A method, device and system for identifying safety alarm strategies in underground coal mines

Abstract

The present invention belongs to the technical field of underground safety alarms, and specifically discloses a method, device, and system for identifying underground coal mine safety alarm strategies, including collecting original alarm strategies, and abstracting the original alarm strategies to obtain alarm features; using the fields of the alarm strategy information of the alarm features as valid fields, performing semantic retrieval based on the valid fields, and listing the retrieved alarm features with the same semantics as the same family of alarms; performing alarm device comparison and alarm count comparison within the same family of alarms, and deciding whether to discard the same family of alarms based on the comparison results; and displaying the alarm features that have not been discarded on the alarm strategy display device in sequence based on the alarm priority. The present invention can reduce the problem of repeated sending of alarm strategies caused by network environment and other issues, and improve the authenticity of alarms and the accuracy of troubleshooting.

Description

Technical Field

[0001] The present invention belongs to the technical field of underground safety alarms, and in particular relates to a method, device and system for identifying safety alarm strategies in coal mines. Background Art

[0002] The underground coal mine safety monitoring system plays a vital role in coal mining. It promptly transmits safety issues underground to the ground control room, enabling personnel to promptly investigate and eliminate potential safety hazards. Under normal circumstances, the underground safety monitoring system promptly issues underground alarms. However, in certain situations, such as when the network is under attack, it may repeatedly send numerous similar alarms, making troubleshooting cumbersome and placing significant pressure on the ground control center. Summary of the Invention

[0003] The purpose of the present invention is to overcome the defect of repeated sending of alarm information in the underground safety monitoring system in the prior art, and to provide a method, device and system for identifying safety alarm strategies in coal mines.

[0004] A first aspect of the present invention provides a method for identifying safety alarm strategies in coal mines, which is applied to an alarm strategy display device and includes:

[0005] Step 1: Collect the original alarm strategy and perform abstract processing on the original alarm strategy to obtain alarm features, which at least include: alarm device information, alarm strategy information, and alarm frequency information;

[0006] Step 2: Take the fields of the alarm policy information of the alarm feature as valid fields, perform semantic search based on the valid fields, list the retrieved alarm features with the same semantics as the same family of alarms, perform alarm device comparison and alarm count comparison within the same family of alarms, and decide whether to discard the same family of alarms based on the comparison results;

[0007] Step 3: Display the alarm features that have not been discarded on the alarm strategy display device in sequence based on the alarm priority.

[0008] A further solution is that the abstract processing extracts the alarm device information, alarm strategy information, and alarm frequency information of the alarm feature based on the trained first neural network and outputs a file package recording: alarm device information, alarm strategy information, and alarm frequency information.

[0009] A further solution is that in step 2, performing alarm device comparison and alarm frequency comparison within the same family of alarms includes comparing the alarm device information of the alarm features within the same family of alarms, determining the alarm features with the same alarm device information as a repeated alarm feature set, and performing alarm frequency information comparison on the remaining alarm features within the same family of alarms. If the alarm frequency is greater than the alarm frequency threshold, it is determined to be a repeated alarm feature set; otherwise, it is determined to be a non-repeated alarm feature set.

[0010] One of the alarm features in the duplicate alarm feature set is selected and displayed on the alarm strategy display device, and the remaining alarm features in the duplicate alarm feature set are discarded.

[0011] A further solution is that in step 3, the alarm priority is:

[0012] P=X*x+Y*y+Z*z

[0013] Among them, P is the alarm priority, X is the valid field semantics, Y is the device ID, and Z is the number of alarms. x, y, and z are the weight proportions of the valid field semantics, device ID, and number of alarms, respectively.

[0014] A further solution is that the method for determining x, y, and z includes: presetting the weight ratio y of the device ID based on the type of device ID and the location of the device ID, inputting the alarm feature with valid fields into the trained second neural network, extracting the valid field semantics, outputting the weight ratio x of the valid field semantics and the weight ratio y of the device ID, and determining the weight ratio z of the number of alarms by comparing the number of alarms with the alarm frequency threshold.

