Tunnel fire monitoring and ventilation smoke exhaust linkage control system

Abstract

The invention provides a tunnel fire monitoring and ventilation and smoke exhaust linkage control system. The tunnel fire monitoring and ventilation and smoke exhaust linkage control system comprises a tunnel environment and fire monitoring module, a data fusion processing module, a tunnel ventilation and smoke exhaust linkage control module and a remote monitoring and control management system module which are connected through a communication network. The data fusion processing module is configured with a D-S evidence theory algorithm and is used for receiving multi-source environment data, calculating basic probability distribution of data of each sensor, quantitatively evaluating a fire occurrence probability through a combination rule and generating a fire state judgment result; the tunnel ventilation and smoke exhaust linkage control module is used for receiving the fire state judgment result and determining the target wind speed according to the fire scale; a fuzzy PID control unit and a hysteresis control unit are arranged in the tunnel ventilation and smoke exhaust linkage control module. According to the method, the problem of high false alarm rate of traditional monitoring is solved, the engineering problem that discrete ventilation equipment is difficult to continuously and accurately control is solved, and the disaster prevention and reduction capability of the tunnel is remarkably improved.

Description

Technical Field

[0001] This invention relates to a tunnel fire monitoring and control system, specifically a tunnel fire monitoring and ventilation smoke extraction linkage control system, belonging to the field of tunnel fire protection and intelligent control technology. Background Technology

[0002] With the rapid development of transportation infrastructure, the construction scale and operational mileage of highway and railway tunnels have continued to grow. As enclosed or semi-enclosed narrow spaces, tunnels are highly susceptible to fires due to the rapid spread and high concentration of high-temperature smoke, which is difficult to dissipate, easily leading to asphyxiation and injury or damage to transportation facilities. Furthermore, the complex environment within tunnels means that traditional fire monitoring methods, often relying on single sensors, suffer from high false alarm rates and delayed response times. Ventilation and smoke extraction systems often operate in fixed modes, unable to dynamically adjust strategies according to the fire's development stage. This can result in excessively high wind speeds that exacerbate fire spread or insufficient wind speeds that cause smoke stagnation, severely impacting evacuation and fire rescue efficiency. The ventilation and smoke extraction systems operate independently from the fire monitoring module, lacking a linkage mechanism. Manual intervention is required after a fire breaks out to adjust the ventilation strategy, delaying optimal smoke control. In addition, the fan speed control of ventilation and smoke extraction systems uses traditional PID control algorithms with fixed parameters, which cannot cope with the non-linear changes in parameters such as wind speed, temperature, and smoke concentration during tunnel fires, making flexible and stable wind speed control difficult. Since ventilation equipment such as jet fans are discrete actuators, continuous control signals cannot be directly mapped to the number of fans that are turned on, which can easily lead to fluctuations in wind speed or frequent start-ups and shutdowns of the equipment, thus shortening the equipment's lifespan. Summary of the Invention

[0003] Based on the above background, the purpose of this invention is to provide a tunnel fire monitoring and ventilation smoke exhaust linkage control system, which solves the technical problems of existing tunnel fire monitoring systems, such as high false alarm rate due to reliance on a single sensor, lag in ventilation control response, and the inability of traditional PID control to adapt to nonlinear changes in fire and difficulty in directly controlling discrete fans, resulting in frequent equipment start-ups and shutdowns.

[0004] To achieve the above-mentioned objectives, the present invention provides the following technical solution:

[0005] A tunnel fire monitoring and ventilation smoke exhaust linkage control system includes a tunnel environment and fire monitoring module, a data fusion processing module, a tunnel ventilation smoke exhaust linkage control module, and a remote monitoring and control management system module connected via a communication network.

[0006] The tunnel environment and fire monitoring module is used to collect environmental parameters and fire characteristic data in the tunnel in real time, and transmit them to the data fusion processing module through the communication network.

[0007] The data fusion processing module is equipped with the DS evidence theory algorithm, which is used to receive multi-source environmental data, calculate the basic probability distribution of each sensor data, quantify and evaluate the probability of fire occurrence through combination rules, and generate fire status determination results.

