Fire-fighting information tracing analysis system based on artificial intelligence

Abstract

The invention relates to a fire-fighting information tracing analysis system based on artificial intelligence, and belongs to the technical field of fire-fighting systems, the fire-fighting information tracing analysis system comprises a plurality of monitoring modules and a central control module, and the plurality of monitoring modules are in communication connection with the central control module; the monitoring module is used for collecting fire information in a monitoring area and preliminarily judging whether a fire occurs or not, and when the judgment result is yes, the monitoring module uploads the fire information to the central control module; the central control module comprises a data storage module and an analysis module, past fire information is stored in the storage module, and the central control module extracts a past fire spreading speed by using the past fire information, calculates the fire spreading speed and records the fire spreading speed as a first predicted fire spreading speed; and the central control module calculates a comprehensive fire spreading speed according to the passing fire spreading speed and the first predicted fire spreading speed, and sends alarms to different rescue teams when the comprehensive fire spreading speed exceeds different fire thresholds.

Description

Technical Field

[0001] The present invention belongs to the technical field of fire protection systems, and in particular relates to a fire protection information tracing and analysis system based on artificial intelligence. Background Art

[0002] Factories or warehouses are an important part of production and life, but they usually store a large number of flammable and combustible items. Therefore, it is necessary to monitor abnormal situations such as fires in real time, determine the degree of fire risk, and send relevant information to relevant personnel or rescue teams.

[0003] Conventional systems use fire maintenance records only as a reference for repairing fire equipment failures. This superficial use is limited to the use of a single fire maintenance record, lacking the centralized and in-depth application of multiple fire maintenance records. This makes it impossible to provide feedback references for regulating fire maintenance cycles. To this end, Chinese patent CN115601010B discloses an artificial intelligence fire protection smart information management platform, which includes a park fire maintenance record extraction module, a fire maintenance record classification module, an abnormal fire equipment screening module, a fire information database, an abnormal fire equipment abnormal maintenance location analysis module, an abnormal operation anomaly tracing module for abnormal maintenance locations, and a dynamic regulation module for abnormal fire equipment maintenance cycles. The platform classifies fire maintenance records in industrial parks within a set time period according to fire equipment, identifies abnormal fire equipment and its corresponding abnormal maintenance locations, and then traces the causes of the abnormal operation of the abnormal maintenance locations. This allows for dynamic regulation of the maintenance cycles of abnormal fire equipment, achieving centralized and in-depth application of fire maintenance records, providing feedback references for regulating fire equipment maintenance cycles, improving the utilization rate of maintenance information in fire maintenance records, and avoiding waste of fire maintenance resources.

[0004] In the above scheme, although the camera effectiveness corresponding to each associated camera is compared and the associated camera corresponding to the maximum camera effectiveness is selected as the effective monitoring camera corresponding to the abnormal fire-fighting equipment, it only uses the traceability mechanism to deploy cameras and cannot use the traceability mechanism to analyze the fire situation. The utilization rate of the traceability mechanism and past data is not high, and it is impossible to combine the traceability mechanism and the real-time analysis mechanism to judge the fire situation. Therefore, a fire information traceability analysis system based on artificial intelligence is needed that can combine the traceability mechanism to predict the fire situation and has high prediction efficiency. Summary of the Invention

[0005] In order to solve the above problems existing in the prior art, the present invention provides a fire information tracing and analysis system based on artificial intelligence, which has the characteristics of being able to predict fire conditions in combination with a tracing mechanism and having high prediction efficiency.

[0006] The purpose of the present invention can be achieved through the following technical solutions:

[0007] A fire information tracing and analysis system based on artificial intelligence includes several monitoring modules and a central control module, wherein the several monitoring modules are respectively connected to the central control module for mutual communication;

[0008] The monitoring module is used to collect fire information in the monitoring area and preliminarily determine whether a fire has occurred. If the judgment result is yes, the monitoring module uploads the fire information to the central control module;

[0009] The central control module includes a data storage module and an analysis module. Past fire information is stored in the storage module. Each time the central control module receives fire information, it compares the current fire information with the past fire information, filters out the past fire information closest to the current fire information, extracts the fire spread speed from the closest past fire information, and records it as the past fire spread speed. Each time the central control module receives fire information, it simultaneously calls the analysis module to calculate the fire spread speed, which is recorded as the first predicted fire spread speed. The central control module calculates the comprehensive fire spread speed based on the past fire spread speed and the first predicted fire spread speed, and sends an alarm to different rescue teams when the comprehensive fire spread speed exceeds different fire thresholds.

[0010] As a preferred technical solution of the present invention, the central control module calculates the comprehensive fire spread rate Z based on the past fire spread rate G and the first predicted fire spread rate Y, where Z = (k1×G+k2×Y) / (k1+k2), where k1 and k2 are pre-input constants.

