Early warning and control devices and methods for preventing curtain wall spontaneous explosion

Abstract

This application discloses an early warning control device and method for preventing spontaneous explosion of curtain walls, relating to the field of building curtain wall safety monitoring technology. The device includes a sensing module, a data acquisition and communication module, a risk prediction module, an early warning control module, and an active control device. The sensing module includes a temperature sensor array and strain gauges, which measure the temperature difference between the indoor and outdoor surfaces of the glass and the surface micro-strain, respectively. The risk prediction module calculates the probability of glass spontaneous explosion based on real-time temperature difference and strain data. The early warning control module automatically triggers the active control device when the risk probability exceeds a preset safety threshold. The active control device includes a spray system that sprays water to cool the outer surface of the glass. A thermo-coupled risk prediction model is constructed through dual-parameter synchronous monitoring and fusion analysis. Automatic triggering of spray cooling at the risk critical point achieves closed-loop control from passive alarm to active risk elimination, effectively preventing spontaneous explosion accidents of glass curtain walls.

Description

Technical Field

[0001] This invention relates to the field of building curtain wall safety monitoring technology, specifically to an early warning control device and method for preventing curtain wall spontaneous explosion. Background Technology

[0002] Glass curtain walls, as an important external envelope structure of modern buildings, are widely used in large public buildings, high-rise buildings, and commercial complexes. Glass curtain walls not only offer excellent lighting performance and aesthetic appeal but also effectively insulate against heat and sound, enhancing the overall performance of the building. However, glass curtain walls, especially tempered glass, pose a risk of spontaneous breakage, a problem that has always been a significant concern in the field of building safety. Spontaneous glass breakage is mainly caused by the combined effects of internal impurities and external stresses (such as thermal stress and installation stress). Once a spontaneous breakage occurs, it not only causes economic losses but may also endanger pedestrian safety.

[0003] While traditional manual inspection methods can detect some obvious structural defects, they rely on regular manual inspections and cannot detect in real time the temperature difference between the inner and outer surfaces of the glass, a key factor leading to a surge in thermal stress. Especially under high temperatures and sunlight in summer, the temperature of the sun-facing side of the glass can rise sharply in a short period of time. Since manual inspections are usually conducted only 2-3 times a day, it is difficult to capture such instantaneous temperature changes, thus missing the best opportunity for risk intervention. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide an early warning control device and method for preventing curtain wall spontaneous explosion, so as to realize the proactive prevention and control of the risk of curtain wall spontaneous explosion.

[0005] The technical solution adopted by this invention to solve its technical problem is an early warning and control device for preventing curtain wall spontaneous explosion, including a sensing module. The sensing module includes a temperature sensor array for measuring the temperature difference between the indoor and outdoor surfaces of the curtain wall and a strain gauge for measuring the micro-strain of the curtain wall surface. The sensing module is communicatively connected to a data acquisition and communication module, which receives the temperature difference data and strain data collected by the sensing module and transmits them to the risk prediction module in real time. The risk prediction module is connected to the data acquisition and communication module to calculate the probability of spontaneous explosion of the curtain wall based on the received temperature difference data and strain data. The early warning control module communicates with the risk prediction module to issue an alarm when the probability of spontaneous detonation exceeds a safety threshold and to automatically trigger the active control device. The active control device is communicatively connected to the early warning control module. The active control device includes a sprinkler system installed on the curtain wall for spraying water to cool the outer surface of the curtain wall.

[0006] Furthermore, the risk prediction model built into the risk prediction module is trained using machine learning algorithms, and its input features include at least the real-time internal and external temperature difference, the rate of change of temperature difference, the real-time strain value, and the rate of change of strain.

[0007] Furthermore, the machine learning algorithm is either an LSTM neural network or a gradient boosting decision tree algorithm.