[0008] The tunnel ventilation and smoke exhaust linkage control module is used to receive the fire status judgment result and determine the target wind speed according to the fire scale. The tunnel ventilation and smoke exhaust linkage control module has a built-in fuzzy PID control unit and a hysteresis control unit. The fuzzy PID control unit dynamically adjusts the PID parameters according to the deviation between the target wind speed and the actual wind speed to calculate the total air volume required by the system. The hysteresis control unit maps the continuously changing air volume to the discrete number of jet fans to be turned on, and uses hysteresis logic to prevent the fans from frequently starting and stopping at the critical point.

[0009] The remote monitoring and control management system module is used to monitor the tunnel environment, fire status, and equipment operation in real time, and provides data storage, historical query, and hierarchical alarm functions.

[0010] Preferably, the tunnel environment and fire monitoring module includes a data acquisition terminal, temperature sensors, CO concentration sensors, visibility sensors arranged along the longitudinal direction of the tunnel and the tunnel arch, and wind speed sensors arranged at the tunnel entrances and cross passages; wherein, the temperature sensors, CO concentration sensors, and visibility sensors are distributed at a preset interval at the tunnel arch, and the wind speed sensors are arranged in pairs at the tunnel entrance and exit to monitor the longitudinal wind speed of the tunnel, and the temperature sensors, CO concentration sensors, visibility sensors, and wind speed sensors are all electrically connected to the data acquisition terminal.

[0011] Preferably, the data acquisition terminal has a built-in signal conditioning circuit and a microprocessor and is configured with the following data preprocessing logic:

[0012] The original sampled data is smoothed using a moving average filtering algorithm, and the size of the moving window is set to reduce high-frequency noise interference from the environment.

[0013] The 3σ criterion is used to detect outliers in the filtered data. Data that exceeds the standard deviation of the mean by a multiple of the set value is identified as sensor malfunction or sudden interference and is discarded. Only valid data is transmitted through the communication network.

[0014] Preferably, the data fusion processing module constructs an identification framework U={fire occurred, no fire occurred} based on the Dempster evidence theory. It uses the Dempster combination rule to perform orthogonal and normalized fusion calculations on the basic probability assignments of temperature, CO concentration, and visibility to obtain a comprehensive confidence level. When the comprehensive confidence level supporting the occurrence of a fire is significantly higher than the comprehensive confidence level supporting the absence of a fire, it is determined that a fire has occurred.

[0015] Preferably, the data fusion processing module is configured with fire feature thresholds, including temperature threshold, CO concentration threshold and visibility threshold. When sensor data exceeds the threshold, a higher fire occurrence confidence value and a lower non-fire occurrence confidence value are assigned to the basic probability assignment function corresponding to the sensor; conversely, a lower fire occurrence confidence value and a higher non-fire occurrence confidence value are assigned.

[0016] Preferably, the formula for calculating the comprehensive trust level mass(A) is as follows:

[0017]

[0018] In the formula, the conflict factor K is calculated as follows:

[0019]

[0020] In the formula, , , The basic probability assignments are for temperature, CO concentration, and visibility, respectively.

[0021] Preferably, the fuzzy PID control unit uses the deviation between the target wind speed and the actual wind speed and the rate of change of the deviation as two input variables, and the PID parameter increments Δkp, Δki, and Δkd as three output variables. Through the fuzzy inference rule base and the centroid method defuzzification algorithm, it outputs the corrected PID parameters in real time and updates the proportional coefficient, integral coefficient, and derivative coefficient of the PID controller, thereby calculating the real-time air volume required to maintain the target wind speed.

[0022] Preferably, the hysteresis control unit is configured with the rated air volume and hysteresis width of a single fan, and calculates the theoretical number of fans to be turned on based on the real-time air demand. When the real-time air demand increases and exceeds the upper limit of the air volume corresponding to the current number of fans to be turned on, the number of fans to be turned on is increased. When the real-time air demand decreases and is lower than the lower limit of the air volume corresponding to the current number of fans to be turned on, the number of fans to be turned on is decreased.

[0023] Preferably, the tunnel ventilation and smoke exhaust linkage control module further includes a fan drive unit, which is equipped with the following electrical protection logic:

[0024] The system monitors the operating current and voltage of each fan in real time. When the current exceeds the overcurrent threshold or the voltage exceeds the overvoltage threshold, the power supply circuit of the corresponding fan is cut off, and the fault status is fed back to the remote monitoring and control management system module.

[0025] Preferably, the remote monitoring and control management system module is configured with a three-level alarm mechanism based on comprehensive trust level:

[0026] When the overall trust level exceeds the first threshold, a warning alarm is triggered to alert of potential fire risks.