[0011] As a preferred technical solution of the present invention, any of the monitoring modules includes an edge computing module. When the monitoring module determines that a fire has occurred, it samples the fire information to obtain simplified fire information, and uploads the simplified fire information to the central control module. The central control module is used to calculate the comprehensive fire spread speed based on the simplified fire information. The monitoring module calls the edge computing module to calculate a second predicted fire spread speed based on the fire information, and uploads the second predicted fire spread speed to the central control module. The central control module takes the maximum value D between the second predicted fire spread speed or the comprehensive fire spread speed, and sends an alarm to different rescue teams when D exceeds different fire thresholds.

[0012] As a preferred technical solution of the present invention, several fire thresholds are pre-set, and the several fire thresholds are set in one-to-one correspondence with rescue teams of different levels.

[0013] As a preferred technical solution of the present invention, the monitoring module is used to read the network delay between itself and the central control module, and reduce the number of samples in the fire information sampling process when the network delay exceeds a threshold.

[0014] As a preferred technical solution of the present invention, the monitoring module is used to read the network delay C between itself and the central control module, and adjust the number of samples in the fire information sampling process to X times the standard value, where X = C / C0×d, C0 is the pre-input standard network delay.

[0015] As a preferred technical solution of the present invention, the monitoring module uploads X synchronously each time it uploads simplified fire information, and the central control module is used to record the X values corresponding to the simplified fire information as the same group and synchronously store them in the storage module as past fire information.

[0016] As a preferred technical solution of the present invention, it also includes an interactive interface, which is used to display the comprehensive fire spread speed Z and the second predicted fire spread speed.

[0017] The beneficial effects of the present invention are:

[0018] (1) After receiving fire information, the central control module first extracts the past fire spread speed from the past fire information by comparison, and simultaneously analyzes and predicts the fire spread speed. Finally, the comprehensive fire spread speed is calculated based on the past fire spread speed and the first predicted fire spread speed. The fire situation is analyzed using a retrospective mechanism, which improves the utilization rate of past data. The fire situation is judged by combining the retrospective mechanism and the real-time analysis mechanism, which improves the accuracy of the judgment.

[0019] (2) By setting up an edge computing module, in some cases, when the fire spreads rapidly and communicates with the central control module and relies on the central control module for judgment, there is a probability that the computing speed cannot keep up with the fire spread. The edge computing module is used to calculate the second fire spread speed in advance, and an alarm is issued to the corresponding rescue team when either the second fire spread speed or the comprehensive fire spread speed exceeds a certain threshold, taking into account both computing efficiency and computing accuracy;

[0020] (3) By reducing the number of samples in the fire information sampling process when the network delay exceeds the threshold, when the network delay is high and the communication is relatively poor, and the central control module cannot be relied upon to make a judgment on the fire situation, the data transmission volume is reduced, thereby reducing the network burden and increasing the degree of dependence on the edge computing module, further improving the computing efficiency under network delay conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] To facilitate understanding by those skilled in the art, the present invention is further described below with reference to the accompanying drawings.

[0022] Figure 1 This is a topological diagram of each module of the present invention. DETAILED DESCRIPTION

[0023] In order to further illustrate the technical means and effects adopted by the present invention to achieve the predetermined purpose of the invention, the specific implementation methods, structures, features and effects of the present invention are described in detail below in conjunction with the accompanying drawings and preferred embodiments.

[0024] See also Figure 1 , a fire information tracing and analysis system based on artificial intelligence, including several monitoring modules and a central control module, wherein the several monitoring modules are respectively communicated with the central control module;

[0025] The monitoring module is used to collect fire information in the monitoring area and preliminarily determine whether a fire has occurred. If the judgment result is yes, the monitoring module uploads the fire information to the central control module;

[0026] The central control module includes a data storage module and an analysis module. The storage module stores past fire information. Each time the central control module receives fire information, it compares the current fire information with the past fire information, selects the past fire information closest to the current fire information, extracts the fire spread speed from the closest past fire information, and records it as the past fire spread speed. Each time the central control module receives fire information, it simultaneously calls the analysis module to calculate the fire spread speed, which is recorded as the first predicted fire spread speed. The central control module calculates the comprehensive fire spread speed based on the past fire spread speed and the first predicted fire spread speed, and sends an alarm to different rescue teams when the comprehensive fire spread speed exceeds different fire thresholds.

