Fire hazard early warning system and method based on pyrolysis quality data analysis

Abstract

The invention relates to the technical field of fire hazard early warning, in particular to a fire hazard early warning system and method based on pyrolysis quality data analysis. A fire hazard early warning system based on pyrolysis quality data analysis comprises a data acquisition module used for acquiring pyrolysis quality related data and personnel flow related data; the first early warning value acquisition module is used for acquiring a first early warning value by using a first early warning adjustment formula according to the cracking quality related data; and the detection influence coefficient acquisition module is used for acquiring an abnormal maintenance personnel detection influence coefficient according to the personnel flow related data and the cracking quality related data. According to the invention, by introducing the personnel flow related data and the abnormal maintenance personnel detection influence coefficient, the interference gas brought by the personnel is reasonably compensated and corrected, the sensor detection data is ensured to more accurately reflect the actual environment condition, and the false alarm probability caused by personnel interference is significantly reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of fire warning, and in particular to a fire hazard warning system and method based on pyrolysis mass data analysis. Background Art

[0002] In recent years, with the continuous expansion of the construction scale of urban underground pipe corridors and the increasingly strict management requirements, the safety monitoring of underground pipe corridors has become an important part of urban safety management. In underground pipe corridors, pyrolysis mass sensors, as a key environmental monitoring device, are mainly used to capture and monitor the content of pyrolysis particles and related harmful gases in order to timely warn of fires or other safety hazards. However, in the actual application process, there is a technical problem that urgently needs to be solved: when personnel are performing maintenance in the underground pipe corridor, due to the relevant gases monitored by the pyrolysis mass sensors attached to the clothes or tools carried by the personnel themselves, when the personnel move within the monitoring range of the pyrolysis mass sensors, the gas concentration detected by the sensors is abnormal, resulting in frequent false alarms and error alarms. This problem not only increases the workload of maintenance personnel, but also causes chaos in safety monitoring data, reducing the overall accuracy and reliability of the warning system.

[0003] Currently, the existing technology mainly relies on the monitoring data of a single sensor, and does not fully consider the interference of personnel flow and the gases they carry in the environment on the sensor detection data. Some solutions attempt to reduce the false alarm rate through simple filtering and data smoothing techniques, but it is often difficult to distinguish abnormal data caused by personnel influence from real environmental anomalies. Therefore, there is an urgent need for a fire hazard warning system and method that can comprehensively consider personnel flow, sensor distribution, and multi-modal environmental data, so as to accurately eliminate false alarms caused by personnel interference in a complex underground pipe corridor environment and ensure the authenticity and effectiveness of warning information. Summary of the Invention

[0004] In order to overcome the disadvantage of the interference of the gases carried by personnel on the detection data of pyrolysis mass sensors, the present invention provides a fire hazard warning system and method based on pyrolysis mass data analysis.

[0005] The technical solution is as follows: A fire hazard warning system based on pyrolysis mass data analysis, comprising: A data acquisition module for acquiring pyrolysis mass-related data and personnel flow-related data; A first warning value acquisition module for acquiring a first warning value according to the pyrolysis mass-related data using a first warning adjustment formula; A detection influence coefficient acquisition module for acquiring an abnormal maintenance personnel detection influence coefficient according to the personnel flow-related data and the pyrolysis mass-related data; A collaborative data acquisition module, configured to determine a collaborative pyrolysis mass sensor according to a target pyrolysis mass sensor and acquire data related to the collaborative pyrolysis mass sensor; A second warning value acquisition module, configured to use a second warning adjustment formula according to the data related to the collaborative pyrolysis mass sensor to acquire a second warning value; An auxiliary warning value acquisition module, configured to obtain an auxiliary warning value according to a fire warning model; A final warning value acquisition module, configured to use a final warning formula to obtain a final warning value according to a first warning value, a second warning value, and an auxiliary warning value; A warning adjustment module, configured to perform warning according to the final warning value and adjust the pyrolysis mass sensor.