[0027] When the overall trust level exceeds the second threshold, a control alarm is triggered, and the ventilation control program is started.

[0028] When the overall trust level exceeds the third threshold, an emergency alarm is triggered, and alarm information is sent to the management personnel terminal.

[0029] Compared with the prior art, the present invention has the following advantages:

[0030] The present invention provides a tunnel fire monitoring and ventilation smoke exhaust linkage control system. Through the deep integration of multi-source data fusion and adaptive linkage control, it realizes intelligent process from detection to handling of tunnel fires. It not only solves the problem of high false alarm rate of traditional monitoring, but also overcomes the engineering problem of continuous and accurate control of discrete ventilation equipment, and significantly improves the tunnel's disaster prevention and mitigation capabilities.

[0031] This invention filters out high-frequency noise and bad data caused by the harsh tunnel environment at the data source by embedding a moving average filter and a 3σ outlier removal algorithm in the acquisition terminal. It also uses DS evidence theory to fuse multi-dimensional features such as temperature, CO concentration and visibility to effectively avoid the false alarm problem of a single sensor. By dynamically allocating BPA (basic probability allocation) and calculating conflict factors, it can make a correct fire judgment even when there are local conflicts in the sensor data.

[0032] The fuzzy PID algorithm used in this invention can adjust the control parameters in real time according to the wind speed deviation at different stages of fire development, adapting to the nonlinear characteristics of fire airflow. The hysteresis control unit successfully establishes a buffer mapping between the continuous air demand and the number of discrete fans, avoiding frequent start-stop of fans at critical wind speeds, extending the service life of jet fans, and maintaining stable tunnel wind speed. Attached Figure Description

[0033] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0034] Figure 1 This is an architecture diagram of a tunnel fire monitoring and ventilation smoke exhaust linkage control system according to the present invention;

[0035] Figure 2 This is a schematic diagram of the sensor cross-sectional arrangement of the tunnel environment and fire monitoring module in this invention;

[0036] Figure 3 This is a schematic diagram of the longitudinal arrangement of the monitoring and smoke exhaust equipment working in synergy in this invention;

[0037] Figure 4 This is a logic flowchart of the linkage control between tunnel fire monitoring and ventilation and smoke exhaust in this invention;

[0038] Figure 5 This is a control principle diagram of the fuzzy PID control unit in this invention. Detailed Implementation

[0039] The technical solution of the present invention will be further described in detail below through specific embodiments and in conjunction with the accompanying drawings. It should be understood that the implementation of the present invention is not limited to the following embodiments, and any modifications and / or alterations made to the present invention will fall within the protection scope of the present invention.

[0040] In this invention, unless otherwise specified, all parts and percentages are by weight, and the equipment and raw materials used are commercially available or commonly used in the art. Unless otherwise specified, the methods in the following embodiments are conventional methods in the art. Unless otherwise specified, the components or equipment in the following embodiments are general standard parts or components known to those skilled in the art, and their structures and principles can be learned by those skilled in the art through technical manuals or conventional experimental methods.

[0041] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings. In this detailed description, numerous specific details are set forth to facilitate explanation and provide a thorough understanding of the embodiments of the present invention. However, one or more embodiments may be practiced by those skilled in the art without these specific details.

[0042] like Figure 1 As shown, an embodiment of the present invention discloses a tunnel fire monitoring and ventilation smoke exhaust linkage control system, which includes four core modules: a tunnel environment and fire monitoring module, a data fusion processing module, a tunnel ventilation smoke exhaust linkage control module, and a remote monitoring and control management system module.

[0043] like Figure 2-3As shown, the tunnel environment and fire monitoring module includes a data acquisition terminal and sensor groups arranged along the tunnel's longitudinal direction and at the tunnel arch. Sensors are placed every 10 meters at the tunnel arch and at a height of 2 meters above the ground. Among these, the temperature sensor is a thermocouple-type temperature sensor with a measurement range of -40 to 200℃ and an accuracy of ±0.5℃; the CO concentration sensor is an electrochemical CO concentration sensor with a measurement range of 0 to 500 ppm and an accuracy of ±5 ppm; and the visibility sensor is a diffused light visibility sensor with a measurement range of 5 to 500 meters. In addition, ultrasonic or hot-wire anemometers with a measurement range of 0 to 30 m / s are arranged in pairs at the tunnel entrances / exits and cross passages to monitor longitudinal wind speed.

