intelligent smoke exhaust adjustment system for coordinated control of multiple air outlets according to any one of claims 1 to 5, characterized in that: include: Real-time collection of smoke exhaust demand data of each air outlet and public air duct pressure data; Calculate the required air volume of each air outlet and the target pressure of the public air duct based on the exhaust demand data of each air outlet and the pressure data of the public air duct, and generate the fan frequency and valve opening adjustment instructions; Execute coordinated adjustment of independent air outlets and public air ducts according to adjustment instructions to ensure the balance between local smoke exhaust demand and global pressure; Through the frequency optimization algorithm, the fan operates in the optimal efficiency range, and the smoke exhaust volume of each air outlet is dynamically allocated based on the smoke exhaust priority; Monitor fan current, voltage, and duct pressure data in real time, identify equipment abnormalities through anomaly detection models, and trigger multi-level alarms and protection actions.

7. The intelligent ventilation adjustment method for building with multi-air outlet coordinated control according to claim 6 is characterized in that: The calculation of the required air volume of each air outlet includes: ； in, Q: Merchants’ total demand for air volume; : Single burner base air volume; w: adjustment coefficient for stove type (frying stove, steamer, clay pot stove, etc.); The fan frequency is calculated using the following formula: ； in, F: fan frequency setting; : fan aging coefficient; t: current ambient temperature; Q: Merchants’ total ventilation demand; N: number of fans in parallel; : Fan manufacturer performance parameter table / function.

8. The intelligent smoke exhaust adjustment method for building with multi-air outlet coordinated control according to claim 6 is characterized in that: The method further comprises: Adopting a master-slave control architecture, the main induced draft fan acts as the master control unit to broadcast the pressure target value, and each independent air outlet fan acts as a slave control unit to synchronously adjust the frequency; Among them, the valve opening is dynamically allocated according to the air volume ratio required by each air outlet. The calculation formula is: Fan pressure calculation: ; Friction resistance calculation: ; Static resistance calculation: ; Overall balance relationship: ; in, : valve opening; : Merchant fan air volume; : Merchant fan frequency; L: duct length; W: duct width; H: Duct height; : Fan manufacturer's wind pressure table / function; : Friction coefficient of air duct; : Actual wind pressure of the fan; k: pressure / temperature correction coefficient; : Local resistance coefficient (elbows, reducers, purifier equipment...); : air density; Pipeline flow rate; : Pipeline friction coefficient; The merchant's air volume is calculated through the furnace switch status, and then the resistance of the entire building is calculated through the air volume and pipeline specifications, so as to dynamically adjust the valve opening of each merchant and the negative pressure balance of the public fan to achieve balanced exhaust in the building.

9. The intelligent smoke exhaust adjustment method for building with multi-air outlet coordinated control according to claim 6 is characterized in that: The method further comprises: When it is detected that the public air duct pressure exceeds the threshold, it automatically switches to full-frequency operation mode and starts the backup fan; When the system is disconnected from the network, the basic control logic is maintained through the edge computing nodes to ensure the continuous operation of the smoke exhaust system.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, which includes program instructions. When the program instructions are executed by a processor of an electronic device, the processor executes the intelligent building smoke exhaust adjustment method for multi-air vent coordinated control as described in any one of claims 6 to 9.

Images