[0008] Furthermore, the early warning control module supports multi-level early warning. When the temperature difference or strain exceeds the initial threshold but the probability value of spontaneous explosion is lower than the warning value, an attention message is sent. When the probability value exceeds the warning value but is lower than the safety threshold, an audible and visual alarm is issued. When the probability value exceeds the safety threshold, an audible and visual alarm is issued and the sprinkler system is automatically triggered.

[0009] Furthermore, the sprinkler system includes a water storage mechanism, a water pipe, and an atomizing nozzle. The atomizing nozzle is arranged at the top of the curtain wall. One end of the water pipe is connected to the water storage mechanism, and the other end is connected to the atomizing nozzle. A solenoid valve is installed on the water pipe.

[0010] Furthermore, the strain gauge is a fiber optic strain sensor.

[0011] Furthermore, temperature sensor arrays and strain gauges are deployed at the edges and center of the curtain wall.

[0012] Furthermore, it also includes a data storage module, which is communicatively connected to the sensing module, the early warning control module, and the active control device. This module is used to record monitoring data, early warning records, and control actions, and upload them to a cloud database.

[0013] The early warning and control method for preventing curtain wall spontaneous explosion includes the following steps:

[0014] Step S1: Real-time acquisition of indoor and outdoor surface temperatures of the curtain wall using a temperature sensor array, and calculation of the indoor and outdoor temperature difference ΔT; real-time acquisition of micro-strain ε on the curtain wall surface using strain gauges;

[0015] Step S2: Input the real-time collected temperature difference ΔT and strain ε into the risk prediction model to calculate the probability of spontaneous explosion of the curtain wall, Prisk.

[0016] Step S3: Determine whether the risk probability Prisk exceeds the safety threshold. If it does, proceed to step S4; otherwise, return to step S1 to continue monitoring.

[0017] Step S4: Issue an alarm message and automatically trigger the sprinkler system to spray water to cool the exterior surface of the curtain wall;

[0018] Step S5: Monitor the changes in temperature difference ΔT and strain ε during the cooling process, and stop spraying when the risk probability Prisk drops below the safety threshold.

[0019] Furthermore, in step S5, when the risk probability Prisk drops below the safety threshold, and when the temperature difference ΔT decreases to below 30% of the initial value and the strain ε falls back, it is determined that the risk has been eliminated.

[0020] The beneficial effects of this invention are as follows: By integrating temperature difference and strain, two key parameters directly leading to spontaneous breakage of glass curtain walls, this invention constructs a thermo-mechanical coupled risk prediction model, which can accurately assess the comprehensive risk status resulting from the superposition of thermal stress and installation stress. Simultaneously, the introduction of a spray system automatically triggers the system to cool the outer surface of the glass when the risk reaches a critical threshold, rapidly reducing the internal and external temperature difference, alleviating thermal stress at its source, and effectively curbing spontaneous breakage. Attached Figure Description

[0021] Figure 1 This is a schematic diagram of the structure of the present invention; Figure 2 This is a schematic diagram of the installation of a sprinkler system and sensor modules on the curtain wall.

[0022] Reference numerals in the attached figures: 1-Sensing module; 101-Temperature sensor array; 102-Strain gauge; 2-Data acquisition and communication module; 3-Risk prediction module; 4-Early warning and control module; 5-Active control device; 501-Water storage mechanism; 502-Water pipe; 503-Atomizing nozzle; 504-Solenoid valve; 6-Curtain wall; 7-Data storage module; 8-Cloud database. Detailed Implementation

[0023] Embodiments of the present invention are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention.

