Mine underground emergency alarm method and system

Abstract

The invention belongs to the field of emergency warning of an energy exploitation site, and particularly relates to a mine underground emergency warning method and system. Comprising the following steps: inputting collected data containing current environment monitoring information, underground personnel positioning information, equipment state monitoring information and video monitoring information into a trained real-time emergency alarm model as input data, obtaining an emergency alarm result containing a result of judging whether the current underground condition corresponding to each piece of input data has a risk or not, and starting a corresponding real-time emergency scheme according to the emergency alarm result; the emergency alarm model is trained by adopting a training set of data samples corresponding to historical data including environment monitoring information, underground personnel positioning information, equipment state monitoring information and video monitoring information. In this way, the real underground situation can be accurately and reliably reflected from multiple dimensions, and then it is guaranteed that an emergency scheme obtained according to the emergency warning result processed in the mode can accurately and reliably solve the current real underground risk.

Description

Technical Field

[0001] The present invention belongs to the field of emergency alarms at energy mining sites, and in particular relates to an emergency alarm method and system for underground mines. Background Art

[0002] A mine shaft is a passage that is dug underground and leads directly to the ground for mining. It is an important part of the underground development system of a mine. The construction and use of mine shafts can greatly improve the efficiency of ore mining and transportation and reduce production costs.

[0003] However, the underground working environment of mines is complex and there are many potential safety hazards, such as gas explosion, coal dust explosion, roof collapse and water seepage. Once these accidents occur, they often cause serious casualties and property losses. Therefore, through the underground mine safety monitoring system, the underground environmental parameters (such as gas concentration, oxygen content, temperature, humidity, etc.) and equipment operation status can be monitored in real time, abnormal conditions can be discovered in time and alarms can be issued, so as to effectively avoid or reduce the occurrence of accidents and ensure the safety of personnel lives.

[0004] At present, with the continuous advancement of various technologies, the safety monitoring system of underground mines has also made significant progress. However, there are still many problems in realizing the integration of multiple systems underground. First of all, if you want to monitor the environment underground in the mine from multiple angles to achieve more accurate emergency alarms, you need to collect multiple underground environmental parameters for risk assessment (that is, you cannot be limited to risk assessment based on individual types of parameters, but should collect more different types of parameters to conduct risk assessment on the environment underground in the mine). A single manufacturer can usually only provide a system that considers individual parameters and gives emergency plans. For example, the system provided by Manufacturer A specifically provides emergency plans for risks caused by abnormal gas concentrations, while the system provided by Manufacturer B specifically provides emergency plans for risks caused by abnormal temperature and humidity.

[0005] If the systems of manufacturers A and B are directly integrated to obtain an emergency plan that comprehensively considers the three parameters of gas concentration, temperature and humidity, the emergency plan given by system A may be incompatible with the emergency plan given by system B, that is, system A and system B may give two completely different plans; and because only one of the two plans can usually be chosen, it is ultimately difficult to take both emergency plans into account and adopt them. If the two plans are combined blindly, the wrong response measures may be given because the combined plan does not match the actual risk situation underground.

[0006] In summary, if we want to use the collected multi-dimensional data to characterize underground risks, we can only obtain emergency plans corresponding to the data through separate real-time monitoring and corresponding risk analysis of individual data, and it is difficult to conduct a comprehensive assessment of underground risks reflected by different data. In addition, the emergency plan for accidents is still in the manual management stage, and it is impossible to accurately and quickly generate a comprehensive real-time emergency plan that fully fits the current accident situation, which affects the accuracy and effectiveness of emergency plan decision-making, thereby affecting the speed of emergency response of the mine underground safety monitoring system and the accuracy and reliability of risk judgment and monitoring alarms. Summary of the invention

[0007] The purpose of the present invention is to provide an emergency alarm method and system for underground mines, which is used to solve the problem that when multidimensional data from different sources are used to characterize underground risks in the prior art, it is impossible to accurately and quickly generate a comprehensive real-time emergency plan that fully fits the current accident situation, thereby affecting the accuracy and effectiveness of emergency plan decisions.

[0008] In order to achieve the above-mentioned purpose, the present invention provides an emergency alarm method for underground mines, the method comprising: taking the collected data containing current environmental monitoring information, underground personnel positioning information, equipment status monitoring information and video monitoring information as input data, inputting it into a trained real-time emergency alarm model, obtaining an emergency alarm result containing a judgment result of whether there is a risk in the current underground situation corresponding to each input data, and initiating a corresponding real-time emergency plan according to the emergency alarm result; the real-time emergency alarm model is trained using a training set of data samples corresponding to historical data containing environmental monitoring information, underground personnel positioning information, equipment status monitoring information and video monitoring information.

