A full-automatic intelligent ventilation and smoke exhaust system for a factory building

Abstract

This invention provides a fully automated intelligent ventilation and smoke extraction system for factories, comprising an environmental sensing module, a data processing and fusion module, an intelligent decision-making and control module, an actuator drive module, a human-machine interaction monitoring module, and a system diagnosis and maintenance module. This invention adjusts fan speed and damper opening in real time based on a comprehensive ventilation and smoke extraction demand index, avoiding unnecessary energy consumption. It employs a multi-sensor dynamic weighted fusion and predictive compensation algorithm, which can trigger adjustment in the early trend stage of anomalies, faster than traditional switch-based alarm methods. Furthermore, dynamic weighting avoids malfunctions caused by instantaneous interference from a single sensor; adaptive feedforward and dead-zone compensation achieve smooth adjustment, reducing mechanical shock. The designed self-diagnosis module can accurately locate fault types.

Description

Technical Field

[0001] This invention relates to the field of ventilation and smoke extraction technology, specifically to a fully automatic intelligent ventilation and smoke extraction system for factory buildings. Background Technology

[0002] In industrial production, ventilation and smoke extraction systems are crucial for ensuring the safety of workers and the normal operation of equipment. Currently, most domestic machining, chemical, painting, and electronic assembly plants still primarily use traditional constant air volume exhaust fans combined with independent smoke detectors for ventilation and smoke extraction. Specifically, when a smoke or heat detector detects a signal exceeding a preset threshold, it sends an alarm signal, which is typically connected to a fire alarm controller. Upon receiving the alarm, the controller starts the smoke extraction fan and opens the corresponding smoke extraction valve through manual confirmation or automatic linkage. Under this architecture, the daily operation of the ventilation system relies on time relays or manual buttons for start / stop control, rarely undergoing dynamic adjustment based on the actual pollutant concentration or heat load inside the plant.

[0003] The aforementioned traditional solutions have several obvious technical limitations. First, because smoke detectors only output switching signals and have high activation thresholds (e.g., light reduction exceeding 5% / m), alarms are often only triggered after significant smoke is generated from the fire source, resulting in a delay in the activation of the ventilation and smoke extraction system. Second, daily ventilation is not adjusted based on feedback from gas concentrations such as CO, CO2, and VOCs, or temperature and humidity, leading to significant wasted energy. Furthermore, the system is completely unresponsive when harmful gases leak slowly but do not reach the smoke alarm threshold. Third, the sensors operate independently, lacking multi-source data fusion capabilities, making it impossible to distinguish between high-temperature smoke and ordinary heat source interference, and difficult to determine the specific location and direction of fire or leak. Consequently, smoke extraction is poor, and smoke frequently spreads to non-fire source areas. Fourth, some improved systems use PLCs and frequency converters to regulate fan speed, but the control logic is usually a single PID or simple temperature threshold comparison, failing to consider the impact of indoor and outdoor air pressure differences on smoke flow and lacking any predictive algorithms. In large-space factory buildings with drastic changes in thermal and wind pressure, this can easily lead to insufficient smoke extraction or excessive air supply. Fifth, existing systems generally lack self-diagnostic functions for actuators and sensors. When the air valve is stuck, the sensor zero point drifts, or communication is interrupted, the system cannot automatically identify and alarm, and continues to operate according to the erroneous signal, posing a significant safety hazard.

[0004] To address the aforementioned issues, some recent studies have attempted to combine wireless sensor networks with host computer monitoring software. For example, multiple temperature, humidity, and smoke monitoring nodes are deployed within the factory, transmitting data via ZigBee to an industrial control computer for display, and allowing manual adjustments to ventilation strategies. However, these solutions remain at the data acquisition and display level, failing to establish closed-loop automatic control, and wireless communication suffers from reliability issues in metal-structured factory buildings. A few papers have also proposed using BP neural networks or support vector machines to predict smoke concentration, but model training requires a large amount of historical data under specific operating conditions, and the computational load is substantial, making real-time operation in low-cost embedded controllers difficult. Therefore, designing a fully automatic intelligent ventilation and smoke extraction system capable of real-time fusion of multi-type sensor information, possessing self-learning and predictive capabilities, and simultaneously achieving closed-loop automatic adjustment and fault self-diagnosis is of great significance for improving factory safety and reducing energy consumption. Summary of the Invention

[0005] To address the aforementioned problems, this invention discloses a fully automated intelligent ventilation and smoke extraction system for factory buildings.