[0024] like Figure 1 and Figure 2 As shown, the early warning and control device for preventing curtain wall spontaneous explosion of the present invention includes a sensing module 1. The sensing module 1 includes a temperature sensor array 101 for measuring the temperature difference between the indoor and outdoor surfaces of the curtain wall 6 and a strain gauge 102 for measuring the micro-strain on the surface of the curtain wall 6. The sensing module 1 is communicatively connected to a data acquisition and communication module 2. The data acquisition and communication module 2 is used to receive the temperature difference data and strain data collected by the sensing module 1 and transmit them to the risk prediction module 3 in real time. The risk prediction module 3 is connected to the data acquisition and communication module 2 and is used to calculate the probability of spontaneous explosion of the curtain wall 6 based on the received real-time temperature difference data and strain data. The early warning control module 4 is communicatively connected to the risk prediction module 3, and is used to issue an alarm message and automatically trigger the active control device 5 when the probability of spontaneous detonation exceeds the safety threshold; The active control device 5 is communicatively connected to the early warning control module 4. The active control device 5 includes a spray system installed on the curtain wall 6, which is used to spray water to cool the outer surface of the curtain wall 6.

[0025] The sensing module 1 includes a temperature sensor array 101 and strain gauges 102. The temperature sensor array 101 consists of multiple temperature sensors, respectively deployed on the indoor and outdoor surfaces of the curtain wall 6. One set of temperature sensors is deployed at the edge and another at the center of each curtain wall 6. The edge sensors monitor temperature changes in stress concentration areas, while the center sensors monitor the overall temperature distribution of the curtain wall 6. The temperature sensor array 101 calculates the internal and external temperature difference ΔT by measuring the indoor and outdoor surface temperatures, providing key parameters for thermal stress assessment. The strain gauges 102 are fiber Bragg grating strain sensors, deployed at the edges and center of the curtain wall 6, corresponding to the temperature sensor positions. The edge strain gauges 102 monitor the strain state in edge stress concentration areas, while the center strain gauges 102 monitor the overall strain distribution of the curtain wall 6. Fiber Bragg grating strain sensors have advantages such as strong anti-electromagnetic interference capability, high measurement accuracy, and good long-term stability. The measurement accuracy can reach 1 micro-strain, with a sampling frequency of 10 times / second, enabling real-time measurement of the micro-strain value ε on the surface of the curtain wall 6, reflecting the stress state of the curtain wall 6.

[0026] The data acquisition and communication module 2 includes a data acquisition module and a communication module. The data acquisition and communication module 2 receives temperature difference and strain data collected by the sensing module 1 via wired or wireless means and transmits it to the risk prediction module 3 in real time. The data acquisition module uses a high-precision analog-to-digital converter to convert the analog signals output by the temperature sensor and strain gauge 102 into digital signals with a conversion accuracy of 16 bits, ensuring the accuracy of data acquisition. The communication module uses wireless transmission, employing the Zigbee or LoRa protocol, which can cover the monitoring range of the large building curtain wall 6. The transmission frequency is synchronized with the sensor sampling frequency, ensuring the real-time performance of the data.

[0027] The risk prediction module 3 receives real-time temperature difference ΔT and strain ε data transmitted from the data acquisition and communication module 2, and calculates the spontaneous explosion risk probability Prisk of the curtain wall 6 using its built-in risk prediction model. The risk prediction model is trained using a machine learning algorithm, specifically an LSTM neural network algorithm. This algorithm excels at processing time-series data, capturing the dynamic characteristics of temperature difference and strain changes over time, and accurately predicting the evolution trend of the risk probability. The input features of the risk prediction model include four parameters: real-time internal and external temperature difference ΔT, temperature difference change rate dΔT / dt, real-time strain value ε, and strain change rate dε / dt. The temperature difference change rate reflects the speed of temperature change, and the strain change rate reflects the speed of stress accumulation. The combination of these two parameters enables a forward-looking prediction of the risk development trend. The risk prediction model is trained based on historical data, including stress distribution data of the curtain wall 6 under different temperature difference and strain conditions, as well as actual spontaneous explosion case data. By learning the complex nonlinear relationship between temperature difference, strain, and stress distribution of the curtain wall 6, a thermo-mechanical coupled risk prediction model is established. The model outputs a risk probability Prisk ranging from 0 to 1; the closer Prisk is to 1, the higher the spontaneous explosion risk. The risk prediction module 3 updates the risk probability calculation results every minute to assess the safety status of the curtain wall 6 in real time.