[0009] Beneficial effects: The present invention provides a new emergency alarm method for underground mines. The method first establishes an emergency alarm model, and then uses the collected environmental monitoring information, underground personnel positioning information, equipment status monitoring information, video monitoring information and other data as input data, and inputs them into the trained emergency alarm model to obtain an emergency alarm result. According to the emergency alarm result, the corresponding emergency plan (that is, the emergency plan specially designed for the emergency alarm result) is initiated. The emergency alarm result includes a determination result on whether the underground situation reflected by each type of information poses a risk; for example, in a certain emergency alarm result, only the environmental monitoring information is determined to be at risk, and other information is not determined to be at risk. In this case, the emergency plan corresponding to the emergency alarm result only provides corresponding solutions or countermeasures when there is a risk in the environmental monitoring information and no risk in other information.

[0010] This method can use the emergency warning model to uniformly process the information from different sources that can reflect the real situation underground through a neural network, so as to integrate and analyze the information from different sources and provide a set of emergency warning solutions that coordinate various risks (the risks corresponding to the real situation underground reflected by the information from different sources mentioned above). That is, it can accurately and reliably reflect the real situation underground from multiple dimensions, thereby ensuring that the emergency solution obtained according to the emergency warning results processed in this way can accurately and reliably solve the current real underground risks.

[0011] Furthermore, it also includes: using the collected data including current environmental monitoring information, underground personnel positioning information, equipment status monitoring information and video monitoring information as input data, inputting them into the trained emergency warning model, and obtaining prediction results including future changes of each underground environmental parameter corresponding to the input data; obtaining emergency warning results according to the prediction results, and initiating corresponding emergency plans according to the emergency warning results; the methods for obtaining emergency warning results according to the prediction results include: If the prediction result contains a future change of a downhole environmental parameter that satisfies the condition that, starting from the present, it will fall into the set alarm range corresponding to the downhole environmental parameter within the set change time, then an emergency warning result including the risk of the future downhole situation corresponding to the downhole environmental parameter is obtained; If the future change of a certain downhole environmental parameter in the prediction result satisfies that, starting from the present, the amplitude of the increase or decrease of the value within the set mutation time exceeds the corresponding set amplitude range, an emergency warning result indicating the risk of future downhole conditions corresponding to the downhole environmental parameter is obtained.

[0012] Furthermore, the method of collecting data including environmental monitoring information, underground personnel positioning information, equipment status monitoring information and video monitoring information includes: collecting environmental monitoring information through the set environmental monitoring subsystem; collecting the location information of underground personnel through the set personnel positioning subsystem; collecting the operating status of key equipment through the set equipment status monitoring subsystem; collecting video monitoring information through the set video monitoring subsystem.

[0013] Furthermore, the method of initiating the corresponding real-time emergency plan according to the emergency alarm result includes: triggering the real-time emergency plan corresponding to the emergency alarm result according to different alarm signals corresponding to different emergency alarm results through the set emergency command subsystem.

[0014] Furthermore, the method of initiating the corresponding emergency plan according to the emergency warning result includes: triggering the emergency plan corresponding to the emergency warning result according to different warning signals corresponding to different emergency warning results through the set emergency command subsystem.

[0015] Furthermore, both the real-time emergency warning model and the emergency early warning model are trained through unsupervised learning algorithms.

[0016] Furthermore, the unsupervised learning algorithm is a One-Class SVM algorithm.

[0017] Furthermore, the method of collecting environmental monitoring information through the set environmental monitoring subsystem includes: collecting concentration information of toxic gases to be monitored through the set gas monitoring sensors; collecting temperature and humidity information of the downhole environment through the set temperature and humidity sensors; collecting concentration information of dust particles in the downhole air through the set dust sensors; collecting air pressure information through the set pressure sensors; and collecting sound information of ground vibration or rock fracture through the set vibration sensors or sound wave sensors.

[0018] The present invention also provides an emergency alarm system for an underground mine, comprising a processor, wherein the processor is used to execute a computer program to implement the steps of the above-mentioned emergency alarm method for an underground mine.

[0019] The underground mine emergency alarm system can achieve the same beneficial effects as the above-mentioned underground mine emergency alarm method. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 This is a system application example diagram of the emergency alarm method for underground mines in the embodiment of the emergency alarm method for underground mines of the present invention. DETAILED DESCRIPTION

[0021] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention is further described in detail below with reference to the accompanying drawings and embodiments.