[0006] The specific plan is as follows:

[0007] A fully automated intelligent ventilation and smoke extraction system for factory buildings includes an environmental sensing module, a data processing and fusion module, an intelligent decision-making and control module, an actuator driving module, a human-machine interaction monitoring module, and a system diagnosis and maintenance module. The environmental sensing module collects real-time data on temperature, smoke concentration, carbon monoxide concentration, and indoor / outdoor air pressure difference inside and outside the factory building, converting the raw physical signals into digital signals from the sensors. The data processing and fusion module filters, compensates for missing values, and normalizes the sensor data, and calculates the comprehensive ventilation and smoke extraction demand index (VDI) by fusing multi-source data, while also providing residual data required for diagnosis. The intelligent decision-making and control module calculates the VDI based on the VDI. The system determines the ventilation mode (standby, normal ventilation, emergency smoke exhaust) and uses a fuzzy PID combined with a predictive compensation algorithm to calculate the fan speed command and damper opening command. The actuator drive module receives the control commands and drives the smoke exhaust fan and electric damper to perform corresponding actions, while simultaneously providing feedback on the actual operating status. The human-machine interface monitoring module provides a local human-machine interface and remote monitoring functions, displays system operating data and alarm information, and supports manual / automatic mode switching and parameter settings. The system diagnostic and maintenance module periodically detects sensor drift, actuator jamming, and communication interruption faults and outputs diagnostic results.

[0008] Furthermore, the environmental sensing module includes a temperature sensor, a smoke concentration sensor, a CO concentration sensor, an indoor-outdoor air pressure difference sensor, and a data acquisition terminal; wherein, the temperature sensor and the smoke concentration sensor are arranged on the top of the factory building, the CO concentration sensor is arranged in the work area and near the smoke exhaust outlet, and the indoor and outdoor sides of the indoor-outdoor air pressure difference sensor are respectively placed inside and outside the factory building, and at the same height; the data from each sensor are collected to the data acquisition terminal via an RS-485 bus.

[0009] Furthermore, the data processing and fusion module employs an edge computing gateway, uses time-weighted moving average filtering and normalization processing, and calculates VDI using the following formula:

[0010]

[0011] in, These are the normalized values ​​for temperature, smoke, and CO concentration, respectively. For adaptive weights, The difference between indoor and outdoor air pressure. Maximum permissible pressure difference This is a factor affecting air pressure.

[0012] Furthermore, the intelligent decision control module employs a programmable logic controller (PLC), combining fuzzy PID with a predictive compensation algorithm. The control quantity calculation formula is as follows:

[0013]

[0014] in, This refers to the control voltage for the fan or the opening command for the damper. The deviation between the target demand index and the actual VDI; PID parameters are dynamically adjusted for fuzzy rules; To predict the compensation coefficient; For the future Predicted smoke concentration per second.

[0015] Furthermore, the actuator drive module employs an adaptive feedforward compensation algorithm based on the VDI change rate to improve control accuracy; the feedforward term adjusts the output in advance based on the VDI change rate and the predicted smoke concentration to overcome the large inertial delay of the system; dead zone compensation is used to overcome mechanical friction and clearance. The total output formula is:

[0016]

[0017] Among them, the rate of change of the ventilation demand index reflects the speed of abnormal development; The same as the prediction term in the intelligent decision control module; This is the feedforward gain coefficient. is the influencing factor of the smoke prediction item; is the basic instruction calculated and fed back by the intelligent decision control module; means performing dynamic dead zone compensation on the signals in the parentheses.

[0018] Further, the human-computer interaction monitoring module includes an industrial touch screen and a cloud platform gateway.

[0019] Further, the system diagnosis and maintenance module includes a self-diagnosis unit, which is embedded in the edge computing gateway of the data processing and fusion module, and sends the fault diagnosis result to the intelligent decision control module and the human-computer interaction monitoring module.

[0020] Further, the operation method of the system is as follows:

[0021] S1. The PLC in the intelligent decision control module loads plant parameters; the edge computing gateway in the data processing and fusion module calibrates the zero point of each sensor.

