Coal and gas outburst danger area identification and early warning method based on multi-parameter fusion

Abstract

The invention relates to the technical field of coal mine safety and coal and gas prediction, and discloses a coal and gas outburst danger area identification and early warning method based on multi-parameter fusion, and the method comprises the steps: obtaining multi-parameter data through a sensor network, connecting the sensor network through a monitoring transmission strategy, and transmitting the multi-parameter data to a cloud platform; extracting features of the multi-parameter data transmitted to the cloud platform through a feature fusion strategy, and performing fusion processing; an outburst early warning model and an outburst early warning strategy are constructed according to the multi-parameter characteristics, and the outburst early warning model is trained by obtaining historical outburst events and normal events, so that the prediction accuracy and reliability of coal and gas outburst danger are improved; identifying an abnormal state through the outburst early warning model and the outburst early warning strategy; the signals and the signal areas are visually displayed in the cloud platform and are transmitted to management personnel and professional analysts at the same time, and remote analysis and prediction effect judgment are carried out; and dynamically adjusting the highlighting early warning model and the highlighting early warning strategy according to the multi-parameter data and the prediction effect.

Description

Technical Field

[0001] The invention relates to the technical field of coal mine safety and coal and gas prediction, and in particular to a coal and gas outburst danger zone identification and early warning method based on multi-parameter fusion. Background Art

[0002] At present, researchers collect data through various sensors and conduct comprehensive analysis on the data collected by sensors, so as to predict and warn the dangerous areas of coal and gas outburst, which has important practical significance in reducing the risk of coal mine disasters and improving the safety of miners. The research results show that no matter how large the ground stress is, if it is in a static and constant state, it will not trigger outburst. The gas, which mainly exists in an adsorbed state, does not have the energy to drive the outburst. At the same time, when the mining method and mining intensity have changed greatly, the stress tension of the working face will be greatly aggravated, and there will not be enough space and time for the concentrated stress of the working face to relieve the stress tension. The direct cause of the outburst does not depend on the gas in the coal seam, but on whether the concentration of ground stress can destroy the roof rock that plays the main bearing role and generate a strong impact force. Therefore, realizing remote early warning monitoring, analysis, guidance and standardized management of the working face is a very effective way and method to solve and prevent outburst accidents. However, most of them have not solved how to fuse the acquired multi-parameter data and extract features to build early warning models and early warning strategies to identify mine anomalies, generate early warning signals and early warning danger zones and visualize them, and dynamically adjust the early warning models and early warning strategies according to the early warning effect.

[0003] For example, a Chinese patent with publication number CN113374525A discloses a comprehensive early warning method for identifying coal and gas outburst hazard zones based on multi-parameter data fusion. This method evaluates the outburst hazard of the coal seam working face with conventional indicators that can be quantitatively characterized, assigns weight coefficients to the conventional indicators, and then introduces basic information used for qualitative characterization as comprehensive indicator coefficients. The quantitative conventional indicators and the qualitative multi-parameter basic information are integrated to achieve comprehensive early warning. This early warning method can describe the consistency of the actual situation in the monitoring area, reflect the essence of things in a more realistic and comprehensive manner, make up for the shortcomings of conventional indicator early warnings, and provide more accurate early warnings for the identification of coal and gas outburst hazard zones at coal mine sites.

[0004] For example, a Chinese patent with authorization announcement number CN110043317B discloses a multi-parameter local danger zone identification and early warning method for mine disasters, which can accurately identify the location and danger level of local danger zones, can guide mines to carry out disaster protection in a targeted manner, and can meet the on-site needs of each mine. The above-mentioned early warning method includes step 1, spatial positioning of mine disaster monitoring points; step 2, obtaining the initial danger index I single point initial of each monitoring point; step 3, retrieving the conditions of other monitoring points around each monitoring point; step 4, correcting the initial danger index of the monitoring point; step 5, obtaining the location and danger level of the local danger zone. It solves the problem that the existing early warning method cannot accurately identify the location and danger level of the local danger zone, and cannot guide mines to carry out disaster protection in a targeted manner.

[0005] The above patents have the problems raised by the background technology: the above coal and gas outburst hazard zone identification comprehensive early warning method evaluates the outburst hazard of the coal seam working face through conventional indicators that can be quantitatively characterized, and assigns weight coefficients to the conventional indicators, and then introduces the basic information used for qualitative characterization as the comprehensive indicator coefficient, and integrates the conventional indicators of quantitative characterization and the multi-parameter basic information of qualitative characterization to achieve comprehensive early warning; the above mine disaster multi-parameter local hazard zone identification early warning method obtains the initial hazard index I single point initial of each monitoring point by spatially locating the mine disaster monitoring point, and retrieves the situation of other monitoring points around each monitoring point, and corrects the initial hazard index of the monitoring point to obtain the location and hazard level of the local hazard zone. The above two patents do not solve the problem of how to fuse the acquired multi-parameter data and extract features to construct an early warning model and early warning strategy to identify mine anomalies, generate early warning signals and early warning hazard zones and visualize them, and dynamically adjust the early warning model and early warning strategy according to the early warning effect. To solve this problem, the present invention proposes a coal and gas outburst hazard zone identification early warning method based on multi-parameter fusion. Summary of the invention

[0006] The purpose of this section is to summarize some aspects of the embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section and the specification abstract and the invention title of this application to avoid blurring the purpose of this section, the specification abstract and the invention title, and such simplifications or omissions cannot be used to limit the scope of the present invention.

[0007] In view of the above-mentioned problems existing in the existing coal and gas outburst danger zone identification and early warning method based on multi-parameter fusion, the present invention is proposed.

[0008] Therefore, the purpose of the present invention is to provide a coal and gas outburst danger zone identification and early warning method based on multi-parameter fusion.