[0022] Embodiment of emergency warning method in mine This embodiment provides a technical solution for an emergency alarm method in a mine. The method processes the collected information that can reflect the actual situation underground through a pre-established and trained emergency alarm model, and outputs an emergency alarm result for avoiding or solving risks. The model can make a comprehensive judgment on the above information that can reflect the actual situation underground to obtain a judgment result. The emergency alarm result output after processing by the model includes the judgment results corresponding to each piece of information, so that the corresponding emergency alarm scheme is finally obtained according to the emergency alarm result. The emergency alarm result can accurately and reliably reflect the various risks currently existing. The emergency measures in the corresponding emergency alarm scheme are also determined by evaluating the current underground risks from various dimensions, rather than judging the risks reflected by a single piece of information, so as to avoid the situation where the risk assessment is biased due to too few consideration dimensions. The scheme output by the model can ensure that the emergency measures it provides are sufficiently accurate and reliable for the current risks, thereby improving the speed of emergency response and the accuracy and reliability of accident handling effects of the mine underground safety monitoring system to judge risks and monitor alarms.

[0023] The method in this embodiment specifically includes: analyzing and processing different information reflecting the current underground situation collected and uploaded by the environment monitoring subsystem, personnel positioning subsystem, equipment status monitoring subsystem and video monitoring subsystem through a data analysis and early warning subsystem (an emergency alarm model is set up in the subsystem) to provide an emergency alarm plan for the current underground risk.

[0024] In this embodiment, the collected data including current environmental monitoring information, underground personnel positioning information, equipment status monitoring information and video monitoring information are used as input data and input into a trained real-time emergency warning model (the model is set in the data analysis and early warning subsystem), and an emergency warning result including a judgment result of whether there is a risk in the current underground situation corresponding to each input data is obtained, and the corresponding real-time emergency plan is initiated according to the emergency warning result. The collected information can display in detail the personnel location (i.e., underground personnel positioning information), equipment status monitoring information, environmental parameters (i.e., environmental monitoring information), video monitoring information (including real-time video and historical video), historical change trends and historical action trajectories, etc. Figure 1 As shown, the method of collecting the above information is to collect environmental monitoring information through the set environmental monitoring subsystem; collect the location information of underground personnel through the set personnel positioning subsystem; collect the operating status of key equipment through the set equipment status monitoring subsystem; and collect video monitoring information through the set video monitoring subsystem).

[0025] The above-mentioned real-time emergency warning model uses a training set of data samples corresponding to historical data including environmental monitoring information, underground personnel positioning information, equipment status monitoring information and video monitoring information, and trains the real-time emergency warning model through an unsupervised learning algorithm. In this embodiment, the unsupervised learning algorithm is a One-Class SVM algorithm. In other embodiments, other existing unsupervised algorithms or supervised algorithms can also be used for training, which will not be described in detail here.

[0026] In this embodiment, the above-mentioned method of collecting environmental monitoring information through the set environmental monitoring subsystem includes: 1) Collect the concentration information of the toxic gas to be monitored (such as methane, carbon dioxide, hydrogen sulfide, etc.) through the set gas monitoring sensor; 2) Collect the temperature and humidity information of the underground environment through the set temperature and humidity sensors; 3) Collect the concentration information of dust particles in the underground air through the dust sensor; 4) Collecting air pressure information through the set pressure sensor helps to judge the ventilation conditions; 5) Vibration sensors or acoustic sensors are installed to collect sound information of ground vibration or rock fracture to warn of geological disasters.

[0027] In this embodiment, in addition to collecting environmental monitoring information, other information that can reflect the actual situation underground also needs to be collected, including underground personnel positioning information, equipment status monitoring information, and video monitoring information.

[0028] The specific method of collecting underground personnel positioning information through the above-mentioned personnel positioning subsystem is: equip each staff member (i.e. underground personnel, whose work needs require them to go down the mine) with a helmet with a UWB tag, which can send location information with a timestamp. At the same time, UWB base stations corresponding to UWB tags are deployed at key locations underground to receive signals from the tags and calculate the time difference or flight time. Through the collaboration between the base stations, the specific location of the tags is calculated using triangulation. The signal source location is determined by converting the time difference between the receivers into a distance difference and using geometric methods to construct multiple hyperbolic intersections.

[0029] In this embodiment, the specific method of collecting device status monitoring information through the device status monitoring subsystem is: a) Install corresponding vibration sensors for key equipment to monitor the vibration of the equipment and identify potential mechanical failures.