[0028] The early warning control module 4, based on the risk probability Prisk calculated by the risk prediction module 3, executes a multi-level early warning strategy and automatically triggers the active control device 5. The early warning strategy has three levels: when the temperature difference ΔT or strain ε exceeds the initial threshold (ΔT>10℃ or ε>100 micro-strain) but the risk probability Prisk is lower than the early warning value of 0.50, it is determined to be a low-risk state. A warning message is sent to the monitoring center, prompting maintenance personnel to pay attention to the status of the curtain wall 6 in that area, but the active control device 5 is not triggered. When the risk probability Prisk is greater than or equal to the early warning value of 0.50 but less than the safety threshold of 0.75, it is determined to be a medium-risk state. An audible and visual early warning signal is issued, prompting maintenance personnel to strengthen inspections. When the risk probability Prisk is greater than or equal to the safety threshold of 0.75, it is determined to be a high-risk state. In addition to issuing an audible and visual alarm, the active control device 5 is automatically triggered to start the sprinkler system for cooling intervention. The multi-level early warning strategy adopts differentiated response measures according to the degree of risk, avoiding frequent high-level alarms that waste resources, while ensuring timely active control measures are taken at risk thresholds. The warning value and safety threshold are set based on historical spontaneous explosion case statistics and model verification. When Prisk exceeds 0.75, the probability of spontaneous explosion of curtain wall 6 increases significantly, requiring immediate proactive intervention.

[0029] See Figure 2The active control device 5 includes a spray system controlled by the early warning control module 4, used to spray water to cool the outer surface of the curtain wall 6. The spray system includes a water storage mechanism 501, a water pipe 502, and atomizing nozzles 503. The atomizing nozzles 503 are arranged at the top of the curtain wall 6. One end of the water pipe 502 is connected to the water storage mechanism 501, and the other end is connected to the atomizing nozzles 503. A solenoid valve 504 is installed on the water pipe 502. The atomizing nozzles 503 are evenly distributed horizontally along the top of the curtain wall 6, with a spacing of 2 meters, ensuring complete coverage of the outer surface of the curtain wall 6. The water spraying method using the atomizing nozzles 503 can quickly remove heat from the surface of the curtain wall 6 while avoiding additional stress caused by water flow impact. The spraying action is controlled by the solenoid valve 504, which has a response time of less than 1 second, enabling it to quickly open or close the spray after receiving a control command, achieving precise control. The water storage mechanism 501 can be a building fire water pipe, utilizing existing facilities to reduce renovation costs. The fire water pipe pressure is 0.3-0.5MPa, meeting the pressure requirements of the sprinkler system. The water storage mechanism 501 can also be a water tank, using a water pump to supply water to the water pipe 502. The working principle of the sprinkler system is that when the early warning control module 4 triggers the control command, the solenoid valve 504 opens, and the water in the water storage mechanism 501 is atomized by the atomizing nozzle 503 and sprayed evenly on the outer surface of the curtain wall 6. The temperature of the outer surface of the curtain wall 6 is quickly reduced by the heat absorption of water evaporation, thereby reducing the internal and external temperature difference ΔT, alleviating thermal stress, and bringing the risk probability Prisk back to the safe range.

[0030] Furthermore, the temperature sensor array 101 and strain gauge 102 are deployed at the edges and center of the curtain wall 6. The placement of the temperature sensor array 101 and strain gauge 102 is based on the stress distribution characteristics of the curtain wall 6. The edge positions are stress concentration areas and also high-risk areas for spontaneous explosion, while the center positions can reflect the overall state of the curtain wall 6. The combination of the two can comprehensively grasp the temperature and strain distribution of the curtain wall 6, providing an accurate data basis for risk prediction.