[0022] S2. The temperature, smoke, CO and air pressure difference sensors in the environment perception module send the raw data to the data processing and fusion module via the data acquisition terminal;

[0023] S3. The data processing and fusion module performs filtering, normalization and calculates VDI, and sends the residual to the system diagnosis and maintenance module at the same time;

[0024] S4. The intelligent decision control module reads the VDI value: VDI ≤ 0.2: standby mode, the fan runs at the minimum speed; 0.2 < VDI ≤ 0.6: conventional ventilation, fuzzy PID adjustment; VDI > 0.6: emergency smoke exhaust, the fan runs at full speed, and the air damper is fully opened;

[0025] S5. The intelligent decision control module sends the control instruction to the actuator drive module;

[0026] S6. The actuator drive module returns the actual rotation speed and opening degree to the intelligent decision control module;

[0027] S7. The system diagnosis and maintenance module runs residual detection, and sends the fault result to the intelligent decision control module and the human-computer interaction monitoring module;

[0028] S8. The industrial touch screen of the human-computer interaction monitoring module displays data, and the cloud platform gateway pushes an alarm.

[0029] The beneficial effects of this invention are as follows: It adjusts the fan speed and damper opening in real time based on the comprehensive ventilation and smoke extraction demand index, avoiding unnecessary energy consumption; it employs a multi-sensor dynamic weighted fusion and predictive compensation algorithm, which can trigger adjustment in the early trend stage of anomalies, faster than traditional switch-based alarm methods. Furthermore, dynamic weighting avoids malfunctions caused by instantaneous interference from a single sensor; adaptive feedforward and dead-zone compensation achieve smooth adjustment, reducing mechanical shock. The designed self-diagnostic module can accurately locate the fault type. Attached Figure Description

[0030] Figure 1 This is a system block diagram of the present invention.

[0031] Figure 2 This is a flowchart of the operation method of the present invention. Detailed Implementation

[0032] The present invention will be further illustrated below with reference to the accompanying drawings and specific embodiments. It should be understood that the following specific embodiments are for illustrative purposes only and are not intended to limit the scope of the invention.

[0033] As shown in the figure, this invention provides a fully automated intelligent ventilation and smoke extraction system for factories, including an environmental sensing module, a data processing and fusion module, an intelligent decision control module, an actuator driving module, a human-machine interaction monitoring module, and a system diagnosis and maintenance module. The environmental sensing module collects real-time data on temperature, smoke concentration, carbon monoxide concentration, and indoor / outdoor air pressure difference inside and outside the factory, converting the raw physical signals into digital signals from the sensors. The data processing and fusion module filters, compensates for missing values, and normalizes the sensor data, and calculates the comprehensive ventilation and smoke extraction demand index (VDI) by fusing multi-source data, while also providing residual data required for diagnosis. The intelligent decision control module calculates the VDI based on the VDI... The system determines the ventilation mode (standby, normal ventilation, emergency smoke exhaust) and uses a fuzzy PID combined with a predictive compensation algorithm to calculate the fan speed command and damper opening command. The actuator drive module receives the control commands and drives the smoke exhaust fan and electric damper to perform corresponding actions, while simultaneously providing feedback on the actual operating status. The human-machine interface monitoring module provides a local human-machine interface and remote monitoring functions, displays system operating data and alarm information, and supports manual / automatic mode switching and parameter settings. The system diagnostic and maintenance module periodically detects sensor drift, actuator jamming, and communication interruption faults and outputs diagnostic results.

[0034] In this embodiment, the environmental sensing module includes a temperature sensor, a smoke concentration sensor, a CO concentration sensor, an indoor-outdoor air pressure difference sensor, and a data acquisition terminal. The temperature sensor and the smoke concentration sensor are located on the roof of the factory building, the CO concentration sensor is located in the work area and near the smoke exhaust outlet, and the indoor and outdoor sides of the indoor-outdoor air pressure difference sensor are located inside and outside the factory building, respectively, at the same height. The data from each sensor are collected to the data acquisition terminal via an RS-485 bus.

[0035] Furthermore, the data processing and fusion module employs an edge computing gateway, uses time-weighted moving average filtering and normalization processing, and calculates VDI using the following formula:

[0036]

[0037] in, These are the normalized values ​​for temperature, smoke, and CO concentration, respectively. For adaptive weights, The difference between indoor and outdoor air pressure. Maximum permissible pressure difference This is a factor affecting air pressure.