[0009] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0010] The coal and gas outburst danger zone identification and early warning method based on multi-parameter fusion includes: obtaining multi-parameter data through a sensor network, connecting the sensor network through a monitoring transmission strategy, and transmitting the multi-parameter data to a cloud platform;

[0011] The features of the multi-parameter data transmitted to the cloud platform are extracted through feature fusion strategy, multi-parameter features are generated, and fusion processing is performed;

[0012] Construct a prominent warning model and prominent warning strategy based on multi-parameter features, and train the prominent warning model by obtaining historical prominent events and normal events;

[0013] Identify abnormal conditions through prominent warning models and prominent warning strategies. If an abnormality is identified, generate a signal and locate it in the signal area, and send a control instruction at the same time;

[0014] Visualize the signals and signal areas in the cloud platform and transmit them to managers and professional analysts, so as to remotely analyze and judge the prediction results;

[0015] Dynamically adjust the prominent warning model and prominent warning strategy based on multi-parameter data and prediction results.

[0016] As a preferred solution of the coal and gas outburst danger zone identification and early warning method based on multi-parameter fusion described in the present invention, wherein: the sensor network includes a microseismic sensor, a gas sensor, a wind speed sensor, an equipment start-stop sensor, a power-off controller, a voice playback alarm and a visual camera;

[0017] The multi-parameter data include microseismic signals, gas concentration, gas outflow, wind speed and equipment operating conditions;

[0018] The multi-parameter data is obtained through the sensor network, and the sensor network is connected through the monitoring transmission strategy to transmit the multi-parameter data to the cloud platform. The monitoring transmission strategy includes:

[0019] Integrate the multi-parameter data acquired by the sensor network, synchronize the multi-parameter data in time series, compress the multi-parameter data into data packets, dynamically select the transmission path of the data packets according to the network conditions, set a data buffer at the receiving end of the cloud platform, put the received data packets into the data buffer, and dynamically adjust the transmission bandwidth of the data packets according to the data volume.

[0020] As a preferred solution of the coal and gas outburst danger zone identification and early warning method based on multi-parameter fusion of the present invention, wherein: the transmission path of the data packet is dynamically selected according to the network status, and the function expression of the transmission path selection is as follows:

[0021]

[0022] Where t represents the transmission path selected according to the network conditions, d represents the network delay, max(d) represents the maximum value of the network delay, min(d) represents the minimum value of the network delay, l represents the network packet loss rate, max(l) represents the maximum value of the network packet loss rate, and min(l) represents the minimum value of the network packet loss rate.

[0023] As a preferred solution of the coal and gas outburst danger zone identification and early warning method based on multi-parameter fusion described in the present invention, the outburst early warning strategy includes:

[0024] The number of events and released energy of the microseismic signal are obtained, the amplitude and duration of the microseismic signal are extracted, the threshold value, the amplitude threshold value and the duration threshold value are configured, and the state of the microseismic signal is judged. If the amplitude of the microseismic signal is greater than or equal to the amplitude threshold value and the duration of the microseismic signal is greater than or equal to the duration threshold value, it indicates that a major event has occurred, which is recorded as abnormal ground stress activity;

[0025] If the amplitude of the microseismic signal is greater than the threshold value and the amplitude of the microseismic signal is less than the amplitude threshold value, it indicates that a small event has occurred, which is recorded as abnormal ground stress activity;

[0026] If the amplitude of the microseismic signal is less than or equal to the threshold value and the duration of the microseismic signal is less than or equal to the duration threshold value, the working face shows that the ground stress activity is normal.

[0027] As a preferred solution of the coal and gas outburst danger zone identification and early warning method based on multi-parameter fusion described in the present invention, the outburst early warning strategy also includes:

[0028] Obtaining the extreme value of the gas concentration, configuring a concentration safety threshold, and determining the state of the gas concentration. If the extreme value of the gas concentration is greater than or equal to the concentration safety threshold, it is recorded as abnormal concentration.

[0029] If the extreme value of the gas concentration is less than the concentration safety threshold, the working face displays a normal concentration;

[0030] Obtaining the extreme value of the gas outflow, configuring an outflow safety threshold, and judging the state of the gas outflow. If the extreme value of the gas outflow is greater than or equal to the outflow safety threshold, it is recorded as an outflow abnormality.

[0031] If the extreme value of the gas outflow is less than the outflow safety threshold, the working face shows that the outflow is normal;

[0032] Obtain abnormal wind speed fluctuations, extract wind speed variation, configure wind speed variable thresholds, and determine the state of the wind speed. If the wind speed variation is greater than or equal to the wind speed variable threshold, it is recorded as abnormal ventilation state.

[0033] If the wind speed change is less than the wind speed variable threshold, the working face shows that the ventilation status is normal;

[0034] Obtain device abnormality and determine the device status. If the device is abnormal, it will be recorded as a device abnormality.

[0035] If there is no abnormality with the equipment, the working surface will show that the equipment is normal.

[0036] As a preferred solution of the coal and gas outburst danger zone identification and early warning method based on multi-parameter fusion described in the present invention, wherein: the abnormal state is identified by the outburst early warning model and the outburst early warning strategy, and the strategy for identifying the abnormal state includes:

[0037] Determine the output result of the outburst warning model. If it shows that there is no outburst danger, it means that the working face is in a normal state. Continuously monitor the multi-parameter data to generate a safety signal.

[0038] If a danger of outburst is displayed, the abnormal state is comprehensively identified through the outburst warning strategy. If the working face is in a normal state, it means that no abnormality is identified, and the multi-parameter data is continuously monitored to generate a safety signal;

[0039] If there is an abnormal state on the working surface, determine the state of the equipment. If the working surface shows that the equipment is abnormal, identify the abnormality and generate an early warning signal;

[0040] If the working face shows that the equipment is normal, the state of the microseismic signal is determined; if the working face shows that the ground stress activity is abnormal, the state of the microseismic energy is determined; if the microseismic energy is abnormal, the abnormality is identified and an emergency signal is generated;

[0041] If the microseismic energy is normal, the state of the microseismic event and the abnormality of the ground stress activity are comprehensively judged;

[0042] If the working face shows that the ground stress activity is normal, the gas concentration, the gas outflow and the wind speed are determined; if the working face shows that the concentration is abnormal, the outflow is abnormal and the ventilation state is abnormal, the abnormality is identified and an early warning signal is generated;

[0043] If the working surface shows abnormal concentration, abnormal outflow or abnormal ventilation status, the multi-parameter data is continuously monitored to generate a safety signal.