[0006] Preferably, the data acquisition module is configured to acquire pyrolysis mass-related data and personnel flow-related data, including: acquiring pyrolysis mass-related data and personnel flow-related data within a target area, where the pyrolysis mass-related data includes the standard content of related gases, the positions of pyrolysis mass sensors, the monitoring ranges of pyrolysis mass sensors, and the number of pyrolysis mass sensors; the personnel flow-related data includes the stay duration of abnormal maintenance personnel and the positions of abnormal maintenance personnel; using a first warning adjustment formula according to the pyrolysis mass-related data to acquire a first warning value, where the abnormal maintenance personnel are maintenance personnel carrying related interfering gases; the pyrolysis mass-related data within the target area is pyrolysis mass-related data and personnel flow-related data in an underground pipe gallery; the pyrolysis mass sensor is used to capture and monitor related gases; the related gases are pyrolysis particles.

[0007] Preferably, the first warning value acquisition module is configured to use a first warning adjustment formula according to the pyrolysis mass-related data to acquire a first warning value, including: after performing normalization adjustment on the target pyrolysis mass sensor according to the pyrolysis mass-related data, using the first warning adjustment formula to acquire a first warning value, and when the distance between the pyrolysis mass sensor and the abnormal maintenance personnel is less than a second preset range, using this pyrolysis mass sensor as the target pyrolysis mass sensor, where the first warning adjustment formula is: ; In the formula, is the first warning value; is the detection influence coefficient of abnormal maintenance personnel; is the detected content of related gases monitored by the target pyrolysis mass sensor; is the standard content of related gases; is the number of pyrolysis mass sensors that alarm within the first preset range of the target pyrolysis mass sensor; is the adjustment coefficient of the first warning adjustment formula.

[0008] Preferably, the detection influence coefficient acquisition module is configured to obtain an abnormal maintenance personnel detection influence coefficient according to the personnel flow related data and the pyrolysis mass related data, including: after performing normalization adjustment on the personnel flow related data and the pyrolysis mass related data, obtaining the abnormal maintenance personnel detection influence coefficient, and respectively obtaining the residence time of the abnormal maintenance personnel when an alarm occurs within the monitoring range of each pyrolysis mass sensor and the distance between the abnormal maintenance personnel and each pyrolysis mass sensor when an alarm occurs within the monitoring range of each pyrolysis mass sensor according to the position of the abnormal maintenance personnel, the monitoring range of the pyrolysis mass sensor, and the position of the pyrolysis mass sensor. ; In the formula, is the abnormal maintenance personnel detection influence coefficient; is the residence time of the abnormal maintenance personnel when an alarm occurs within the monitoring range of the th pyrolysis mass sensor; is the distance between the abnormal maintenance personnel and the th pyrolysis mass sensor when an alarm occurs within the monitoring range of the th pyrolysis mass sensor; is the weight adjustment coefficient of the th pyrolysis mass sensor; is the adjustment factor.

[0009] Preferably, the collaborative data acquisition module is configured to determine collaborative pyrolysis mass sensors according to the target pyrolysis mass sensor and obtain collaborative pyrolysis mass sensor related data, including: obtaining collaborative pyrolysis mass sensor related data that has not been passed by abnormal maintenance personnel within the third preset range of the target pyrolysis mass sensor, where the collaborative pyrolysis mass sensor related data includes the position of the collaborative pyrolysis mass sensor, the early warning accuracy rate of the collaborative pyrolysis mass sensor, and the number of collaborative pyrolysis mass sensors, and obtaining a second early warning value according to the collaborative pyrolysis mass sensor related data using a second early warning adjustment formula.

[0010] Preferably, the second early warning value acquisition module is configured to obtain a second early warning value according to the collaborative pyrolysis mass sensor related data using a second early warning adjustment formula, including: obtaining the distance between each collaborative pyrolysis mass sensor and the target pyrolysis mass sensor according to the positions of each collaborative pyrolysis mass sensor and the target pyrolysis mass sensor, and using the second early warning adjustment formula to obtain the second early warning value, where the second early warning formula is: ; In the formula, is the second early warning value; is the early warning accuracy rate of the th collaborative pyrolysis mass sensor; is the distance between the th collaborative pyrolysis mass sensor and the target pyrolysis mass sensor; is the relevant gas detection content of the th collaborative pyrolysis mass sensor; is the standard content of the relevant gas; is the number of collaborative pyrolysis mass sensors; is the weight adjustment coefficient of the