[0015] A second aspect of the present invention provides a coal mine safety alarm strategy identification system, the system using the above-mentioned coal mine safety alarm strategy identification method, the system comprising:

[0016] The alarm strategy collection module abstracts the original alarm strategy to obtain alarm features, which include at least: alarm device information, alarm strategy information, and alarm frequency information;

[0017] The alarm strategy analysis module is used to sort and number alarm features according to the collection time, use the alarm strategy information field of the alarm feature as the valid field, perform semantic search based on the valid field, list the retrieved alarm features with the same semantics as the same family alarms, compare the alarm devices and the number of alarms within the same family, and decide whether to discard the same family alarms based on the comparison results;

[0018] The alarm strategy display module is used to display the alarm features that have not been discarded on the alarm strategy display device in sequence based on the alarm priority.

[0019] The third aspect of the present invention provides a coal mine underground safety alarm strategy identification device, which is equipped with the above-mentioned coal mine underground safety alarm strategy identification system. The device includes a first neural network for abstractly processing the original alarm strategy, and a second neural network for determining the priority of the alarm.

[0020] Compared with the prior art, the present invention has the following beneficial effects:

[0021] The present invention abstracts a large number of irregular alarm strategies based on a trained neural network to obtain alarm features. The alarm features are then compared and judged to distinguish between repetitive alarm feature sets and non-repetitive alarm feature sets. One alarm feature in the repetitive alarm feature set is then selected to form a new set to be displayed together with the alarm features in the non-repetitive alarm feature set. The set to be displayed is then displayed on the alarm strategy display device in sequence according to the alarm priority, and the remaining alarm features in the repetitive alarm feature set are discarded. This can reduce the problem of repeated alarm strategy transmission caused by network environment and other issues, improve the authenticity of alarms and the accuracy of problem investigation. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The following drawings are merely provided for illustrative purposes only and are not intended to limit the scope of the present invention.

[0023] Figure 1 : Flowchart of the alarm strategy identification method of the present invention;

[0024] Figure 2 : Alarm strategy analysis method flow chart of the present invention. DETAILED DESCRIPTION

[0025] In order to make the purpose, technical solution, design method and advantages of the present invention more clear, the present invention is further described in detail below through specific embodiments in conjunction with the accompanying drawings. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0026] Example 1

[0027] The present invention provides a method for identifying safety alarm strategies in coal mines. Figure 1 and Figure 2 As shown, the following steps are included:

[0028] S1. Collect the original alarm strategy and abstract the original alarm strategy into an alarm feature with alarm device information, alarm strategy information, and alarm number information;

[0029] S2. Using the fields of the alarm strategy information of the alarm feature as valid fields, performing semantic search based on the valid fields, and listing the retrieved alarm features with the same semantics as the same family of alarms;

[0030] S3. Compare the alarm devices and alarm times within the same alarm group, and decide whether to discard the same alarm group based on the comparison results;

[0031] S4. Display the alarm features that have not been discarded on the alarm strategy display device in sequence based on the alarm priority.

[0032] Among them, the abstract processing of S1 extracts the alarm features: alarm device information, alarm strategy information, and alarm number information based on the trained first neural network and outputs a file package recording: alarm device information, alarm strategy information, and alarm number information.

[0033] In S2, alarm device comparison and alarm frequency comparison are performed within the same family of alarms, including comparing the alarm device information of the alarm features within the same family of alarms, determining the alarm features with the same alarm device information as a repeated alarm feature set, and performing alarm frequency information comparison on the remaining alarm features within the same family of alarms. If the alarm frequency is greater than the alarm frequency threshold, it is determined to be a repeated alarm feature set, otherwise it is determined to be a non-repetitive alarm feature set; one of the alarm features in the repeated alarm feature set is selected to form a set to be displayed together with the non-repetitive alarm feature set, and the remaining alarm features in the repeated alarm feature set are displayed on the alarm strategy display device and discarded.