[0030] b) Set up corresponding temperature sensors for key equipment to monitor the temperature of the equipment during operation and prevent failures caused by overheating; c) Set up corresponding pressure sensors for key equipment to monitor pressure changes in the hydraulic system to ensure the normal operation of the hydraulic equipment; d) Set up corresponding location sensors for key equipment to monitor the movement of the equipment; this method is suitable for location tracking of mobile devices.

[0031] In this embodiment, the specific method of collecting video surveillance information through the video surveillance subsystem is: A) Install high-definition cameras in key areas underground; B) Use the video analysis module to analyze the collected video images in real time to identify abnormal situations; C) The data acquisition module is used to collect video data and transmit it to the central processing unit through the wireless network.

[0032] In this embodiment, the information that can reflect the actual situation underground collected by the above-mentioned environment monitoring subsystem, personnel positioning subsystem, equipment status monitoring subsystem and video monitoring subsystem is successfully and stably transmitted to the central processing unit through the data acquisition module (this module is responsible for collecting sensor data and performing preliminary processing; this module adopts wireless communication technologies such as LoRa, ZigBee, Wi-Fi or wired communication technologies such as optical fiber and cable with a variety of data transmission schemes to ensure the stability of data transmission).

[0033] When the central processing unit receives the information summarized by the data acquisition module, it first processes and analyzes it, and transmits the processing results to the emergency alarm model set in the data analysis and early warning subsystem for further processing.

[0034] In addition to using the emergency warning model for real-time emergency warning, this embodiment also includes: using the collected data containing current environmental monitoring information, underground personnel positioning information, equipment status monitoring information and video monitoring information as input data, inputting it into the trained emergency warning model (in this embodiment, the model and the emergency warning model are also set in the data analysis and warning subsystem; in other embodiments, it can also be set in other modules or devices), and obtaining the prediction results containing the future changes of each underground environmental parameter corresponding to the input data (that is, through the set data analysis module to analyze the trend and periodic characteristics of the time series data, predict possible problems in the future, and predict the future state of the equipment. Such as the vibration level of the motor, the temperature of the bearing, etc.; analyze the changing trend of the gas concentration, and analyze and predict the future trend). The emergency warning result is obtained according to the prediction result, and the corresponding emergency plan is activated according to the emergency warning result. The method in this embodiment includes: through the set emergency command subsystem, according to the different warning signals corresponding to different emergency warning results, the emergency plan corresponding to the emergency warning result is triggered.

[0035] Specifically, the above-mentioned methods of obtaining emergency warning results based on the prediction results include: ① Judge by setting the alarm range: If the future change of a certain underground environmental parameter in the prediction result satisfies the condition that it will fall into the set alarm range corresponding to the underground environmental parameter within the set change time from now on, then an emergency warning result including the risk of future underground conditions corresponding to the underground environmental parameter is obtained. For example, when the temperature of the equipment inside the mine continues to rise slowly, although it has not reached the alarm point (that is, the preset temperature alarm threshold), through the analysis and calculation of the temperature rise trend, it can be predicted that at a certain subsequent time point (or after a certain period of time), the temperature of the equipment inside the mine will exceed the temperature alarm threshold. Therefore, the above method can achieve early prediction of some risks in the mine.

[0036] ② Judge by setting the amplitude range: If the future change of a certain underground environmental parameter in the prediction result satisfies that from now on, the amplitude of the increase or decrease of the value within the set mutation time exceeds the corresponding set amplitude range, then an emergency warning result containing the underground environmental parameter corresponding to the future underground situation with risks is obtained. For example, when the concentration of a certain gas (such as hydrogen or carbon monoxide) in the mine changes from 0 to a larger value, although it does not reach the alarm point (that is, the preset gas concentration alarm threshold), due to the large change in the gas concentration, it can be preliminarily judged that the equipment is likely to be abnormal. Therefore, the above method can achieve early prediction of some risks in the mine.

[0037] In this embodiment, the emergency plan output by the emergency alarm model and the emergency plan output by the emergency warning are both initiated through the emergency command subsystem.

[0038] Specifically, the method of initiating the corresponding emergency plan according to the emergency alarm result includes: triggering the real-time emergency plan corresponding to the emergency alarm result according to different alarm signals corresponding to different emergency alarm results through the set emergency command subsystem.

[0039] The method of initiating the corresponding emergency plan according to the emergency warning result includes: through the set emergency command subsystem, according to different warning signals corresponding to different emergency warning results, triggering the emergency plan corresponding to the emergency warning result.