[0011] Preferably, the auxiliary early warning value acquisition module is used to obtain an auxiliary early warning value according to a fire early warning model, including: obtaining an auxiliary early warning value through a fire early warning model of multimodal fusion of non-pyrolysis mass sensors, collecting multimodal sensor data through an environmental temperature sensor, a humidity sensor, a smoke concentration sensor, an infrared image sensor and a sound monitoring sensor, and performing denoising, normalization, time series correction and data enhancement processing, using a deep learning algorithm combining a convolutional neural network and a recurrent neural network, training to obtain a fire early warning model, and performing real-time analysis and calculation on the comprehensive fire risk feature vector based on the fire early warning model to obtain an auxiliary early warning value.

[0012] Preferably, the final early warning value acquisition module is used to obtain a final early warning value according to the first early warning value, the second early warning value and the auxiliary early warning value, including: obtaining a final early warning value according to the first early warning value, the second early warning value and the auxiliary early warning value, where the final early warning formula is: ; In the formula, is the final early warning value; is the first early warning value; is the second early warning value; is the auxiliary early warning value; is the influence coefficient of abnormal maintenance personnel detection; is the weight adjustment coefficient.

[0013] Preferably, the early warning adjustment module is used to give an early warning according to the final early warning value and adjust the pyrolysis mass sensor, including: when the final early warning value is greater than or equal to a preset early warning threshold and the duration is greater than a preset early warning time, sending an early warning message to relevant personnel, and adjusting the number and position of the pyrolysis mass sensors according to the final early warning value.

[0014] Preferably, a fire hazard early warning method based on pyrolysis mass data analysis further includes: S1: Obtain pyrolysis mass-related data and personnel flow-related data; S2: Obtain a first early warning value according to the pyrolysis mass-related data using a first early warning adjustment formula; S3: Use the second warning adjustment formula based on the relevant data of the co-pyrolysis mass sensor to obtain the second warning value; S4: Obtain the auxiliary warning value according to the fire warning model; S5: Use the final warning formula to obtain the final warning value based on the first warning value, the second warning value, and the auxiliary warning value; S6: Give a warning according to the final warning value and adjust the pyrolysis mass sensor.

[0015] The present invention has the following advantages: 1. By introducing the personnel flow data and the influence coefficient of abnormal maintenance personnel detection, the present invention reasonably compensates and corrects the interfering gases brought in by personnel, ensures that the detection data of the sensor more accurately reflects the actual environmental conditions, and significantly reduces the false alarm probability caused by personnel interference; 2. By adopting a multi-module collaborative working mode of data acquisition, collaborative data processing, warning value calculation, and deep learning-assisted warning, the present invention comprehensively utilizes the data of various sensors such as pyrolysis mass sensors, temperature, humidity, smoke, infrared images, and sounds to form a multi-angle and all-round monitoring network, thereby improving the reliability and real-time performance of fire hazard warning; 3. By using the warning adjustment module, the present invention intelligently adjusts the quantity and position of the pyrolysis mass sensors according to the final warning value, can adapt to the actual scenario with variable working conditions in the underground pipe gallery environment, and realizes a dynamic adaptive warning response mechanism. Description of the Drawings

[0016] Figure 1 is a schematic structural diagram of a fire hazard warning system based on pyrolysis mass data analysis of the present invention; Figure 2 is a flowchart of a fire hazard warning method based on pyrolysis mass data analysis of the present invention. Detailed Embodiments

[0017] The technical solutions in the embodiments of the present invention will be clearly and completely described below. Apparently, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0018] Embodiment 1: A fire hazard warning system based on pyrolysis mass data analysis, as Figure 1 shown, includes: A data acquisition module for acquiring pyrolysis mass-related data and personnel flow-related data; Obtain the pyrolysis mass-related data and personnel flow-related data within the target area. The pyrolysis mass-related data includes the standard content of the relevant gas, the position of the pyrolysis mass sensor, the monitoring range of the pyrolysis mass sensor, and the number of pyrolysis mass sensors. The personnel flow-related data includes the stay duration of the abnormal maintenance personnel and the position of the abnormal maintenance personnel. Use the first warning adjustment formula to obtain the first warning value according to the pyrolysis mass-related data, where the abnormal maintenance personnel are the maintenance personnel carrying the relevant interfering gas. The pyrolysis mass-related data within the target area is the pyrolysis mass-related data and personnel flow-related data in the underground pipe gallery. The pyrolysis mass sensor is used to capture and monitor the relevant gas. The relevant gas is pyrolytic particles.