[0034] In the above, the alarm priority calculation formula in S3 is:

[0035] P=X*x+Y*y+Z*z

[0036] Among them, P is the alarm priority, X is the valid field semantics, Y is the device ID, and Z is the number of alarms. x, y, and z are the weight proportions of the valid field semantics, device ID, and number of alarms, respectively.

[0037] The method for determining x, y, and z includes: presetting the weight ratio y of the device ID based on the type of device ID and the location of the device ID, inputting the alarm features with valid fields into the trained second neural network, extracting the valid field semantics, outputting the weight ratio x of the valid field semantics and the weight ratio y of the device ID, and determining the weight ratio z of the number of alarms by comparing the number of alarms with the alarm frequency threshold.

[0038] During implementation, it is assumed that the coal mine safety monitoring system sends n alarm strategies to the ground control room within a unit of time, and these n alarm strategies come from m devices underground. First, collect n alarm strategies, input the n alarm strategies into the trained first neural network to extract the alarm device information, alarm strategy information, and alarm frequency information of the alarm strategy, and output a file package containing: alarm device information, alarm strategy information, and alarm frequency information. The n alarm strategies correspond to the output of n file packages, and the alarm strategy information of each file package is used as a valid field for semantic retrieval to obtain different alarm strategy types: Fault 1, Fault 2, and Fault 3. Among them, the file package corresponding to Fault 1 contains a (alarms of the same family), the file package corresponding to Fault 2 contains b, and the file package corresponding to Fault 3 contains c. In order to facilitate the explanation of the technical solution of the present invention, the fault type is set to 3. Then, a + b + c = n. Alarm device and alarm count comparisons are performed on a alarm signature within the same alarm family, b alarm signatures within the same alarm family, and c alarm signatures within the same alarm family. If a1 identical alarm device information exists among a alarm signatures, meaning that a1 alarm signatures originate from the same device, these a1 alarm signatures are considered a duplicate alarm signature set. The remaining a2 alarm signatures are then compared for alarm count information. If the alarm count within a unit time exceeds the alarm frequency threshold, the set is considered duplicate; otherwise, it is considered a non-duplicate alarm signature set. Based on this, the alarm signature analysis is complete, and all alarm signatures are divided into duplicate and non-duplicate alarm signature sets. One alarm signature is randomly selected from the duplicate alarm signature set and combined with the non-duplicate alarm signature set to form the set to be displayed. The remaining alarm signatures within the set to be displayed are displayed on the alarm strategy display device in order based on the alarm priority.

[0039] Example 2

[0040] This embodiment provides a coal mine safety alarm strategy identification system, characterized in that the system uses the coal mine safety alarm strategy identification method described in Example 1 to complete alarm strategy screening and display, and the system includes:

[0041] The alarm strategy collection module abstracts the original alarm strategy to obtain alarm features, which include at least: alarm device information, alarm strategy information, and alarm frequency information;

[0042] The alarm strategy analysis module is used to sort and number alarm features according to the collection time, use the alarm strategy information field of the alarm feature as the valid field, perform semantic search based on the valid field, list the retrieved alarm features with the same semantics as the same family alarms, compare the alarm devices and the number of alarms within the same family, and decide whether to discard the same family alarms based on the comparison results;

[0043] The alarm strategy display module is used to display the alarm features that have not been discarded on the alarm strategy display device in sequence based on the alarm priority.

[0044] Example 3

[0045] The present embodiment provides a device for identifying coal mine safety alarm strategies, which is provided with a coal mine safety alarm strategy identification system as described in Example 2, and the device includes a first neural network for abstractly processing the original alarm strategy, and a second neural network for determining the priority of the alarm situation. The abstract processing is based on the trained first neural network to extract the alarm characteristics: alarm device information, alarm strategy information, and alarm number information and output a file package containing: alarm device information, alarm strategy information, and alarm number information; when calculating the priority, the weight ratio y of the device ID is preset by the type of device ID and the location where it is located, the alarm feature with valid fields is input into the trained second neural network, the valid field semantics are extracted, the weight ratio x of the valid field semantics and the weight ratio y of the device ID are output, the weight ratio z of the alarm number is determined by comparing the alarm number with the alarm frequency threshold, and the formula is used.