[0044] As a preferred solution of the coal and gas outburst danger zone identification and early warning method based on multi-parameter fusion described in the present invention, the strategy for identifying abnormal conditions also includes:

[0045] Determine the state of the microseismic event and the state of the microseismic energy, extract the microseismic event frequency and microseismic energy of the microseismic signal, configure the event frequency threshold and the microseismic energy threshold. If the microseismic event frequency is less than or equal to the event frequency threshold, it means that the microseismic event is normal.

[0046] If the frequency of microseismic events is greater than the event frequency threshold, it indicates that the microseismic events are abnormal;

[0047] If the microseismic energy is less than or equal to the microseismic energy threshold, it means that the microseismic energy is normal;

[0048] If the microseismic energy is greater than the microseismic energy threshold, it indicates that the microseismic energy is abnormal.

[0049] As a preferred solution of the coal and gas outburst danger zone identification and early warning method based on multi-parameter fusion described in the present invention, the strategy for identifying abnormal conditions also includes:

[0050] Comprehensively judge the status of microseismic events and abnormal ground stress activity. If a major event occurs and the microseismic events are normal, identify the anomaly and generate a danger signal;

[0051] If a major event occurs and the microseismic event is abnormal, the anomaly is identified and a critical signal is generated;

[0052] If a small event occurs and the microseismic event is normal, the anomaly is identified and an early warning signal is generated;

[0053] If a small event occurs and the microseismic event is abnormal, the gas concentration, the gas outflow and the wind speed are determined. If the working face shows abnormal concentration, abnormal outflow and normal ventilation, the abnormality is identified and an emergency signal is generated.

[0054] If the working face shows abnormal concentration, abnormal outflow or abnormal ventilation status, the abnormality is identified and a danger signal is generated.

[0055] As a preferred solution of the coal and gas outburst danger zone identification and early warning method based on multi-parameter fusion described in the present invention, the strategy for identifying abnormal conditions also includes:

[0056] If the safety signal is generated, the signal area is a safety area;

[0057] If the warning signal is generated, the signal area is a threat area;

[0058] If the danger signal is generated, the signal area is a danger area;

[0059] If the emergency signal is generated, the signal area is an emergency area.

[0060] As a preferred solution of the coal and gas outburst danger zone identification and early warning method based on multi-parameter fusion of the present invention, wherein: visually displaying the signal and the signal area in the cloud platform includes:

[0061] A three-dimensional map is constructed according to the working surface, and the position of the sensor network is marked on the three-dimensional map, and the multi-parameter data is superimposed on the three-dimensional map, and the multi-parameter data is marked according to the state of the multi-parameter data to distinguish the signal and the signal area. If it is the safety signal and the safety area, the three-dimensional map is displayed normally;

[0062] If it is not the safety signal or the safety zone, the three-dimensional map will be highlighted.

[0063] Beneficial effects of the present invention: The present invention obtains multi-parameter data through a sensor network, connects the sensor network through a monitoring transmission strategy, transmits the multi-parameter data to a cloud platform, realizes real-time transmission of multi-parameter data and remote monitoring of the cloud platform, realizes comprehensive monitoring of mine outburst warning, and transmits field data to the cloud platform in real time through an effective monitoring transmission strategy to ensure the timeliness of the data, which is crucial for rapid response to possible dangerous situations; extracts the features of the multi-parameter data transmitted to the cloud platform through a feature fusion strategy, generates multi-parameter features, and performs fusion processing to more accurately characterize the safety status of the mine; constructs an outburst warning model and an outburst warning strategy based on the multi-parameter features, and obtains historical outburst events and normal event training The outburst warning model improves the accuracy and reliability of predictions of coal and gas outburst hazards; abnormal conditions are identified through the outburst warning model and outburst warning strategy. If an abnormality is identified, a signal is generated and located in the signal area. At the same time, a control instruction is sent to automatically identify the abnormal condition and locate the danger zone, which speeds up the decision-making process, reduces dependence on human intervention, and can effectively allocate escape and rescue resources; the signal and signal area are visualized on the cloud platform and transmitted to management personnel and professional analysts at the same time, and the prediction effect is judged remotely, which helps to reduce casualties and accidents and improve the overall safety of miners; the outburst warning model and outburst warning strategy are dynamically adjusted according to multi-parameter data and prediction effects to adapt to the ever-changing mine environment and mining conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor. Among them:

[0065] Figure 1It is a method flow chart of the method for identifying and warning the coal and gas outburst danger zone based on multi-parameter fusion of the present invention;

[0066] Figure 2 It is a flow chart of the monitoring and transmission strategy of the coal and gas outburst danger zone identification and early warning method based on multi-parameter fusion of the present invention;

[0067] Figure 3 It is a flow chart of ventilation status judgment of the coal and gas outburst danger zone identification and early warning method based on multi-parameter fusion of the present invention;

[0068] Figure 4 The present invention is a flow chart of the abnormal state recognition strategy of the coal and gas outburst danger zone identification and early warning method based on multi-parameter fusion. DETAILED DESCRIPTION

[0069] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0070] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention can also be implemented in other ways different from those described herein, and those skilled in the art can make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0071] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0072] Example 1

[0073] In this embodiment, a method flow chart of a method for identifying and warning a coal and gas outburst danger zone based on multi-parameter fusion is provided, such as Figure 1 As shown in the figure, the coal and gas outburst danger zone identification and early warning method based on multi-parameter fusion includes:

[0074] S1. Acquire multi-parameter data through the sensor network, connect the sensor network through the monitoring transmission strategy, transmit the multi-parameter data to the cloud platform, realize the real-time transmission and remote monitoring of multi-parameter data, provide a comprehensive description of the mine environment, so as to make the monitoring more comprehensive and accurate, and ensure the timeliness of multi-parameter data, so as to respond to possible dangerous situations in a timely manner.