[0031] Furthermore, it also includes a data storage module 7, which is communicatively connected to the sensing module 1, the early warning control module 4, and the active control device 5. This module records monitoring data, early warning records, and control actions, and uploads them to the cloud database 8. The data storage module 7 records all monitoring data collected by the sensing module 1 (temperature difference ΔT, strain ε), the risk probability Prisk calculated by the risk prediction module 3, early warning records issued by the early warning control module 4, and the control actions of the active control device 5 (sprinkler start time, duration, and stop time), and uploads this data to the cloud database 8 via the internet. The cloud database 8 adopts a distributed storage architecture to ensure data security and reliability. The data storage module 7 enables data traceability throughout the entire operation of the device, providing a basis for post-event analysis. Through big data analysis, it can uncover the patterns and characteristics of spontaneous explosion of the curtain wall 6, continuously optimize the parameters of the risk prediction model, verify the effectiveness of active control measures, and achieve self-learning and continuous improvement of the device.

[0032] Example 1

[0033] Under high summer temperatures and strong sunlight, the device operated normally before 10:00 AM, with the indoor temperature maintained at 25°C, the outer surface temperature of the curtain wall at 28°C, the temperature difference between the inside and outside (ΔT) at 3°C, the strain (ε) at 50 microstrain, and the risk probability (Prisk) at 0.15, indicating a safe state. After 10:00 AM, as solar radiation intensity increased, the outer surface temperature of curtain wall 6 rose rapidly. The temperature sensor array 101 monitored in real time that the outdoor surface temperature rose from 28°C to 55°C, while the indoor surface temperature remained at 25°C, and the temperature difference (ΔT) increased from 3°C to 30°C. Simultaneously, strain gauge 102 monitored that the strain value (ε) continuously increased from 50 microstrain to 280 microstrain. The increase in strain was due to the thermal expansion of the outer surface of curtain wall 6, while the inner surface, being at a lower temperature, expanded less, creating a temperature gradient that generated thermal stress inside curtain wall 6. This thermal stress, combined with the inherent installation stress of curtain wall 6, significantly increased the overall strain level of curtain wall 6.

[0034] After receiving real-time data including temperature difference ΔT = 30℃, strain ε = 280 microstrain, temperature difference change rate dΔT / dt = 5℃ / 10min, and strain change rate dε / dt = 30 microstrain / 10min, the risk prediction module 3 performs calculations using an LSTM neural network model. The model first analyzes the current values ​​of temperature difference and strain, determining that the thermal stress level is approaching the bearing limit of curtain wall 6. Then, it analyzes the temperature difference change rate and strain change rate, predicting that the temperature difference and strain will continue to rise within the next 10 minutes. After comprehensive evaluation, the current risk probability Prisk = 0.85 is calculated. The risk probability Prisk = 0.85 exceeds the safety threshold of 0.75, indicating that curtain wall 6 is in a high-risk state. If not intervened in time, curtain wall 6 may spontaneously explode within a short period.

[0035] Upon receiving a signal with a risk probability Prisk=0.85, the early warning control module 4 immediately executes a high-risk response strategy: first, it issues an audible and visual alarm, with the alarm sounding and a red alarm message displayed on the monitoring screen, notifying maintenance personnel that the curtain wall 6 in that area is in a high-risk state; simultaneously, it automatically triggers the active control device 5, sending an opening command to the solenoid valve 504 of the sprinkler system. The solenoid valve 504 opens within 1 second, and water from the fire hose is atomized by the atomizing nozzle 503 and sprayed evenly onto the outer surface of the curtain wall 6. The atomized water droplets quickly cover the surface of the curtain wall 6 and evaporate, absorbing heat.