[0027] Specifically, for the preliminary determination of whether a fire has occurred, each monitoring module is electrically connected to a temperature sensor and a smoke sensor. When the monitoring module determines that the temperature is greater than 70°C or the smoke concentration is greater than 5%, it determines that a fire has occurred;

[0028] For the calculation of past fire spread speed, each fire situation information consists of a series of images or videos arranged in time sequence. The monitoring module includes at least a camera, which is used to collect images or videos of the on-site fire and upload them to the control module. The images or videos contain flame outlines. The central control module is pre-input with a CNN image recognition model and a time series prediction algorithm. The central control module uses the CNN image recognition model to read the changes in the flame outline in the images or videos arranged in time sequence, and calculates the fire spread speed and flame area based on the time series prediction algorithm. Therefore, for each fire spread speed parameter, the unit is meters per second or pixels per second;

[0029] In the past fire information, the fire information belonging to the same fire is archived as a group. The past fire information consists of several groups of data. When a fire is handled, the fire spread speed and flame area are also archived as part of the data in the past fire information of this fire.

[0030] Each time the central control module receives fire information, it will identify the past fire information that is closest to the current fire information in terms of fire spread speed and flame area, and record the fire spread speed in the closest past fire information as the past fire spread speed.

[0031] For several fire thresholds, the control module is pre-input with different fire thresholds and contact information of different rescue teams. Different rescue teams are divided according to their size and rescue capabilities. Teams with lower size or rescue capabilities correspond to lower fire thresholds. That is, when the overall fire spread speed is low and the fire risk is low, rescue teams with lower rescue capabilities or smaller sizes are assigned to carry out rescue. When the overall fire spread speed is high and the fire risk is high, rescue teams with stronger rescue capabilities or larger sizes are assigned to carry out rescue, forming a graded response.

[0032] For sending alarms to different rescue teams, the central control module itself is provided with a communication module, which is used to send alarm signals to signal receivers such as terminals of corresponding rescue teams;

[0033] Specifically, the central control module calculates the comprehensive fire spread rate Z based on the past fire spread rate G and the first predicted fire spread rate Y, where Z = (k1×G+k2×Y) / (k1+k2), where k1 and k2 are pre-entered constants that serve as weights in the weighted average calculation of the past fire spread rate G and the first predicted fire spread rate Y. k1 and k2 are assigned by the operator after estimating the weights of the fire spread rate and the first predicted fire spread rate in the calculation based on historical data.

[0034] By making the central control module first extract the past fire spread speed from the past fire information through comparison each time it receives fire information, and simultaneously analyze and predict the fire spread speed, and finally calculate the comprehensive fire spread speed based on the past fire spread speed and the first predicted fire spread speed, the fire situation is analyzed using the retrospective mechanism, which improves the utilization rate of past data. The fire situation is judged by combining the retrospective mechanism and the real-time analysis mechanism, which improves the accuracy of the judgment.

[0035] In some cases, the fire spreads relatively quickly. At this time, the process of collecting data, sending the data to the central control module, processing the data by the central control module, and issuing the processing results takes a long time, and it may not be possible to make a judgment in time. For this reason, any monitoring module includes an edge computing module. When the monitoring module determines that a fire has occurred, it samples the fire information to obtain simplified fire information, and uploads the simplified fire information to the central control module. The central control module is used to calculate the comprehensive fire spread speed based on the simplified fire information. The monitoring module calls the edge computing module to calculate the second predicted fire spread speed based on the fire information, and uploads the second predicted fire spread speed to the central control module. The central control module takes the maximum value of the second predicted fire spread speed or the comprehensive fire spread speed, recorded as D, and sends an alarm to different rescue teams when D exceeds different fire thresholds;

[0036] Compared with the solution of uploading the entire acquired fire information to the central control module, only the simplified fire information and the second predicted fire spread speed are transmitted. The transmission speed is faster and the edge computing module is responsible for part of the fire information processing task, further improving the computing efficiency.

[0037] By setting up an edge computing module, we can avoid situations where the fire spreads rapidly and the computing speed may not be able to keep up with the fire spread when communicating with and relying on the central control module for judgment. Instead, we use the edge computing module to calculate the second fire spread speed in advance and send an alarm to the corresponding rescue team when either the second fire spread speed or the comprehensive fire spread speed exceeds a certain threshold, thus balancing computing efficiency and accuracy.

[0038] When the network delay is high, communication is relatively poor, and the central control module cannot be relied upon to make a judgment on the fire situation, it is necessary to enhance the role of the edge computing module in fire judgment, reduce the amount of data transmitted to the central control module, and further simplify the fire data. When the network delay is low, more fire data can be smoothly transmitted to the central control module. To this end, the monitoring module is used to read the network delay between itself and the central control module, and reduce the number of samples in the fire information sampling process when the network delay exceeds the threshold.