[0075] The sensor network includes microseismic sensors, gas sensors, wind speed sensors, equipment start and stop sensors, power off controllers, voice playback alarms and visual cameras.

[0076] Multi-parameter data include microseismic signals, gas concentration, gas outflow, wind speed and equipment operating conditions;

[0077] The multi-parameter data is obtained through the sensor network, and the sensor network is connected through the monitoring transmission strategy to transmit the multi-parameter data to the cloud platform. The monitoring transmission strategy is as follows Figure 2 As shown, specifically including:

[0078] Integrate the multi-parameter data obtained by the sensor network, synchronize the multi-parameter data in time series, compress the multi-parameter data into data packets, dynamically select the transmission path of the data packets according to the network conditions, set a data buffer at the receiving end of the cloud platform, put the received data packets into the data buffer, and dynamically adjust the transmission bandwidth of the data packets according to the data volume.

[0079] The transmission path of the data packet is dynamically selected according to the network conditions. The function expression of the transmission path selection is as follows:

[0080]

[0081] Where t represents the transmission path selected according to the network conditions, d represents the network delay, max(d) represents the maximum value of the network delay, min(d) represents the minimum value of the network delay, l represents the network packet loss rate, max(l) represents the maximum value of the network packet loss rate, and min(l) represents the minimum value of the network packet loss rate.

[0082] The transmission bandwidth of the data packet is dynamically adjusted according to the data volume. The function expression of the transmission bandwidth adjustment is as follows:

[0083]

[0084] Wherein, w represents the transmission bandwidth adjusted according to the data volume, p represents the data volume, max(p) represents the maximum data volume, min(p) represents the minimum data volume, n represents the available bandwidth, max(n) represents the maximum available bandwidth, and min(n) represents the minimum available bandwidth.

[0085] Microseismic signals are obtained through microseismic sensors, gas concentration and gas outflow are obtained through gas sensors, and the surrounding wind speed is obtained through wind speed sensors. The equipment start and stop sensors are used to detect the start and stop status of the equipment. The power-off controller is used to detect the disconnection and reconnection of the equipment. The voice broadcast alarm is used for voice broadcasts, especially when locating threat areas, danger areas and critical areas, it can promptly notify the staff underground and above the mine. The visualization camera is used to transmit real-time images of the situation underground, in order to better visualize the signals and signal areas on the cloud platform. The equipment operating conditions refer to the operating conditions of large-scale mining equipment on the working face.

[0086] Preliminary data fusion is performed locally in the sensor network to integrate multi-parameter data from different sensor networks. At the same time, an efficient data compression algorithm is used to compress the multi-parameter data into data packets. The transmission path of the data packet is dynamically selected according to the network status, including considering the network delay and network packet loss rate in the network status, and selecting the best transmission path based on historical performance and current network conditions. At the same time, a data buffer is set at the receiving end of the cloud platform, and the received data packets are first placed in the data buffer. Then, the transmission bandwidth of the data packet is dynamically adjusted according to the data volume, including considering the data volume of the data packet itself and the remaining available bandwidth for the transmission bandwidth. At the same time, the priority of data packet transmission is determined according to the importance of the data in the data packet.

[0087] S2. The features of the multi-parameter data transmitted to the cloud platform are extracted through feature fusion strategy, multi-parameter features are generated, and fusion processing is performed to more accurately characterize the safety status of the mine.

[0088] The features of the multi-parameter data transmitted to the cloud platform are extracted and fused through feature fusion strategies. The feature fusion strategies include:

[0089] Feature extraction is performed on the multi-parameter data transmitted to the cloud platform to generate multi-parameter features, calculate the correlation of multi-parameter features, identify and eliminate redundant features, and fuse the multi-parameter features.

[0090] The functional expression of the correlation of multi-parameter features is as follows:

[0091]

[0092] In the formula, r XY Represents the correlation coefficient between feature X and feature Y, X i represents the i-th observation value of feature X, represents the average value of the observations of feature X, Y i represents the i-th observation value of feature Y, represents the average value of the observations of feature Y, and m represents the total number of observations.

[0093] If r XY →1, it means that when feature X increases, feature Y increases;

[0094] If r XY →0, it means that feature X is not related to feature Y;

[0095] If r XY →-1, it means that when feature X increases, feature Y decreases.

[0096] Feature extraction is performed on the multi-parameter data transmitted to the cloud platform to generate multi-parameter features, including the number of microseismic signal events and the energy released, the extreme values ​​of gas concentration, the extreme values ​​of gas outburst, abnormal wind speed fluctuations, and equipment abnormalities, etc. The number of microseismic signal events refers to the number of times coal rock fractures and damages occur.

[0097] When the correlation coefficient r between feature X and feature Y XY The closer it is to 1, the stronger the positive correlation between the two features. XY The closer it is to -1, the stronger the negative correlation between the two features. XY The closer it is to 0, the less correlation there is between the two features. For the correlation calculation between each pair of features, we set the feature threshold r and exclude |r XY |<r features, for |r XY For each pair of features with |≥r, the feature that is more important to the prominent warning model is selected and retained, and the other feature is eliminated. The multi-parameter features after redundant feature identification and elimination are fused.

[0098] S3. Construct a surge warning model and a surge warning strategy based on multi-parameter characteristics, and train the surge warning model by obtaining historical surge events and normal events to ensure that the surge warning model can accurately predict surge hazards.

[0099] According to the multi-parameter characteristics, a prominent warning model and a prominent warning strategy are constructed. The function expression of the prominent warning model is as follows:

[0100]

[0101] Where M final represents the final selected prominent warning model, It indicates that the prominent warning model with the maximum value of P(M,T) is found in the indicator set. P(M,T) indicates the indicator set for performance evaluation of the prominent warning model. M indicates the constructed prominent warning model. T indicates the training data set, including historical prominent events and normal events.