[0044] The data acquisition terminal uses an industrial-grade microprocessor (MCU) as its core and has built-in signal conditioning circuitry. The terminal connects to each sensor node via a Zigbee wireless sensor network and communicates with upper-layer modules via industrial Ethernet.

[0045] The data fusion processing module is deployed in the industrial computer of the tunnel management station. It runs the DS evidence theory algorithm program and is responsible for processing multi-source heterogeneous data and outputting fire determination results.

[0046] The tunnel ventilation and smoke extraction linkage control module is implemented using a PLC. This module has a built-in fuzzy PID control program and hysteresis control logic, and outputs control signals to the fan drive unit. The fan drive unit consists of a relay control cabinet, which controls the start and stop of multiple jet fans in the tunnel.

[0047] The remote monitoring and control management system module includes an industrial touch screen monitoring terminal and a database server, which are used to display environmental parameter curves, fire alarm status and the number of operating fans in real time, and record historical events.

[0048] To ensure data quality, this system executes the following data processing flow.

[0049] The first step is data preprocessing logic, where the data acquisition terminal cleans the raw data sampled by the sensors. First, a moving average filtering algorithm is used, with a sliding window size of 10, to calculate the arithmetic mean of the data within the window to filter out random noise. Second, the 3σ criterion (Laida criterion) is applied to calculate the mean and standard deviation of the data. If the deviation of a sampling point's data from the mean exceeds three times the standard deviation, it is judged as interference or faulty data and discarded, with only valid data being uploaded.

[0050] Secondly, there is the DS evidence theory fusion logic. The data fusion processing module constructs an identification framework U = {fire occurred (D), no fire occurred (¬D)}. The preset fire characteristic thresholds are: temperature > 50°C, CO concentration > 20ppm, and visibility < 50m.

[0051] When sensor data exceeds the aforementioned thresholds, the system automatically adjusts the Basic Probability Allocation (BPA). For example, when the temperature exceeds the limit, the system sets mass1(D) = 0.8 and mass1(¬D) = 0.2 for the temperature sensor; if the temperature does not exceed the limit, then mass1(D) = 0.2 and mass1(¬D) = 0.8. Similarly, the BPA is set for CO concentration (mass2) and visibility (mass3).

[0052] Calculate the overall trust level using Dempster's combination rule formula:

[0053]

[0054] In the formula, the conflict factor K, which reflects the degree of conflict between pieces of evidence, is calculated as follows:

[0055]

[0056] In the formula, , , The basic probability assignments are for temperature, CO concentration, and visibility, respectively.

[0057] The system determines that a fire has occurred when the calculated mass(D) is significantly higher than mass(¬D).

[0058] Once a fire is detected, the tunnel ventilation and smoke exhaust linkage control module is activated and executes the following control steps.

[0059] One approach is fuzzy PID control. The system sets a target wind speed based on the fire scale; for example, the critical wind speed for preventing backflow of smoke is 3 m / s. The deviation *e* between the target wind speed and the actual wind speed, and the rate of change of this deviation *ec*, are calculated. *e* and *ec* are mapped to the universe of discourse [-6, 6] and divided into seven fuzzy levels: NB (negative maximum), NM, NS, ZO, PS, PM, and PB. Based on the fuzzy rule base, the fuzziness is resolved using the centroid method, outputting the PID parameter increments Δkp, Δki, and Δkd. The PID parameters are then updated, and the total required air volume for the system at the current moment is calculated.

[0060] Secondly, there is hysteresis control mapping. The system needs to convert the continuous total air demand Q into a discrete number of fans to be activated. Setting the hysteresis width ΔQ = 5% Q, and the rated air volume of a single fan being 29.4 m³ / s, the formula for calculating the number of fans to be activated is as follows: ( (This is a rounding function). When the air demand fluctuation exceeds the hysteresis range, the N value is adjusted (±1 units). When Q > Qt + ΔQ, one fan is added; when Q < Qt - ΔQ, one fan is removed. This logic effectively prevents frequent fan start-stop when the air demand fluctuates slightly around the rated air capacity of a certain fan.

[0061] Thirdly, there is electrical protection. The fan drive unit monitors the current and voltage in real time. If it detects that the operating current of a fan exceeds 10A (overcurrent) or the voltage exceeds 250V (overvoltage), it cuts off the circuit, reports an error to the remote system, and simultaneously controls and dispatches other standby fans to supplement the air volume.