[0036] After spraying for 3 minutes, the temperature sensor array 101 monitored that the outer surface temperature of curtain wall 6 dropped from 55℃ to 35℃, with a cooling rate of approximately 7℃ / min. The internal and external temperature difference ΔT decreased from 30℃ to 10℃, a reduction of 67%. The reason for the decrease in temperature difference is that the sprayed water mist evaporated and absorbed heat from the surface of curtain wall 6. According to the evaporation heat absorption formula Q=mL, where m is the amount of water evaporated and L is the latent heat of vaporization of water (approximately 2260kJ / kg), the amount of water evaporated in 3 minutes of spraying is approximately 4.5L×0.3 (evaporation rate)=1.35kg, absorbing approximately 3051kJ of heat, which effectively reduces the surface temperature of curtain wall 6. As the temperature difference decreases, the internal thermal stress of curtain wall 6 decreases significantly. The strain gauge 102 monitored that the strain value ε dropped from 280 microstrain to 120 microstrain. The reason for the decrease in strain is that after the thermal stress decreases, the stress distribution inside curtain wall 6 tends to be more uniform, and the strain level decreases accordingly.

[0037] The risk prediction module 3 recalculates the risk probability based on the temperature difference ΔT = 10℃ after cooling and the strain ε = 120 microstrain, obtaining Prisk = 0.18. The risk probability drops from 0.85 to 0.18, a decrease of 79%. Since the risk probability Prisk = 0.18 is lower than the warning value of 0.5, it indicates that the curtain wall 6 has returned to a safe state, and the risk of spontaneous explosion has been basically eliminated. Upon receiving the signal Prisk = 0.18, the warning control module 4 determines that the risk has been eliminated and sends a closing command to the solenoid valve 504 of the sprinkler system. The solenoid valve 504 closes, and the sprinkler system stops. Simultaneously, a "Risk Eliminated" report is generated and pushed to the maintenance personnel's terminal. The report contains complete data for this warning: trigger time, peak temperature difference of 30℃, peak strain of 280 microstrain, peak risk probability of 0.85, sprinkler duration of 3 minutes, temperature difference after cooling of 10℃, strain after cooling of 120 microstrain, and final risk probability of 0.18.

[0038] The early warning and control method for preventing curtain wall spontaneous explosion includes the following steps:

[0039] Step S1: Real-time acquisition of indoor and outdoor surface temperatures of curtain wall 6 using temperature sensor array 101, and calculation of the indoor and outdoor temperature difference ΔT; real-time acquisition of micro-strain ε on the surface of curtain wall 6 using strain gauge 102;

[0040] Step S2: Input the real-time collected temperature difference ΔT and strain ε into the risk prediction model to calculate the probability of spontaneous explosion Prisk of curtain wall 6;

[0041] Step S3: Determine whether the risk probability Prisk exceeds the preset safety threshold. If it does, proceed to step S4; otherwise, return to step S1 to continue monitoring.

[0042] Step S4: Issue an alarm message and automatically trigger the sprinkler system to spray water to cool the outer surface of the curtain wall 6;

[0043] Step S5: Monitor the changes in temperature difference ΔT and strain ε during the cooling process, and stop spraying when the risk probability Prisk drops below the safety threshold.

[0044] Furthermore, in step S5, when the risk probability Prisk drops below the safety threshold, and when the temperature difference ΔT decreases to below 30% of the initial value and the strain ε falls back, it is determined that the risk has been eliminated.

[0045] The embodiments described herein are preferred embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Therefore, all equivalent changes made in accordance with the structure, shape, and principle of the present invention should be covered within the scope of protection of the present invention.

Claims

1. An early warning and control device for preventing curtain wall spontaneous explosion, comprising a sensing module (1), wherein the sensing module (1) comprises a temperature sensor array (101) for measuring the temperature difference between the indoor and outdoor surfaces of the curtain wall (6) and a strain gauge (102) for measuring the micro-strain of the surface of the curtain wall (6); characterized in that The sensing module (1) is connected to the data acquisition and communication module (2). The data acquisition and communication module (2) is used to receive the temperature difference data and strain data collected by the sensing module (1) and transmit them to the risk prediction module (3) in real time. The risk prediction module (3) is connected to the data acquisition and communication module (2) to calculate the probability of spontaneous explosion of the curtain wall (6) based on the received temperature difference data and strain data; The early warning control module (4) is connected to the risk prediction module (3) to issue an alarm message when the probability of self-destruction exceeds the safety threshold and to automatically trigger the active control device (5); The active control device (5) is connected to the early warning control module (4) in communication. The active control device (5) includes a spray system installed on the curtain wall (6) for spraying water to cool the outer surface of the curtain wall (6).