[0046] P=X*x+Y*y+Z*z

[0047] Calculate the alarm priority, where P is the alarm priority, X is the valid field semantics, and Y is the device ID. The larger the P value, the higher the priority.

[0048] While various embodiments of the present invention have been described above, the above descriptions are intended to be illustrative, non-exhaustive, and not limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is selected to best explain the principles of the embodiments, their practical applications, or technological improvements in the marketplace, or to enable others skilled in the art to understand the embodiments disclosed herein.

Claims

1. A method for identifying safety alarm strategies in coal mines, applied to an alarm strategy display device, characterized in that: include: Step 1: Collect the original alarm strategy and perform abstract processing on the original alarm strategy to obtain alarm features, which at least include: alarm device information, alarm strategy information, and alarm frequency information; Step 2: Take the fields of the alarm policy information of the alarm feature as valid fields, perform semantic search based on the valid fields, list the retrieved alarm features with the same semantics as the same family of alarms, perform alarm device comparison and alarm count comparison within the same family of alarms, and decide whether to discard the same family of alarms based on the comparison results; Step 3: Display the alarm features that have not been discarded on the alarm strategy display device in order based on the alarm priority; In step 2, performing alarm device comparison and alarm frequency comparison within the same family of alarms includes comparing the alarm device information of the alarm features within the same family of alarms, determining the alarm features with the same alarm device information as a repeated alarm feature set, and performing alarm frequency information comparison on the remaining alarm features within the same family of alarms. If the alarm frequency is greater than the alarm frequency threshold, it is determined to be a repeated alarm feature set, otherwise it is determined to be a non-repeated alarm feature set; One of the alarm features in the repeated alarm feature set is selected to form a to-be-displayed set together with the non-repeated alarm feature set, and the remaining alarm features in the repeated alarm feature set are displayed on the alarm strategy display device and discarded.

2. A method for identifying underground coal mine safety alarm strategies according to claim 1, characterized in that: The abstract processing extracts alarm features based on the trained first neural network and outputs a file package recording alarm device information, alarm strategy information, and alarm number information.

3. A method for identifying underground coal mine safety alarm strategies according to claim 2, characterized in that: In step 3, the alarm priority calculation formula is: ； Among them, P is the alarm priority, X is the valid field semantics, Y is the device ID, and Z is the number of alarms. x, y, and z are the weight proportions of the valid field semantics, device ID, and number of alarms, respectively.

4. A method for identifying underground coal mine safety alarm strategies according to claim 3, characterized in that: The method for determining x, y, and z includes: presetting the weight ratio y of the device ID based on the type of device ID and the location of the device ID, inputting the alarm feature with valid fields into the trained second neural network, extracting the valid field semantics, outputting the weight ratio x of the valid field semantics and the weight ratio y of the device ID, and determining the weight ratio z of the number of alarms by comparing the number of alarms with the alarm frequency threshold.

5. A coal mine safety alarm strategy identification system, characterized in that: The system uses a coal mine underground safety alarm strategy identification method according to any one of claims 1 to 4 to complete the screening and display of alarm strategies, and the system includes: The alarm strategy collection module abstracts the original alarm strategy to obtain alarm features, which include at least: alarm device information, alarm strategy information, and alarm frequency information; The alarm strategy analysis module is used to sort and number alarm features according to the collection time, use the alarm strategy information field of the alarm feature as the valid field, perform semantic search based on the valid field, list the retrieved alarm features with the same semantics as the same family alarms, compare the alarm devices and the number of alarms within the same family, and decide whether to discard the same family alarms based on the comparison results; The alarm strategy display module is used to display the alarm features that have not been discarded on the alarm strategy display device in sequence based on the alarm priority.

6. A coal mine safety alarm strategy identification device, characterized in that: A coal mine safety alarm strategy identification system as described in claim 5 is provided, wherein the device includes a first neural network for abstractly processing the original alarm strategy and a second neural network for determining the priority of the alarm situation.

Images