[0062] The remote monitoring and control management system module is equipped with a three-level alarm mechanism based on comprehensive trust level:

[0063] When the overall trust level exceeds the first threshold of 0.5, a warning alarm is triggered to indicate a potential fire risk.

[0064] When the overall trust level exceeds the second threshold of 0.7, a control alarm is triggered, and the ventilation control program is started.

[0065] When the overall trust level exceeds the third threshold of 0.9, an emergency alarm is triggered, and alarm information is sent to the management personnel terminal.

[0066] Embodiments of the present invention also disclose a method for linkage control of tunnel fire monitoring and ventilation / smoke extraction, such as... Figure 4 As shown, the method includes the following steps:

[0067] Step S1: System Initialization and Data Acquisition

[0068] The system powers on and resets, and each sensor node completes a self-test. The data acquisition terminal collects data on temperature, CO concentration, visibility, and longitudinal wind speed at the tunnel entrance at 100ms intervals. During the acquisition process, the terminal processor executes a moving average filter (window size N=10) and a 3σ outlier removal procedure in real time to generate effective environmental feature vectors.

[0069] Step S2: Multi-source evidence fusion determination

[0070] The data fusion processing module receives the feature vector and first maps it to a basic probability assignment (BPA).

[0071] If the temperature T>50℃, then m(D) is assigned a value of 0.8; if the CO concentration C>20ppm, then m(D) is assigned a value of 0.75; if the visibility V<50m, then m(D) is assigned a value of 0.8.

[0072] The Dempster combination rule is used to orthogonally fuse the above BPA to calculate the comprehensive confidence level Bel(D). If Bel(D) > 0.7 and the duration exceeds 3 seconds, a fire is determined to have occurred, and the process proceeds to step S3; otherwise, the process returns to step S1 to continue monitoring.

[0073] Step S3: Goal Setting and Fuzzy Inference

[0074] Once a fire is confirmed, the system locks the location of the fire source and sets the target wind speed.

[0075] As Figure 5 shown, the controller calculates the wind speed deviation e(k) and the deviation change rate ec(k) at the current moment, inputs e and ec into the fuzzy controller, maps them to the fuzzy domain through fuzzy processing, queries the pre-set fuzzy rule table, and then performs defuzzification by the centroid method to output the correction amount of the PID parameters.

[0076] Step S4: Calculation of required air volume and PID update

[0077] Utilize the corrected PID parameters for PID operation to calculate the current total required air volume of the system. This step ensures that the control quantity can be adaptively adjusted according to the non-linear characteristics of the fire development.

[0078] Step S5: Hysteresis quantization and actuator action

[0079] Input the continuously changing required air volume into the hysteresis control unit. Read the current number of fans Nc in operation. Calculate the air volume interval [Ql, Qu] corresponding to Nc fans in operation, where Qu = Nc × qu + ΔQ and Ql = Nc × qu - ΔQ.

[0080] If Qr > Qu, output an instruction to increase the number of fans and start the next standby fan; if Qr < Ql, output an instruction to reduce the number of fans and shut down one operating fan; if Qr is within the interval, keep the current number of fans unchanged. Qr represents the real-time required air volume calculated by the system, that is, the total air volume value required to maintain the tunnel target wind speed calculated by the fuzzy PID control unit based on the current wind speed deviation. qu represents the rated air volume of a single fan, that is, the theoretical air supply capacity of a single jet fan under the standard working voltage. ΔQ represents the hysteresis width.

[0081] The fan drive unit receives the instruction and acts, and at the same time, it real-time feedbacks the current and voltage states for closed-loop control.

[0082] In a specific application example, taking a double-hole extra-long tunnel as the application object, the left line of the tunnel is 12 km long in total, the right line is 11.8 km long in total, the cross-sectional area of a single hole is 65.65 m 2 , the spacing of the transverse channels is 500 m, and 30 jet fans are configured (the air volume of a single fan is 29.4 m 3 / s), and the system deployment and operation effects are as follows.

[0083] System deployment parameters:

[0084] Multi-source monitoring module: 1000 sensor sets are installed every 10m along the tunnel arch and at a height of 2m; 10 sensor sets are installed every 5m in the cross passages, for a total of 240 sets. Wind speed sensors are installed at the tunnel entrance, exit, and both ends of the cross passages, with 2 sensors each. Two data acquisition terminals are installed (one primary and one backup, spaced 500m apart).