[0019] It should be noted that through multiple pyrolysis mass sensors pre-deployed in the underground pipe gallery, the real-time monitoring of the relevant gas (mainly pyrolytic particles) is realized. Each sensor has been calibrated before installation, and the standard content data of the relevant gas is used as the benchmark for subsequent data comparison. The position, monitoring range and number of each sensor have been set during the initial configuration, and the status is updated and the data is verified regularly during the operation process to ensure that the obtained pyrolysis mass-related data truly reflects the actual environmental status in the underground pipe gallery and provides accurate basic data for the subsequent calculation of the first warning value. All personnel entering the pipe gallery are monitored in real time, and the stay duration and position data of the abnormal maintenance personnel are mainly collected. The abnormal maintenance personnel refer to the maintenance personnel carrying the relevant interfering gas that interferes with the detection of the pyrolysis mass sensor during the maintenance process.

[0020] The first warning value acquisition module is used to obtain the first warning value according to the pyrolysis mass-related data using the first warning adjustment formula. After normalizing and adjusting the target pyrolysis mass sensor according to the pyrolysis mass-related data, use the first warning adjustment formula to obtain the first warning value. When the distance between the pyrolysis mass sensor and the abnormal maintenance personnel is less than the second preset range, this pyrolysis mass sensor is used as the target pyrolysis mass sensor. The first warning adjustment formula is as follows: ; In the formula, is the first warning value; is the abnormal maintenance personnel detection influence coefficient; is the detected content of the relevant gas monitored by the target pyrolysis mass sensor; is the standard content of the relevant gas; is the number of pyrolysis mass sensors that alarm within the first preset range of the target pyrolysis mass sensor; is the adjustment coefficient of the first warning adjustment formula.

[0021] It should be noted that the collected pyrolysis mass-related data is normalized to determine the target pyrolysis mass sensor. When the distance between the pyrolysis mass sensor and the abnormal maintenance personnel is less than the second preset range, the sensor is determined as the target pyrolysis mass sensor, ensuring that the data of the pyrolysis mass sensor can be recognized in time when interfered by personnel, and the data is standardized through normalization adjustment, quickly locating the interfered sensor and monitoring it key points. Furthermore, by adjusting the parameters in the formula, the monitored data of the sensor reflects the change of gas concentration in the real environment after weight adjustment, thereby improving the accuracy of early warning data; The detection influence coefficient for abnormal maintenance personnel is calculated from the personnel flow-related data and the pyrolysis mass data, and is used to reflect the interference degree of abnormal maintenance personnel on the sensor detection data, which can effectively reduce the false alarm signals generated by abnormal maintenance personnel carrying interfering gases.

[0022] The detection influence coefficient acquisition module is used to obtain the detection influence coefficient of abnormal maintenance personnel according to the personnel flow-related data and the pyrolysis mass-related data; After normalization adjustment based on the personnel flow-related data and the pyrolysis mass-related data, the detection influence coefficient of abnormal maintenance personnel is obtained. According to the position of the abnormal maintenance personnel, the monitoring range of the pyrolysis mass sensor, and the position of the pyrolysis mass sensor, the residence time when the abnormal maintenance personnel alarms within the monitoring range of each pyrolysis mass sensor and the distance between the abnormal maintenance personnel and each pyrolysis mass sensor when the abnormal maintenance personnel alarms within the monitoring range of each pyrolysis mass sensor are obtained respectively, ; In the formula, is the detection influence coefficient of abnormal maintenance personnel; is the residence time when the abnormal maintenance personnel alarms within the monitoring range of the th pyrolysis mass sensor; is the distance between the abnormal maintenance personnel and the th pyrolysis mass sensor when the abnormal maintenance personnel alarms within the monitoring range of the th pyrolysis mass sensor; is the weight adjustment coefficient of the th pyrolysis mass sensor; is the adjustment factor.