2. The early warning and control device for preventing curtain wall spontaneous explosion according to claim 1, characterized in that... The risk prediction module (3) has a built-in risk prediction model trained using machine learning algorithms. Its input features include at least the real-time internal and external temperature difference, the rate of change of temperature difference, the real-time strain value, and the rate of change of strain.

3. The early warning and control device for preventing curtain wall spontaneous explosion according to claim 2, characterized in that... The machine learning algorithm is either an LSTM neural network or a gradient boosting decision tree algorithm.

4. The early warning and control device for preventing curtain wall spontaneous explosion according to claim 1, characterized in that... The early warning control module (4) supports multi-level early warning. When the temperature difference or strain exceeds the initial threshold but the risk probability value of spontaneous explosion is lower than the warning value, it sends attention information; when the risk probability value exceeds the warning value but is lower than the safety threshold, it issues an audible and visual alarm; when the risk probability value exceeds the safety threshold, it issues an audible and visual alarm and automatically triggers the sprinkler system.

5. The early warning and control device for preventing curtain wall spontaneous explosion according to claim 1, characterized in that... The sprinkler system includes a water storage mechanism (501), a water pipe (502) and an atomizing nozzle (503). The atomizing nozzle (503) is arranged on the top of the curtain wall (6). One end of the water pipe (502) is connected to the water storage mechanism (501), and the other end is connected to the atomizing nozzle (503). A solenoid valve (504) is provided on the water pipe (502).

6. The early warning and control device for preventing curtain wall spontaneous explosion according to claim 1, characterized in that... The strain gauge (102) is a fiber optic strain sensor.

7. The early warning and control device for preventing spontaneous explosion of curtain walls according to claim 1, characterized in that... Temperature sensor array (101) and strain gauge (102) are arranged at the edges and center of the curtain wall (6).

8. The early warning and control device for preventing spontaneous explosion of curtain walls according to claim 1, characterized in that... It also includes a data storage module (7), which is connected in communication with the sensing module (1), the early warning control module (4), and the active control device (5) to record monitoring data, early warning records and control actions, and upload them to the cloud database (8).

9. A method for early warning and control of curtain wall spontaneous explosion, characterized in that... This includes the following steps: Step S1: Real-time acquisition of indoor and outdoor surface temperatures of the curtain wall (6) using a temperature sensor array (101) to calculate the indoor and outdoor temperature difference ΔT; real-time acquisition of micro-strain ε on the surface of the curtain wall (6) using a strain gauge (102); Step S2: Input the real-time collected temperature difference ΔT and strain ε into the risk prediction model to calculate the spontaneous explosion risk probability Prisk of the curtain wall (6); Step S3: Determine whether the risk probability Prisk exceeds the preset safety threshold. If it does, proceed to step S4; otherwise, return to step S1 to continue monitoring. Step S4: Issue an alarm message and automatically trigger the sprinkler system to spray water to cool the outer surface of the curtain wall (6); Step S5: Monitor the changes in temperature difference ΔT and strain ε during the cooling process, and stop spraying when the risk probability Prisk drops below the safety threshold.

10. The early warning and control method for preventing spontaneous explosion of curtain walls according to claim 9, characterized in that... In step S5, the risk probability Prisk drops below the safety threshold, and when the temperature difference ΔT decreases to below 30% of the initial value and the strain ε falls back, the risk is determined to be eliminated.

Images