[0038] In this embodiment, the intelligent decision control module adopts a programmable logic controller (PLC), fuzzy PID combined with a predictive compensation algorithm, and the control quantity calculation formula is as follows:

[0039]

[0040] in, This refers to the control voltage for the fan or the opening command for the damper. The deviation between the target demand index and the actual VDI; PID parameters are dynamically adjusted for fuzzy rules; To predict the compensation coefficient; For the future Predicted smoke concentration per second.

[0041] In this embodiment, the actuator drive module employs an adaptive feedforward compensation algorithm based on the VDI change rate to improve control accuracy. The feedforward term adjusts the output in advance based on the VDI change rate and predicted smoke concentration to overcome the system's large inertial delay. Dead zone compensation is used to overcome mechanical friction and clearance. The total output formula is:

[0042]

[0043] Among them, the rate of change of the ventilation demand index reflects the speed of abnormal development; The same as the prediction term in the intelligent decision control module; This is the feedforward gain coefficient. is the influencing factor of the smoke prediction item; is the basic instruction calculated and fed back by the intelligent decision control module; means performing dynamic dead zone compensation on the signals in the parentheses.

[0044] In this embodiment, the human-computer interaction monitoring module includes an industrial touch screen and a cloud platform gateway.

[0045] In this embodiment, the system diagnosis and maintenance module includes a self-diagnosis unit, which is embedded in the edge computing gateway of the data processing and fusion module, and sends the fault diagnosis result to the intelligent decision control module and the human-computer interaction monitoring module.

[0046] In this embodiment, the operation method of the system is as follows:

[0047] S1. The PLC in the intelligent decision control module loads plant parameters; the edge computing gateway in the data processing and fusion module calibrates the zero point of each sensor.

[0048] S2. The temperature, smoke, CO and air pressure difference sensors in the environment perception module send the original data to the data processing and fusion module via the data acquisition terminal;

[0049] S3. The data processing and fusion module performs filtering, normalization and calculates VDI, and sends the residual to the system diagnosis and maintenance module at the same time;

[0050] S4. The intelligent decision control module reads the VDI value: VDI ≤ 0.2: standby mode, the fan operates at the minimum speed; 0.2 < VDI ≤ 0.6: normal ventilation, fuzzy PID regulation; VDI > 0.6: emergency smoke exhaust, the fan runs at full speed, and the air damper is fully opened;

[0051] S5. The intelligent decision control module sends the control instruction to the actuator driving module;

[0052] S6. The actuator driving module returns the actual rotation speed and opening degree to the intelligent decision control module;

[0053] S7. The system diagnosis and maintenance module runs residual detection, and sends the fault result to the intelligent decision control module and the human-computer interaction monitoring module;

[0054] S8. The industrial touch screen of the human-computer interaction monitoring module displays data, and the cloud platform gateway pushes an alarm.

[0055] The technical means disclosed in this invention are not limited to those disclosed in the above embodiments, but also include technical solutions composed of any combination of the above technical features. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of this invention, and these improvements and modifications are also considered within the scope of protection of this invention.

Claims

1. A fully automatic intelligent ventilation and smoke extraction system for factory buildings, characterized in that, The system comprises an environmental sensing module, a data processing and fusion module, an intelligent decision-making and control module, an actuator driving module, a human-machine interaction monitoring module, and a system diagnosis and maintenance module. The environmental sensing module collects real-time data on temperature, smoke concentration, carbon monoxide concentration, and indoor / outdoor air pressure difference inside and outside the plant, converting the raw physical signals into digital signals from the sensors. The data processing and fusion module filters, compensates for missing values, and normalizes the sensor data, and calculates the Comprehensive Ventilation and Smoke Exhaust Demand Index (VDI) by fusing multi-source data, while also providing residual data for diagnosis. The intelligent decision-making and control module determines the ventilation mode based on the VDI level and uses a fuzzy PID combined with a predictive compensation algorithm to calculate the fan speed command and damper opening command. The actuator driving module receives control commands and drives the smoke exhaust fan and electric damper to perform corresponding actions, while also providing feedback on the actual operating status. The human-machine interaction monitoring module provides a local human-machine interface and remote monitoring functions, displays system operating data and alarm information, and supports manual / automatic mode switching and parameter settings. The system diagnosis and maintenance module periodically detects sensor drift, actuator jamming, and communication interruption faults, and outputs diagnostic results.