[0023] It should be noted that the personnel flow-related data and the pyrolysis mass-related data collected from the underground pipe gallery are normalized. Through normalization processing, the influence caused by the difference in dimension and value between data is eliminated, providing unified standard data input for subsequent calculations. is the residence time when the abnormal maintenance personnel alarms within the monitoring range of the th pyrolysis mass sensor, specifically when the When a cracking mass sensor issues an alarm, analyze the residence time of abnormal maintenance personnel within the monitoring range of the th cracking mass sensor. The residence time can reflect the degree of emission of interfering gases from the abnormal maintenance personnel. When the residence time is longer, the confidence level of the cracking mass sensor decreases; is the distance between the abnormal maintenance personnel and the th cracking mass sensor when the alarm is triggered within the monitoring range of the th cracking mass sensor. Specifically, ensure that the abnormal maintenance personnel are within the monitoring range of the th cracking mass sensor. At the same time, when the th cracking mass sensor issues an alarm, it is the distance between the abnormal maintenance personnel and the th cracking mass sensor, The maximum value of which is the monitoring range of the th cracking mass sensor; comprehensively consider the presence time of abnormal maintenance personnel within the monitoring range of each sensor and their relative distance from the sensor, so as to obtain the detection influence coefficient of abnormal maintenance personnel. The detection influence coefficient of abnormal maintenance personnel will play a role in reducing false alarms caused by interfering data in subsequent early warning value adjustment, ensuring that the early warning results are more real and effective.

[0024] The collaborative data acquisition module is used to determine the collaborative cracking mass sensor according to the target cracking mass sensor and acquire the relevant data of the collaborative cracking mass sensor; Acquire the relevant data of the collaborative cracking mass sensor that has not been passed by abnormal maintenance personnel within the third preset range of the target cracking mass sensor. The relevant data of the collaborative cracking mass sensor includes the position of the collaborative cracking mass sensor, the early warning accuracy rate of the collaborative cracking mass sensor, and the number of collaborative cracking mass sensors. According to the relevant data of the collaborative cracking mass sensor, use the second early warning adjustment formula to obtain the second early warning value.

[0025] It should be noted that based on the target cracking mass sensor, the collaborative data acquisition module further acquires the relevant data of the collaborative cracking mass sensor that has not been passed by abnormal maintenance personnel within the third preset range. The third preset range is a preset spatial range used to exclude sensors with abnormal data caused by the interference of abnormal maintenance personnel, ensuring that the relevant data of the acquired collaborative cracking mass sensor has a high early warning accuracy rate and credibility; the position data of the collaborative cracking mass sensor will be used to calculate the distance between it and the target cracking mass sensor, while the early warning accuracy rate reflects the performance and reliability of the sensor in early warning of fire hazards in historical data.

[0026] The second early warning value acquisition module is used to obtain the second early warning value according to the relevant data of the collaborative cracking mass sensor using the second early warning adjustment formula; According to the positions of each collaborative pyrolysis mass sensor and the target pyrolysis mass sensor, the distances between each collaborative pyrolysis mass sensor and the target pyrolysis mass sensor are obtained, and the second warning adjustment formula is used to obtain the second warning value, where the second warning formula is: ; In the formula, is the second warning value; is the warning accuracy rate of the th collaborative pyrolysis mass sensor; is the th distance between the collaborative pyrolysis mass sensor and the target pyrolysis mass sensor; is the th relevant gas detection content of the collaborative pyrolysis mass sensor; is the standard content of the relevant gas; is the number of collaborative pyrolysis mass sensors; is the th weight adjustment coefficient of the collaborative pyrolysis mass sensor.

[0027] It should be noted that is the th distance between the collaborative pyrolysis mass sensor and the target pyrolysis mass sensor, which is used to reflect the distance of the spatial position between each collaborative pyrolysis mass sensor and the target pyrolysis mass sensor. The farther the distance, the relatively lower the correlation between the sensors; through the second warning value, it is ensured that the collaborative pyrolysis mass data can play a role in supplementing and verifying the target sensor data in the warning, thereby improving the overall accuracy and reliability of the fire hazard warning.