[0102] The training data set represents the set of all training samples, including feature vectors and corresponding labels, wherein the corresponding labels are obtained through the constructed outburst warning model according to the number of microseismic signal events and the energy released, the extreme value of gas concentration, the extreme value of gas outburst volume, the abnormal wind speed fluctuation and the feature vectors of equipment abnormality, 1 indicates that there is an outburst hazard and 0 indicates that there is no outburst hazard. The construction methods of the outburst warning model include but are not limited to support vector machines, convolutional neural networks and recurrent neural networks.

[0103] Prominent early warning strategies include:

[0104] Obtain the number of microseismic signal events and the energy released, extract the amplitude and duration of the microseismic signal, configure the threshold, amplitude threshold and duration threshold, and judge the state of the microseismic signal. If the amplitude of the microseismic signal is greater than or equal to the amplitude threshold and the duration of the microseismic signal is greater than or equal to the duration threshold, it means that a major event has occurred, which is recorded as abnormal ground stress activity.

[0105] If the amplitude of the microseismic signal is greater than the threshold and the amplitude of the microseismic signal is less than the amplitude threshold, it indicates that a small event has occurred, which is recorded as abnormal ground stress activity;

[0106] If the amplitude of the microseismic signal is less than or equal to the threshold and the duration of the microseismic signal is less than or equal to the duration threshold, the working face shows normal ground stress activity;

[0107] Get the extreme value of gas concentration, configure the concentration safety threshold, and judge the state of gas concentration. If the extreme value of gas concentration is greater than or equal to the concentration safety threshold, it is recorded as abnormal concentration.

[0108] If the extreme value of gas concentration is less than the concentration safety threshold, the working face will show that the concentration is normal;

[0109] Obtain the extreme value of gas outflow, configure the outflow safety threshold, and determine the state of gas outflow. If the extreme value of gas outflow is greater than or equal to the outflow safety threshold, it is recorded as outflow abnormality.

[0110] If the extreme value of gas outflow is less than the outflow safety threshold, the working face shows that outflow is normal;

[0111] like Figure 3 As shown, abnormal wind speed fluctuations are obtained, wind speed changes are extracted, wind speed variable thresholds are configured, and the state of wind speed is judged. If the wind speed change is greater than or equal to the wind speed variable threshold, it is recorded as abnormal ventilation state;

[0112] If the wind speed change is less than the wind speed variable threshold, the working face shows that the ventilation status is normal;

[0113] Obtain device abnormality and determine the device status. If the device is abnormal, it will be recorded as a device abnormality.

[0114] If there is no abnormality with the equipment, the working surface will show that the equipment is normal.

[0115] The number of microseismic signal events and the energy released, the extreme value of gas concentration, the extreme value of gas outburst, the abnormal fluctuation of wind speed and the abnormal condition of equipment are judged respectively, so as to comprehensively identify the abnormal state with the subsequent joint outburst warning model. The above threshold value can more accurately identify small events, and the amplitude threshold value is generally the maximum amplitude released when the main bearing rock formation is destroyed, which can more accurately identify large events. If small events occur throughout the day, especially when the working face is unmanned, it means that the concentrated stress of the working face is constantly shifting and intensifying. At this time, the working face has entered the safe zone. Threat zone, and when major events begin to occur, it means that the stress intensity of the hard roof rock has reached a critical state. The monitoring personnel must pay close attention to the changes in other multi-parameter data and remind mine managers to take appropriate safety measures. This means that the working face has entered the danger zone of protrusion. When major events occur intensively, it means that the hard roof rock has been fractured on a large scale, and the energy accumulated in the fractured rock layer begins to impact the coal seam. The energy of the collected microseismic signal will mutate, and the energy value will reach more than tens of thousands of joules. The impact force on the coal seam will change the permeability of the coal seam, causing changes in the gas properties of the coal seam. At this time, it is necessary to notify the staff to evacuate in time. When the gas concentration changes, observe whether the abnormal fluctuation of wind speed leads to the change of gas concentration. Under the condition of stable ventilation, the change of gas concentration means that the gas properties of the coal seam have changed, and the change of gas outburst volume under the condition of constant wind speed also means that the gas properties of the coal seam have changed. At the same time, whether the operating conditions of large-scale mining equipment on the working face are abnormal will also affect whether the concentration of ground stress is caused by the operation of mining equipment. Most of the outburst accidents on the working face are related to the operation of mechanical equipment. The operating status of underground mining equipment will affect the stress tension of the working face. Many outburst accidents are induced by the construction of mechanical equipment. The continuous operation of mining equipment will aggravate the stress tension of the working face and increase the risk of outburst on the working face. Stopping the operation of mining equipment can relieve the stress tension of the working face, reduce the maximum concentrated stress, and transfer it to the deep part of the working face, creating a larger safety distance in front of the working face, reducing the risk of outburst. The working status of the mining equipment can be detected by the equipment start-stop sensor, and cooperate with other early warning sensors to control the mining equipment to operate within a safe working range.

[0116] S4. Identify abnormal conditions through prominent warning models and prominent warning strategies. If an abnormality is identified, generate a signal and locate it in the signal area, and send a control instruction at the same time.

[0117] By highlighting the early warning model and the early warning strategy, the abnormal state is identified. The strategy for identifying the abnormal state is as follows: Figure 4 As shown, specifically including:

[0118] Determine the output result of the outburst warning model. If it shows no outburst danger, it means that the working face is in a normal state. Continuously monitor multi-parameter data to generate safety signals.

[0119] If a sudden outburst hazard is displayed, the abnormal state is comprehensively identified through the sudden outburst warning strategy. If the working face is in a normal state, it means that no abnormality has been identified. The multi-parameter data is continuously monitored to generate a safety signal.