[0039] Specifically, the monitoring module is used to read the network delay C between itself and the central control module, and adjust the number of samples in the fire information sampling process to X times the standard value, where X = C / C0×d, C0 is the pre-input standard network delay.

[0040] By reducing the number of samples in the fire information sampling process when the network delay exceeds the threshold, when the network delay is high, the communication is relatively poor, and the central control module cannot be relied upon to make a judgment on the fire situation, the data transmission volume is reduced to reduce the network burden, increase the dependence on the edge computing module, and further improve the computing efficiency under network delay.

[0041] When using past fire information for calculations, the number of samples, representing the degree of simplification of the past fire information, is also used as a reference. For fire information with a small number of samples, the first fire spread rate derived from it is likely to be less accurate. In this case, the reference weight of this past fire information can be appropriately reduced. The monitoring module simultaneously uploads X each time it uploads simplified fire information. The central control module is used to record the X values corresponding to the simplified fire information as the same group and synchronously store them in the storage module as past fire information.

[0042] In order to facilitate the operators to view the real-time fire situation, an interactive interface is also included, which is used to display the comprehensive fire spread speed Z and the second predicted fire spread speed.

[0043] The above description is merely a preferred embodiment of the present invention and does not constitute any form of limitation to the present invention. Although the present invention has been disclosed as above in terms of a preferred embodiment, it is not intended to limit the present invention. Any person skilled in the art can, without departing from the scope of the technical solution of the present invention, make some changes or modifications to equivalent embodiments using the technical contents disclosed above. However, any brief modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention are still within the scope of the technical solution of the present invention.

Claims

1. A fire information tracing and analysis system based on artificial intelligence, characterized by: It includes several monitoring modules and a central control module, wherein the several monitoring modules are respectively connected to the central control module for mutual communication; The monitoring module is used to collect fire information in the monitoring area and preliminarily determine whether a fire has occurred. If the judgment result is yes, the monitoring module uploads the fire information to the central control module; The central control module includes a data storage module and an analysis module. Past fire information is stored in the storage module. Each time the central control module receives fire information, it compares the current fire information with the past fire information, filters out the past fire information closest to the current fire information, extracts the fire spread speed from the closest past fire information, and records it as the past fire spread speed. Each time the central control module receives fire information, it simultaneously calls the analysis module to calculate the fire spread speed, which is recorded as the first predicted fire spread speed. The central control module calculates the comprehensive fire spread speed based on the past fire spread speed and the first predicted fire spread speed, and sends an alarm to different rescue teams when the comprehensive fire spread speed exceeds different fire thresholds.

2. The fire information tracing and analysis system based on artificial intelligence according to claim 1 is characterized in that: The central control module calculates the comprehensive fire spread rate Z based on the past fire spread rate G and the first predicted fire spread rate Y, where Z=(k1×G+k2×Y) / (k1+k2), where k1 and k2 are pre-input constants.

3. The fire information tracing and analysis system based on artificial intelligence according to claim 2 is characterized in that: Any of the monitoring modules includes an edge computing module. When the monitoring module determines that a fire has occurred, it samples the fire information to obtain simplified fire information, and uploads the simplified fire information to the central control module. The central control module is used to calculate the comprehensive fire spread speed based on the simplified fire information. The monitoring module calls the edge computing module to calculate a second predicted fire spread speed based on the fire information, and uploads the second predicted fire spread speed to the central control module. The central control module takes the maximum value D between the second predicted fire spread speed or the comprehensive fire spread speed, and sends an alarm to different rescue teams when D exceeds different fire thresholds.

4. The fire information tracing and analysis system based on artificial intelligence according to claim 3 is characterized by: There are several fire thresholds preset, and the fire thresholds are set in one-to-one correspondence with rescue teams of different levels.

5. The fire information tracing and analysis system based on artificial intelligence according to claim 4 is characterized in that: The monitoring module is used to read the network delay between itself and the central control module, and reduce the number of samples in the fire information sampling process when the network delay exceeds a threshold.

6. The fire information tracing and analysis system based on artificial intelligence according to claim 5 is characterized in that: The monitoring module is used to read the network delay C between itself and the central control module, and adjust the sampling number in the fire information sampling process to X times the standard value, where X=C / C0×d, C0 is the pre-input standard network delay.

7. The fire information tracing and analysis system based on artificial intelligence according to claim 6 is characterized by: The monitoring module uploads X synchronously each time it uploads simplified fire information. The central control module is used to record the X values corresponding to the simplified fire information as the same group and synchronously store them in the storage module as past fire information.

8. The fire information tracing and analysis system based on artificial intelligence according to claim 7 is characterized in that: The invention also includes an interactive interface for displaying the comprehensive fire spread speed Z and the second predicted fire spread speed.

Images