[0028] The auxiliary warning value acquisition module is used to obtain the auxiliary warning value according to the fire warning model; Through the fire warning model of multi-modal fusion of non-pyrolysis mass sensors, the auxiliary warning value is obtained. Multi-modal sensor data is collected through environmental temperature sensors, humidity sensors, smoke concentration sensors, infrared image sensors and sound monitoring sensors, and denoising, normalization, time series correction and data enhancement processing are carried out. A deep learning algorithm combining convolutional neural network and recurrent neural network is used for training to obtain the fire warning model, and the auxiliary warning value is obtained by real-time analysis and calculation of the comprehensive fire risk feature vector based on the fire warning model.

[0029] It should be noted that multi-modal sensor data from the site is collected through an environmental temperature sensor, a humidity sensor, a smoke concentration sensor, an infrared image sensor, and a sound monitoring sensor. These sensors are distributed within the target area and are respectively used to capture environmental temperature, humidity, smoke concentration, infrared images, and sound signals. The data of various sensors can comprehensively reflect multiple physical characteristics related to the on-site environment and fire hazards, providing multi-angle information for fire risk assessment. The collected multi-modal sensor data undergoes preprocessing steps, including denoising, normalization, time series correction, and data augmentation operations. The denoising process filters out random interference signals in the environment, normalization makes the data of different sensors within the same dimension range, time series correction ensures the accurate alignment of data in the time dimension, and data augmentation improves the adaptability and robustness of the model to different scenarios. The preprocessed multi-modal data is trained through a deep learning algorithm that combines a convolutional neural network and a recurrent neural network to construct a fire warning model, effectively capturing the subtle changes in the environmental state before a fire occurs, thereby providing an auxiliary decision-making basis for fire warning.

[0030] A final warning value acquisition module, configured to obtain a final warning value according to a first warning value, a second warning value, and an auxiliary warning value by using a final warning formula; According to the first warning value, the second warning value, and the auxiliary warning value, a final warning value is obtained by using a final warning formula, where the final warning formula is: ; In the formula, is the final warning value; is the first warning value; is the second warning value; is the auxiliary warning value; is the influence coefficient of abnormal maintenance personnel detection; is the weight adjustment coefficient.

[0031] It should be noted that the first warning value is calculated by the first warning value acquisition module, and this value is adjusted based on the relevant data of the pyrolysis mass sensor and the influence coefficient of abnormal maintenance personnel detection, reflecting the warning information obtained from the monitoring data of the target pyrolysis mass sensor; the second warning value is calculated by the second warning value acquisition module, and this value combines factors such as the distance between collaborative pyrolysis mass sensors, the warning accuracy rate, and the detected content, and is obtained after being operated by the second warning adjustment formula; the auxiliary warning value is obtained by the auxiliary warning value acquisition module through real-time analysis and calculation of the fire warning model and multi-modal sensor data, mainly reflecting the auxiliary judgment of fire hazards by non-pyrolysis mass sensor information such as environmental temperature, humidity, smoke concentration, infrared images, and sounds; through the final warning formula, dynamic compensation for data interference caused by abnormal maintenance personnel is realized, making the final warning value more comprehensively and accurately reflect the fire hazard risk.

[0032] An early warning adjustment module, configured to perform early warning based on the final early warning value and adjust the pyrolysis mass sensor.

[0033] When the final early warning value is greater than or equal to a preset early warning threshold and the duration is greater than a preset early warning time, an early warning message is sent to relevant personnel, and the quantity and position of the pyrolysis mass sensor are adjusted according to the final early warning value.

[0034] It should be noted that when the final early warning value simultaneously meets the conditions of being greater than or equal to the preset early warning threshold and the duration being greater than the preset early warning time, the early warning adjustment module immediately sends an early warning message to relevant personnel. The early warning message includes a fire hazard risk reminder, an emergency evacuation notice, and other necessary safety instructions. At the same time, the quantity and position of the pyrolysis mass sensor are adjusted according to the final early warning value, and the area with potential hazards is strengthened for monitoring.