[0040] In summary, this embodiment provides an emergency plan for the current underground risk situation through a trained emergency warning model; and provides an early warning plan for future risks through a trained emergency warning model; that is, this method combines the above two models to achieve a comprehensive assessment of underground risk situations and the provision of plans. In order to accurately and quickly generate a comprehensive real-time emergency plan that fully fits the current accident situation, the mine underground safety monitoring system can perform an accurate and reliable emergency response to ensure a good handling effect on the current risk.

[0041] Embodiment of emergency warning system in mine This embodiment provides a technical solution for an emergency alarm system in an underground mine. The system includes a processor having executable program instructions stored therein. The executable program instructions are used to implement the emergency alarm method in an underground mine as described in the above-mentioned embodiment of the emergency alarm method in an underground mine.

[0042] Since the specific working mode and working principle of the emergency alarm system in the mine of this embodiment have been described in detail in the above-mentioned embodiment of the emergency alarm method in the mine, they will not be repeated here.

[0043] It should be understood that the above specific embodiments of the present invention are only used to illustrate or explain the principles of the present invention, but do not constitute a limitation to the present invention.

Claims

1. An emergency alarm method for underground mines, characterized in that: include: The collected data including current environmental monitoring information, underground personnel location information, equipment status monitoring information and video monitoring information are used as input data and input into the trained real-time emergency warning model to obtain an emergency warning result including a determination result of whether there is a risk in the current underground situation corresponding to each input data, and the corresponding real-time emergency plan is initiated according to the emergency warning result; The real-time emergency warning model is trained using a training set of data samples corresponding to historical data including environmental monitoring information, underground personnel positioning information, equipment status monitoring information and video monitoring information.

2. The mine underground emergency alarm method according to claim 1, characterized in that: Also includes: The collected data including current environmental monitoring information, underground personnel location information, equipment status monitoring information and video monitoring information are used as input data and input into the trained emergency warning model to obtain the prediction results of the future changes of each underground environmental parameter corresponding to the input data; Obtaining an emergency warning result according to the prediction result, and initiating a corresponding emergency plan according to the emergency warning result; Methods for obtaining emergency warning results according to the prediction results include: If the prediction result contains a future change of a downhole environmental parameter that satisfies the condition that, starting from the present, it will fall into the set alarm range corresponding to the downhole environmental parameter within the set change time, then an emergency warning result including the risk of the future downhole situation corresponding to the downhole environmental parameter is obtained; If the future change of a certain downhole environmental parameter in the prediction result satisfies that, starting from the present, the amplitude of the increase or decrease of the value within the set mutation time exceeds the corresponding set amplitude range, an emergency warning result indicating the risk of future downhole conditions corresponding to the downhole environmental parameter is obtained.

3. The mine underground emergency alarm method according to claim 1 or 2, characterized in that: The method of collecting data including environmental monitoring information, underground personnel positioning information, equipment status monitoring information and video monitoring information includes: collecting environmental monitoring information through the set environmental monitoring subsystem; collecting the location information of underground personnel through the set personnel positioning subsystem; collecting the operating status of key equipment through the set equipment status monitoring subsystem; collecting video monitoring information through the set video monitoring subsystem.

4. The mine underground emergency alarm method according to claim 1 or 2, characterized in that: The method of starting the corresponding real-time emergency plan according to the emergency alarm result includes: triggering the real-time emergency plan corresponding to the emergency alarm result according to different alarm signals corresponding to different emergency alarm results through the set emergency command subsystem.

5. The emergency alarm method for underground mines according to claim 2, characterized in that: The method of initiating the corresponding emergency plan according to the emergency warning result includes: triggering the emergency plan corresponding to the emergency warning result according to different warning signals corresponding to different emergency warning results through the set emergency command subsystem.

6. The mine underground emergency alarm method according to claim 1 or 2, characterized in that: Both the real-time emergency alert model and the emergency warning model are trained using unsupervised learning algorithms.

7. The mine underground emergency alarm method according to claim 6, characterized in that: The unsupervised learning algorithm is a One-Class SVM algorithm.

8. The mine underground emergency alarm method according to claim 3, characterized in that: The method of collecting environmental monitoring information through the set environmental monitoring subsystem includes: collecting concentration information of toxic gases to be monitored through the set gas monitoring sensor; collecting temperature and humidity information of the underground environment through the set temperature and humidity sensors; collecting concentration information of dust particles in the underground air through the set dust sensor; collecting air pressure information through the set pressure sensor; collecting sound information of ground vibration or rock cracking through the set vibration sensor or sound wave sensor.

9. An emergency warning system for underground mines, comprising a processor, characterized in that: The processor is used to execute a computer program to implement the steps of the emergency alarm method for an underground mine as described in any one of claims 1-8.

Images