[0120] If there is an abnormal state on the working surface, determine the state of the equipment. If the working surface shows that the equipment is abnormal, identify the abnormality and generate an early warning signal;

[0121] If the working face shows that the equipment is normal, the state of the microseismic signal is determined. If the working face shows that the ground stress activity is abnormal, the microseismic event frequency and microseismic energy of the microseismic signal are extracted, and the event frequency threshold and microseismic energy threshold are configured to obtain the state of the microseismic event and the state of the microseismic energy. If the microseismic event frequency is less than or equal to the event frequency threshold, it means that the microseismic event is normal.

[0122] If the frequency of microseismic events is greater than the event frequency threshold, it indicates that the microseismic events are abnormal;

[0123] If the microseismic energy is less than or equal to the microseismic energy threshold, it means that the microseismic energy is normal;

[0124] If the microseismic energy is greater than the microseismic energy threshold, it indicates that the microseismic energy is abnormal;

[0125] Determine the state of microseismic energy. If the microseismic energy is abnormal, identify the abnormality and generate a critical signal;

[0126] If the microseismic energy is normal, the state of the microseismic event and the abnormality of the ground stress activity are comprehensively judged. If a major event occurs and the microseismic event is normal, the anomaly is identified and a danger signal is generated;

[0127] If a major event occurs and the microseismic event is abnormal, the anomaly is identified and a critical signal is generated;

[0128] If a small event occurs and the microseismic event is normal, the anomaly is identified and an early warning signal is generated;

[0129] If a small event occurs and the microseismic event is abnormal, the gas concentration, gas outflow and wind speed are determined. If the working face shows abnormal concentration, abnormal outflow and normal ventilation, the abnormality is identified and a critical signal is generated;

[0130] If the working face shows abnormal concentration, abnormal outflow or abnormal ventilation status, the abnormality is identified and a danger signal is generated;

[0131] If the working face shows normal ground stress activity, the gas concentration, gas outflow and wind speed are judged; if the working face shows abnormal concentration, abnormal outflow and abnormal ventilation status, the abnormality is identified and an early warning signal is generated;

[0132] If the working face shows abnormal concentration, abnormal outflow or abnormal ventilation status, multi-parameter data will be continuously monitored to generate a safety signal.

[0133] When the working face shows continuous microseismic signals that are small event signals with small energy, it indicates that there is ground stress activity on the working face. When the microseismic events are abnormal, it means that small events occur frequently. When the working face shows abnormal concentration, abnormal outburst and normal ventilation status, it indicates that the coal seam characteristics have changed.

[0134] When a major event occurs, hard rock fractures in the coal-bearing strata are identified. At the same time, abnormal concentration and outburst occur when the ventilation of the working face is normal. This is a precursor to the danger of an outburst.

[0135] When the microseismic energy increases significantly and exceeds the microseismic energy threshold, and when the ventilation is normal, the gas concentration and gas outflow increase sharply, and an outburst accident may occur.

[0136] If a safety signal is generated, the signal area is a safety area, and the control command sent is to continuously monitor multi-parameter data;

[0137] If an early warning signal is generated, the signal area is a threat area, indicating that the operation is in a coal outburst threat area. The control instruction sent is to pay close attention to changes in coal seam occurrence conditions, including coal seam thickness and inclination, and take necessary coal outburst hazard verification work;

[0138] If a danger signal is generated, the signal area is a danger zone, indicating that the operation is in a dangerous area of ​​outburst, and the control command sent is to stop or delay the operation of the mining equipment to reduce the danger of outburst;

[0139] If an emergency signal is generated, the signal area becomes an emergency area, and the control command sent is to immediately cut off power and evacuate relevant personnel.

[0140] At the same time, when a danger signal or warning signal is issued, it is determined whether the equipment is in working condition. When the equipment is in working condition and the degree of abnormality on the working face is relatively small, it is determined that the abnormality on the working face is caused by the mining equipment.

[0141] The function expression of the positioning signal area is as follows:

[0142]

[0143] Where q = 0 represents a safety signal, q = 1 represents a warning signal, q = 2 represents a danger signal, e(s, g) represents the location of the warning zone obtained by using the location information s of the sensor network and the GIS integrated data g, and h(s, g) represents the location of the danger zone obtained by using the location information s of the sensor network and the GIS integrated data g.

[0144] It needs to be explained that the location information of the sensor network can be obtained through the positioning function of the sensor itself. GIS integrated data integrates the acquired multi-parameter data with the GIS system, and obtains the precise location of abnormal data through geographic and spatial analysis technology.

[0145] When a safety signal is generated, it is necessary to continuously monitor changes in multi-parameter data, regularly inspect and maintain all sensor networks and monitoring equipment to ensure the accuracy of multi-parameter data, and underground workers should continue to comply with all standard safety operating procedures; when an early warning signal is generated, it is necessary to increase the frequency and range of monitoring multi-parameter data, initiate the early warning response process, adjust the work plan, avoid high-risk areas, increase safety patrols, continue to pay close attention to data changes and on-site conditions of multi-parameter data, and assess whether the predicted situation will deteriorate; when a danger signal is generated, it is necessary to stop the operation of all mining equipment, shut down non-essential power supplies and equipment to reduce the risk of secondary disasters and mitigate the danger of outbursts; when an emergency signal is generated, it is necessary to promptly notify underground workers for emergency evacuation, ensure that all personnel are aware of the nearest safe exits and evacuation routes, immediately shut down all power supplies and equipment, activate the emergency response center, assemble professional rescue teams and implement rescue operations.

[0146] S5. Visualize the signals and signal areas in the cloud platform and transmit them to managers and professional analysts at the same time. Remotely analyze and judge the prediction results, which helps managers quickly understand the situation and make decisions.

[0147] Strategies for visualizing signals and signal areas in the cloud platform include:

[0148] A three-dimensional map is constructed based on the working surface, and the location of the sensor network is marked on the three-dimensional map. The multi-parameter data is superimposed on the three-dimensional map, and the multi-parameter data is marked according to the state of the multi-parameter data to distinguish between signals and signal areas. If it is a safe signal and a safe area, the three-dimensional map will be displayed normally;

[0149] If it is a non-safe signal or non-safe area, the 3D map will be highlighted.