[0035] Embodiment 2: On the basis of Embodiment 1, a fire hazard early warning method based on pyrolysis mass data analysis, as Figure 2 shown, further includes: S1: Obtain pyrolysis mass-related data and personnel flow-related data; S2: Use a first early warning adjustment formula to obtain a first early warning value according to the pyrolysis mass-related data; S3: Use a second early warning adjustment formula according to the relevant data of the collaborative pyrolysis mass sensor to obtain a second early warning value; S4: Obtain an auxiliary early warning value according to the fire early warning model; S5: Use a final early warning formula to obtain a final early warning value according to the first early warning value, the second early warning value, and the auxiliary early warning value; S6: Perform early warning based on the final early warning value and adjust the pyrolysis mass sensor.

[0036] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit the protection scope of the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the essence and scope of the technical solutions of the present invention.

Claims

1. A fire hazard early warning system based on cracking mass data analysis, characterized in that: include: Data acquisition module, used to obtain cracking quality related data and personnel flow related data; A first warning value acquisition module, configured to acquire a first warning value using a first warning adjustment formula according to the pyrolysis mass related data; A detection influence coefficient acquisition module is used to obtain the detection influence coefficient of abnormal maintenance personnel according to the personnel flow related data and the cracking quality related data; A collaborative data acquisition module, used to determine a collaborative pyrolysis mass sensor according to a target pyrolysis mass sensor, and to acquire data related to the collaborative pyrolysis mass sensor; A second warning value acquisition module, configured to obtain a second warning value using a second warning adjustment formula according to the synergistic pyrolysis mass sensor related data; An auxiliary warning value acquisition module is used to obtain an auxiliary warning value according to a fire warning model; A final warning value acquisition module, used to obtain a final warning value using a final warning formula according to the first warning value, the second warning value and the auxiliary warning value; The early warning adjustment module is used to issue an early warning and adjust the pyrolysis mass sensor according to the final early warning value.

2. A fire hazard early warning system based on pyrolysis data analysis according to claim 1, characterized in that: The data acquisition module is used to acquire cracking material related data and personnel flow related data, including: acquiring cracking material related data and personnel flow related data in the target area, the cracking material related data includes the standard content of related gas, the position of the cracking material sensor, the cracking material sensor monitoring range and the number of cracking material sensors; the personnel flow related data includes the stay time and the position of the abnormal maintenance personnel; according to the cracking material related data, a first warning adjustment formula is used to obtain a first warning value, wherein the abnormal maintenance personnel are maintenance personnel carrying related interfering gases; the cracking material related data in the target area are cracking material related data and personnel flow related data in the underground pipe gallery; the cracking material sensor is used to capture and monitor related gases; the related gases are pyrolysis particles.

3. A fire hazard early warning system based on pyrolysis material data analysis according to claim 1, characterized in that: The first warning value acquisition module is used to obtain the first warning value according to the cracking mass related data using the first warning adjustment formula, including: after normalizing and adjusting the target cracking mass sensor according to the cracking mass related data, using the first warning adjustment formula to obtain the first warning value, when the distance between the cracking mass sensor and the abnormal maintenance personnel is less than a second preset range, the cracking mass sensor is used as the target cracking mass sensor, wherein the first warning adjustment formula is: ； In the formula, is the first warning value; Detect influence coefficients for abnormal maintenance personnel; The relevant gas detection content monitored by the target pyrolysis material sensor; is the standard content of the relevant gas; The number of pyrolysis mass sensors that alarm within a first preset range of the target pyrolysis mass sensor; It is the adjustment coefficient of the first warning adjustment formula.

4. A fire hazard early warning system based on pyrolysis data analysis according to claim 3, characterized in that: The detection influence coefficient acquisition module is used to acquire the detection influence coefficient of abnormal maintenance personnel according to the personnel flow related data and the cracking quality related data, including: acquiring the detection influence coefficient of abnormal maintenance personnel after normalization adjustment according to the personnel flow related data and the cracking quality related data, and acquiring the residence time of the abnormal maintenance personnel when the abnormal maintenance personnel alarms within the monitoring range of each cracking quality sensor and the distance between the abnormal maintenance personnel and each cracking quality sensor when the abnormal maintenance personnel alarms within the monitoring range of each cracking quality sensor according to the position of the abnormal maintenance personnel, the monitoring range of the cracking quality sensor and the position of the cracking quality sensor, respectively; ； In the formula, Detect influence coefficients for abnormal maintenance personnel; For abnormal maintenance personnel The length of time a pyrolysis mass sensor stays within the monitoring range when an alarm is sounded; For abnormal maintenance personnel When the first pyrolysis mass sensor alarms within the monitoring range, The distance between the lysate sensors; For the The weight adjustment coefficient of each pyrolysis mass sensor; is the adjustment factor.