2. The fully automatic intelligent ventilation and smoke extraction system for factory buildings according to claim 1, characterized in that, The environmental sensing module includes a temperature sensor, a smoke concentration sensor, a CO concentration sensor, an indoor-outdoor air pressure difference sensor, and a data acquisition terminal. The temperature sensor and the smoke concentration sensor are located on the roof of the factory building, the CO concentration sensor is located in the work area and near the smoke exhaust outlet, and the indoor and outdoor sides of the indoor-outdoor air pressure difference sensor are located inside and outside the factory building, respectively, at the same height. The data from each sensor are collected to the data acquisition terminal via an RS-485 bus.

3. The fully automatic intelligent ventilation and smoke extraction system for factory buildings according to claim 1, characterized in that, The data processing and fusion module employs an edge computing gateway, uses time-weighted moving average filtering and normalization, and calculates VDI using the following formula: in, These are the normalized values ​​for temperature, smoke, and CO concentration, respectively. For adaptive weights, The difference between indoor and outdoor air pressure. Maximum permissible pressure difference This is a factor affecting air pressure.

4. The fully automatic intelligent ventilation and smoke extraction system for factory buildings according to claim 1, characterized in that, The intelligent decision control module adopts a programmable logic controller (PLC), fuzzy PID combined with a predictive compensation algorithm, and the control quantity calculation formula is as follows: in, This refers to the control voltage for the fan or the opening command for the damper. The deviation between the target demand index and the actual VDI; PID parameters dynamically adjusted for fuzzy rules; To predict the compensation coefficient; For the future Predicted smoke concentration per second.

5. A fully automatic intelligent ventilation and smoke extraction system for factory buildings according to claim 1, characterized in that, The actuator drive module employs an adaptive feedforward compensation algorithm based on the VDI change rate to improve control accuracy. The feedforward term adjusts the output in advance based on the VDI change rate and predicted smoke concentration to overcome the system's large inertial delay. Dead zone compensation is used to overcome mechanical friction and clearance. The total output formula is: Among them, the rate of change of the ventilation demand index reflects the speed of abnormal development; The same as the prediction term in the intelligent decision control module; This is the feedforward gain coefficient. The influencing factors for the smoke prediction term; The basic instructions calculated by the intelligent decision-making and control module are fed back. This indicates that dynamic dead-time compensation is performed on the signal within the parentheses.

6. A fully automatic intelligent ventilation and smoke extraction system for factory buildings according to claim 1, characterized in that, The human-machine interaction monitoring module includes an industrial touch screen and a cloud platform gateway.

7. A fully automatic intelligent ventilation and smoke extraction system for factory buildings according to claim 3, characterized in that, The system diagnosis and maintenance module includes a self-diagnosis unit, which is embedded in the edge computing gateway of the data processing and fusion module, and sends the fault diagnosis results to the intelligent decision control module and the human-machine interaction monitoring module.

8. A fully automatic intelligent ventilation and smoke extraction system for factory buildings according to claim 1, characterized in that, The system operates as follows: S1. The PLC in the intelligent decision control module loads the plant parameters; the edge computing gateway in the data processing and fusion module calibrates the zero points of each sensor. S2. The temperature, smoke, CO, and differential pressure sensors in the environmental sensing module send the raw data to the data processing and fusion module via the data acquisition terminal; S3, the data processing and fusion module filters, normalizes and calculates VDI, and sends the residual to the system diagnosis and maintenance module; S4. The intelligent decision control module reads the VDI value: VDI ≤ 0.2: standby mode, fan runs at the lowest speed; 0.2 < VDI ≤ 0.6: normal ventilation, fuzzy PID adjustment; VDI > 0.6: emergency smoke exhaust, fan at full speed, air valve fully open; S5, the intelligent decision control module sends control commands to the actuator drive module; S6. The actuator drive module transmits the actual rotation speed and opening degree back to the intelligent decision control module; S7. The system diagnostic and maintenance module performs residual detection and sends the fault results to the intelligent decision control module and the human-machine interaction monitoring module. The S8 human-machine interaction monitoring module displays data on an industrial touchscreen, and the cloud platform gateway pushes alarms.

Images