[0150] A three-dimensional map is created through GIS technology to accurately display the layout and key locations of the mine. The locations of all sensor networks are marked on the three-dimensional map so that data collection points can be observed intuitively. The multi-parameter data acquired in real time is superimposed on the three-dimensional map, and different labels are used to distinguish the status of the multi-parameter data, including normal and abnormal multi-parameter data, and ensure that the multi-parameter data can be updated in real time to reflect the latest monitoring status. When an early warning, danger or crisis occurs in a certain area, the area is highlighted on the three-dimensional map, and yellow, orange and red colors correspond to the early warning, danger and crisis situations in the area respectively. At the same time, by clicking on the corresponding threat area, danger area and crisis area on the three-dimensional map, detailed threat information, danger information and crisis information can be viewed, including time, possible impact range and corresponding safety measures and urgency, etc.

[0151] The time from the issuance of warning signals and danger signals to the response of managers and staff is recorded to determine the prediction effect. At the same time, the data changes of multi-parameter data after the issuance of warning signals, danger signals and emergency signals are analyzed, including the changing trend of gas concentration, the changing trend of gas outburst volume, the number of microseismic signal events and the energy released, etc., to determine the necessity and timeliness of the prediction, and monitor the warning signals, danger signals and the process of transmission from the cloud platform to managers and professional analysts, provide safety guidance through voice playback alarms, obtain real-time image transmission of the mine through visual cameras, ensure the speed and accuracy of information transmission, and optimize the transmission path and transmission bandwidth based on the feedback of information transmission to reduce network delays and information loss.

[0152] S6. Dynamically adjust the outburst warning model and outburst warning strategy based on multi-parameter data and prediction results, adapt to changes in the mine environment, and continuously optimize prediction accuracy.

[0153] Continuously monitor the performance of the outburst warning model, including indicators such as accuracy, recall rate and F1 score, collect feedback information on the prediction effect, including the accuracy and timeliness of the prediction and the response of miners and managers, and input new multi-parameter data and event results into the training set of the outburst warning model in real time to ensure that the outburst warning model can adapt to the latest data trends. At the same time, dynamically adjust the thresholds of each multi-parameter data according to the prediction effect to reduce false alarms and missed alarms, evaluate the effectiveness of emergency response in real time, and ensure that prompt action can be taken when warning signals, danger signals and critical signals occur to improve efficiency and safety.

[0154] Example 2

[0155] In this embodiment, a computer device is provided, including a memory and a processor, the memory is used to store instructions, and the processor is used to execute the instructions, so that the computer device executes the steps of implementing the above-mentioned coal and gas outburst danger zone identification and early warning method based on multi-parameter fusion.

[0156] Example 3

[0157] In this embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed, the steps of the above-mentioned coal and gas outburst danger zone identification and early warning method based on multi-parameter fusion are implemented.

[0158] The computer-readable storage medium includes: a U disk, a mobile hard disk, a read-only memory, a random access memory, a magnetic disk or an optical disk, and other media for storing program codes.

[0159] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should be included in the scope of the claims of the present invention.

Claims

1. A coal and gas outburst danger zone identification and early warning method based on multi-parameter fusion, characterized by: include: Acquire multi-parameter data through sensor networks, connect sensor networks through monitoring transmission strategies, and transmit multi-parameter data to the cloud platform; The features of the multi-parameter data transmitted to the cloud platform are extracted through feature fusion strategy, multi-parameter features are generated, and fusion processing is performed; Construct a prominent warning model and prominent warning strategy based on multi-parameter features, and train the prominent warning model by obtaining historical prominent events and normal events; Identify abnormal conditions through prominent warning models and prominent warning strategies. If an abnormality is identified, generate a signal and locate it in the signal area, and send a control instruction at the same time; Visualize the signals and signal areas in the cloud platform and transmit them to managers and professional analysts, so as to remotely analyze and judge the prediction results; Dynamically adjust the prominent warning model and prominent warning strategy based on multi-parameter data and prediction results.

2. The method for identifying and warning coal and gas outburst danger zones based on multi-parameter fusion according to claim 1 is characterized in that: The sensor network includes microseismic sensors, gas sensors, wind speed sensors, equipment start and stop sensors, power failure controllers, voice playback alarms and visual cameras; The multi-parameter data include microseismic signals, gas concentration, gas outflow, wind speed and equipment operating conditions; The multi-parameter data is obtained through the sensor network, and the sensor network is connected through the monitoring transmission strategy to transmit the multi-parameter data to the cloud platform. The monitoring transmission strategy includes: Integrate the multi-parameter data acquired by the sensor network, synchronize the multi-parameter data in time series, compress the multi-parameter data into data packets, dynamically select the transmission path of the data packets according to the network conditions, set a data buffer at the receiving end of the cloud platform, put the received data packets into the data buffer, and dynamically adjust the transmission bandwidth of the data packets according to the data volume.

3. The method for identifying and warning coal and gas outburst danger zones based on multi-parameter fusion according to claim 2 is characterized in that: The transmission path of the data packet is dynamically selected according to the network status, and the function expression of the transmission path selection is as follows: Where t represents the transmission path selected according to the network conditions, d represents the network delay, max(d) represents the maximum value of the network delay, min(d) represents the minimum value of the network delay, l represents the network packet loss rate, max(l) represents the maximum value of the network packet loss rate, and min(l) represents the minimum value of the network packet loss rate.