5. A fire hazard early warning system based on pyrolysis material data analysis according to claim 1, characterized in that: The collaborative data acquisition module is used to determine the collaborative cracking mass sensor according to the target cracking mass sensor and obtain the collaborative cracking mass sensor related data, including: obtaining the collaborative cracking mass sensor related data within a third preset range of the target cracking mass sensor that has not been passed by abnormal maintenance personnel, the collaborative cracking mass sensor related data including the collaborative cracking mass sensor position, the collaborative cracking mass sensor early warning accuracy and the number of collaborative cracking mass sensors, and using the second early warning adjustment formula according to the collaborative cracking mass sensor related data to obtain a second early warning value.

6. A fire hazard early warning system based on pyrolysis data analysis according to claim 1, characterized in that: The second warning value acquisition module is used to obtain the second warning value using the second warning adjustment formula according to the cooperative cracking mass sensor related data, including: obtaining the distance between each cooperative cracking mass sensor and the target cracking mass sensor according to the position of each cooperative cracking mass sensor and the position of the target cracking mass sensor, and using the second warning adjustment formula to obtain the second warning value, wherein the second warning formula is: ； In the formula, is the second warning value; For the The early warning accuracy of the synergistic lysate sensor; For the The distance between the cooperative lysate sensor and the target lysate sensor; For the The relevant gas detection content of a synergistic pyrolysis mass sensor; is the standard content of the relevant gas; is the number of synergistic lysate sensors; For the The weight adjustment coefficient of the synergistic lysate sensor.

7. A fire hazard early warning system based on pyrolysis material data analysis according to claim 1, characterized in that: The auxiliary warning value acquisition module is used to obtain the auxiliary warning value according to the fire warning model, including: obtaining the auxiliary warning value through the fire warning model of multimodal fusion of non-fragmentation sensors, collecting multimodal sensor data through ambient temperature sensors, humidity sensors, smoke concentration sensors, infrared image sensors and sound monitoring sensors, and performing denoising, normalization, timing correction and data enhancement processing, using a deep learning algorithm combining a convolutional neural network and a recurrent neural network to obtain a fire warning model after training, and performing real-time analysis and calculation of a comprehensive fire risk feature vector based on the fire warning model to obtain the auxiliary warning value.

8. A fire hazard early warning system based on pyrolysis material data analysis according to claim 1, characterized in that: The final warning value acquisition module is used to obtain the final warning value using the final warning formula according to the first warning value, the second warning value and the auxiliary warning value, including: obtaining the final warning value using the final warning formula according to the first warning value, the second warning value and the auxiliary warning value, wherein the final warning formula is: ； In the formula, is the final warning value; is the first warning value; is the second warning value; It is the auxiliary warning value; Detect influence coefficients for abnormal maintenance personnel; is the weight adjustment coefficient.

9. A fire hazard early warning system based on pyrolysis data analysis according to claim 1, characterized in that: The warning adjustment module is used to issue a warning and adjust the cracking mass sensor according to the final warning value, including: when the final warning value is greater than or equal to a preset warning threshold and the duration is greater than a preset warning time, issuing a warning message to relevant personnel, and adjusting the number and position of the cracking mass sensor according to the final warning value.

10. A fire hazard early warning method based on pyrolysis data analysis, according to any one of claims 1 to 9, a fire hazard early warning system based on pyrolysis data analysis, characterized in that: include: S1: Obtain data related to cracking materials and personnel flow; S2: obtaining a first warning value using a first warning adjustment formula according to the pyrolysis mass related data; S3: using a second warning adjustment formula according to the data related to the cooperative pyrolysis mass sensor to obtain a second warning value; S4: Obtaining auxiliary warning values ​​according to the fire warning model; S5: obtaining a final warning value using a final warning formula according to the first warning value, the second warning value and the auxiliary warning value; S6: issuing an early warning and adjusting the pyrolysis medium sensor according to the final early warning value.

Images