[0085] Data fusion and fire identification module: An industrial computer (IPC-610) is installed at the tunnel management station, running DS evidence theory calculation software compiled based on MATLAB R2020b.

[0086] Adaptive linkage control module: The fuzzy PID controller adopts an S7-1200 PLC (CPU 1214C), the hysteresis control program is integrated into the PLC user program in the form of a ladder diagram, and the fan drive unit adopts a JR-2000 relay cabinet.

[0087] Communication network: 150 Zigbee nodes, 4 industrial Ethernet switches (1 each for the inlet, outlet, middle section, and management station), and a total communication cable length of 30km.

[0088] Required air volume calculation: Target wind speed v = 1 m / s, cross-sectional area A = 65.65 m² 2 The required air volume is Q = 65.65 × 1 = 65.65 m³. 3 / s;

[0089] Number of fans in operation: N = ceil(65.65 / 29.4) = 3 units (initially), and after stabilization, maintain a switching frequency of 6 to 7 units (due to hysteresis control);

[0090] Response time: From the occurrence of the fire to the completion of the fan startup, it took 9.02s (fire identification time) + 1.8s (data transmission time) = 10.82s;

[0091] Wind speed stabilization time: The stabilization time from 0 to 1 m / s is 8.5 s, with an overshoot of 2.8%.

[0092] The control effect is as follows.

[0093] Smoke spread range: controlled within 270m (<300m safety threshold);

[0094] Tunnel vault temperature: Maximum temperature 142℃ (<150℃ safety threshold);

[0095] Human eye temperature: Maximum temperature 58℃ (<60℃ safety threshold);

[0096] High-level CO concentration in the human eye: Maximum concentration 142 ppm (<150 ppm safety threshold);

[0097] Human eye height visibility: minimum visibility 12m (>10m safety threshold).

[0098] As can be seen, this application example verifies that the system can effectively achieve accurate identification of tunnel fires and intelligent control of ventilation and smoke extraction, meet the safety needs of personnel evacuation and fire rescue, and is applicable to fire prevention and control scenarios in various highway and railway tunnels.

[0099] This article uses specific examples to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. It should be noted that those skilled in the art can make several improvements and modifications to the present invention without departing from the principles of the present invention, and these improvements and modifications also fall within the protection scope of the claims of the present invention.

Claims

1. A tunnel fire monitoring and ventilation / smoke exhaust linkage control system, characterized in that: The tunnel fire monitoring and ventilation smoke exhaust linkage control system includes a tunnel environment and fire monitoring module, a data fusion processing module, a tunnel ventilation smoke exhaust linkage control module, and a remote monitoring and control management system module, all connected via a communication network. The tunnel environment and fire monitoring module is used to collect environmental parameters and fire characteristic data in the tunnel in real time, and transmit them to the data fusion processing module through the communication network. The data fusion processing module is equipped with the DS evidence theory algorithm, which is used to receive multi-source environmental data, calculate the basic probability distribution of each sensor data, quantify and evaluate the probability of fire occurrence through combination rules, and generate fire status determination results. The tunnel ventilation and smoke exhaust linkage control module is used to receive the fire status judgment result and determine the target wind speed according to the fire scale. The tunnel ventilation and smoke exhaust linkage control module has a built-in fuzzy PID control unit and a hysteresis control unit. The fuzzy PID control unit dynamically adjusts the PID parameters according to the deviation between the target wind speed and the actual wind speed to calculate the total air volume required by the system. The hysteresis control unit maps the continuously changing air volume to the discrete number of jet fans to be turned on, and uses hysteresis logic to prevent the fans from frequently starting and stopping at the critical point. The remote monitoring and control management system module is used to monitor the tunnel environment, fire status, and equipment operation in real time, and provides data storage, historical query, and hierarchical alarm functions.

2. The tunnel fire monitoring and ventilation smoke extraction linkage control system according to claim 1, characterized in that: The tunnel environment and fire monitoring module includes a data acquisition terminal, temperature sensors, CO concentration sensors, visibility sensors arranged along the longitudinal direction of the tunnel and the tunnel arch, and wind speed sensors arranged at the tunnel entrances and cross passages. The temperature sensors, CO concentration sensors, and visibility sensors are distributed at preset intervals at the tunnel arch, and the wind speed sensors are arranged in pairs at the tunnel entrance and exit to monitor the longitudinal wind speed of the tunnel. The temperature sensors, CO concentration sensors, visibility sensors, and wind speed sensors are all electrically connected to the data acquisition terminal.