4. The method for identifying and warning coal and gas outburst danger zones based on multi-parameter fusion according to claim 3 is characterized in that: The prominent early warning strategies include: The number of events and released energy of the microseismic signal are obtained, the amplitude and duration of the microseismic signal are extracted, the threshold value, the amplitude threshold value and the duration threshold value are configured, and the state of the microseismic signal is judged. If the amplitude of the microseismic signal is greater than or equal to the amplitude threshold value and the duration of the microseismic signal is greater than or equal to the duration threshold value, it indicates that a major event has occurred, which is recorded as abnormal ground stress activity; If the amplitude of the microseismic signal is greater than the threshold value and the amplitude of the microseismic signal is less than the amplitude threshold value, it indicates that a small event has occurred, which is recorded as abnormal ground stress activity; If the amplitude of the microseismic signal is less than or equal to the threshold value and the duration of the microseismic signal is less than or equal to the duration threshold value, the working face shows that the ground stress activity is normal.

5. The method for identifying and warning coal and gas outburst danger zones based on multi-parameter fusion according to claim 4 is characterized in that: The prominent early warning strategy also includes: Obtaining the extreme value of the gas concentration, configuring a concentration safety threshold, and determining the state of the gas concentration. If the extreme value of the gas concentration is greater than or equal to the concentration safety threshold, it is recorded as abnormal concentration. If the extreme value of the gas concentration is less than the concentration safety threshold, the working face displays a normal concentration; Obtaining the extreme value of the gas outflow, configuring an outflow safety threshold, and judging the state of the gas outflow. If the extreme value of the gas outflow is greater than or equal to the outflow safety threshold, it is recorded as an outflow abnormality. If the extreme value of the gas outflow is less than the outflow safety threshold, the working face shows that the outflow is normal; Obtain abnormal wind speed fluctuations, extract wind speed variation, configure wind speed variable thresholds, and judge the state of the wind speed. If the wind speed variation is greater than or equal to the wind speed variable threshold, it is recorded as abnormal ventilation state; If the wind speed change is less than the wind speed variable threshold, the working face shows that the ventilation status is normal; Obtain device abnormality and determine the device status. If the device is abnormal, it will be recorded as a device abnormality. If there is no abnormality with the equipment, the working surface will show that the equipment is normal.

6. The method for identifying and warning coal and gas outburst danger zones based on multi-parameter fusion according to claim 5 is characterized in that: The abnormal state is identified by the prominent warning model and the prominent warning strategy, and the strategy for identifying the abnormal state includes: Determine the output result of the outburst warning model. If it shows that there is no outburst danger, it means that the working face is in a normal state. Continuously monitor the multi-parameter data to generate a safety signal. If a danger of outburst is displayed, the abnormal state is comprehensively identified through the outburst warning strategy. If the working face is in a normal state, it means that no abnormality is identified, and the multi-parameter data is continuously monitored to generate a safety signal; If there is an abnormal state on the working surface, determine the state of the equipment. If the working surface shows that the equipment is abnormal, identify the abnormality and generate an early warning signal; If the working face shows that the equipment is normal, the state of the microseismic signal is determined; if the working face shows that the ground stress activity is abnormal, the state of the microseismic energy is determined; if the microseismic energy is abnormal, the abnormality is identified and an emergency signal is generated; If the microseismic energy is normal, the state of the microseismic event and the abnormality of the ground stress activity are comprehensively judged; If the working face shows that the ground stress activity is normal, the gas concentration, the gas outflow and the wind speed are determined; if the working face shows that the concentration is abnormal, the outflow is abnormal and the ventilation state is abnormal, the abnormality is identified and an early warning signal is generated; If the working surface shows abnormal concentration, abnormal outflow or abnormal ventilation status, the multi-parameter data is continuously monitored to generate a safety signal.

7. The method for identifying and warning coal and gas outburst danger zones based on multi-parameter fusion according to claim 6 is characterized in that: Strategies for identifying abnormal conditions also include: Determine the state of the microseismic event and the state of the microseismic energy, extract the microseismic event frequency and microseismic energy of the microseismic signal, configure the event frequency threshold and the microseismic energy threshold. If the microseismic event frequency is less than or equal to the event frequency threshold, it means that the microseismic event is normal. If the frequency of microseismic events is greater than the event frequency threshold, it indicates that the microseismic events are abnormal; If the microseismic energy is less than or equal to the microseismic energy threshold, it means that the microseismic energy is normal; If the microseismic energy is greater than the microseismic energy threshold, it indicates that the microseismic energy is abnormal.

8. The method for identifying and warning coal and gas outburst danger zones based on multi-parameter fusion according to claim 7 is characterized in that: Strategies for identifying abnormal conditions also include: Comprehensively judge the status of microseismic events and abnormal ground stress activity. If a major event occurs and the microseismic events are normal, identify the anomaly and generate a danger signal; If a major event occurs and the microseismic event is abnormal, the anomaly is identified and a critical signal is generated; If a small event occurs and the microseismic event is normal, the anomaly is identified and an early warning signal is generated; If a small event occurs and the microseismic event is abnormal, the gas concentration, the gas outflow and the wind speed are determined. If the working face shows abnormal concentration, abnormal outflow and normal ventilation, the abnormality is identified and an emergency signal is generated. If the working face shows abnormal concentration, abnormal outflow or abnormal ventilation status, the abnormality is identified and a danger signal is generated.

9. The method for identifying and warning coal and gas outburst danger zones based on multi-parameter fusion according to claim 8, characterized in that: Strategies for identifying abnormal conditions also include: If the safety signal is generated, the signal area is a safety area; If the warning signal is generated, the signal area is a threat area; If the danger signal is generated, the signal area is a danger area; If the emergency signal is generated, the signal area is an emergency area.

10. The method for identifying and warning coal and gas outburst danger zones based on multi-parameter fusion according to claim 9, characterized in that: Visually displaying the signal and the signal area in the cloud platform includes: A three-dimensional map is constructed according to the working surface, and the position of the sensor network is marked on the three-dimensional map, and the multi-parameter data is superimposed on the three-dimensional map, and the multi-parameter data is marked according to the state of the multi-parameter data to distinguish the signal and the signal area. If it is the safety signal and the safety area, the three-dimensional map is displayed normally; If it is not the safety signal or the safety zone, the three-dimensional map will be highlighted.

Images