3. The tunnel fire monitoring and ventilation smoke extraction linkage control system according to claim 2, characterized in that: The data acquisition terminal has a built-in signal conditioning circuit and a microprocessor, and is configured with the following data preprocessing logic: The original sampled data is smoothed using a moving average filtering algorithm, and the size of the moving window is set to reduce high-frequency noise interference from the environment. The 3σ criterion is used to detect outliers in the filtered data. Data that exceeds the standard deviation of the mean by a multiple of the set value is identified as sensor malfunction or sudden interference and is discarded. Only valid data is transmitted through the communication network.

4. The tunnel fire monitoring and ventilation smoke extraction linkage control system according to claim 1, characterized in that: The data fusion processing module constructs an identification framework U={fire occurred, no fire occurred} based on the Dempster evidence theory. It uses the Dempster combination rule to perform orthogonal and normalized fusion calculations on the basic probability assignments of temperature, CO concentration, and visibility to obtain a comprehensive confidence level. When the comprehensive confidence level supporting the occurrence of a fire is significantly higher than the comprehensive confidence level supporting the absence of a fire, it is determined that a fire has occurred.

5. The tunnel fire monitoring and ventilation smoke extraction linkage control system according to claim 4, characterized in that: The data fusion processing module is configured with fire feature thresholds, including temperature threshold, CO concentration threshold, and visibility threshold. When sensor data exceeds the threshold, a higher fire occurrence confidence value and a lower fire non-occurrence confidence value are assigned to the basic probability assignment function corresponding to the sensor; conversely, a lower fire occurrence confidence value and a higher fire non-occurrence confidence value are assigned.

6. The tunnel fire monitoring and ventilation smoke extraction linkage control system according to claim 4, characterized in that: The formula for calculating the overall trust level mass(A) is as follows: In the formula, the conflict factor K is calculated as follows: In the formula, , , The basic probability assignments are for temperature, CO concentration, and visibility, respectively.

7. The tunnel fire monitoring and ventilation smoke extraction linkage control system according to claim 1, characterized in that: The fuzzy PID control unit uses the deviation between the target wind speed and the actual wind speed and the rate of change of the deviation as two input variables, and the PID parameter increments Δkp, Δki, and Δkd as three output variables. Through the fuzzy inference rule base and the centroid method defuzzification algorithm, it outputs the corrected PID parameters in real time and updates the proportional coefficient, integral coefficient, and derivative coefficient of the PID controller, thereby calculating the real-time air volume required to maintain the target wind speed.

8. A tunnel fire monitoring and ventilation / smoke extraction linkage control system according to claim 7, characterized in that: The hysteresis control unit is equipped with the rated air volume and hysteresis width of a single fan. It calculates the theoretical number of fans to be turned on based on the real-time air demand. When the real-time air demand increases and exceeds the upper limit of the air volume corresponding to the current number of fans to be turned on, the number of fans to be turned on is increased. When the real-time air demand decreases and is lower than the lower limit of the air volume corresponding to the current number of fans to be turned on, the number of fans to be turned on is decreased.

9. The tunnel fire monitoring and ventilation smoke extraction linkage control system according to claim 1, characterized in that: The tunnel ventilation and smoke exhaust linkage control module also includes a fan drive unit, which is equipped with the following electrical protection logic: The system monitors the operating current and voltage of each fan in real time. When the current exceeds the overcurrent threshold or the voltage exceeds the overvoltage threshold, the power supply circuit of the corresponding fan is cut off, and the fault status is fed back to the remote monitoring and control management system module.

10. A tunnel fire monitoring and ventilation / smoke extraction linkage control system according to claim 1, characterized in that: The remote monitoring and control management system module is equipped with a three-level alarm mechanism based on comprehensive trust level: When the overall trust level exceeds the first threshold, a warning alarm is triggered to alert of potential fire risks. When the overall trust level exceeds the second threshold, a control alarm is triggered, and the ventilation control program is started. When the overall trust level exceeds the third threshold, an emergency alarm is triggered, and alarm information is sent to the management personnel terminal.

Images