CN-121999586-A - Intelligent sensing and early warning prevention and control system for oil type gas-gas composite disaster

CN121999586ACN 121999586 ACN121999586 ACN 121999586ACN-121999586-A

Abstract

The invention relates to the technical field of coal mine disaster prevention and control, in particular to an oil-gas composite disaster intelligent sensing and early warning prevention and control system which comprises a self-adaptive sensing module, a multi-mode data fusion module, a lightweight intelligent early warning module, a hierarchical linkage regulation and control module and a centralized intelligent management and control platform. The comprehensive and accurate monitoring method for the composite disasters improves the comprehensiveness and accuracy of the composite disaster monitoring, reduces the false alarm rate, shortens the early warning response time, and remarkably improves the safety guarantee capability of the mine.

Inventors

DU XUANHONG
JIANG SHENGLING
YU LAN
ZHAO HAIZHANG
LI JINTAO
CHEN HAO
YuChi Xiaojian
SHU PAN
Cao Juheng
MA DEQI

Assignees

陇东学院

Dates

Publication Date: 20260508
Application Date: 20260114

Claims (10)

1. Oil type gas-gas composite disaster intelligent perception and early warning prevention and control system, its characterized in that includes: The self-adaptive sensing module is used for dynamically adjusting the mounting positions and densities of the CH 4 sensor, the C 2 H 6 sensor, the microseismic sensor, the stress sensor and the environmental temperature and humidity sensor through a self-adaptive deployment algorithm based on the working face propelling speed, the rock stratum damage depth and the oil-gas aggregation risk level, so as to construct the underground full-coverage networking; The multi-mode data fusion module adopts an attention mechanism multi-mode fusion algorithm to perform feature extraction and weight distribution on the gas concentration, the microseismic signals, the stress data, the environmental parameters and the equipment operation parameters acquired by the self-adaptive sensing module to generate a fusion feature set; The lightweight intelligent early warning module is used for training an early warning model through an LSTM (least squares) and random forest hybrid machine learning algorithm based on a disaster mechanism model, historical extraction data and the fusion feature set, and realizing crossing from threshold warning to trend prediction by combining a dynamic threshold adjustment algorithm, and dividing primary, secondary and tertiary early warning grades; The hierarchical linkage regulation and control module is used for linking the existing extraction system and the local ventilation system according to the early warning level, executing extraction power regulation and ventilation control, and additionally triggering a power-off instruction during three-level early warning; The centralized intelligent management and control platform integrates the functions of data storage, real-time display, early warning issue and instruction issuing, and is in butt joint with a mine personnel positioning system and an emergency broadcasting system to form a full-flow automatic closed loop.
2. The intelligent sensing and early warning prevention and control system for the oil-gas composite disaster of claim 1, wherein the self-adaptive deployment algorithm comprises the steps of acquiring a rock stratum damage range and an oil-gas escape path through numerical simulation, dynamically updating a high-risk area by combining real-time working face propulsion data, and raising the sensor density to 0.5-1/10 m 2 in the high-risk area, wherein the low-risk area keeps 1/50 m 2 .
3. The intelligent sensing and early warning prevention and control system for oil-type gas composite disasters according to claim 1, wherein the attention mechanism of the multi-mode data fusion algorithm is characterized in that 30% -40% of weight is distributed to gas concentration data, 40% -50% of weight is distributed to microseism and stress data, and 10% -20% of weight is distributed to environmental and equipment parameters.
4. The intelligent sensing and early warning prevention and control system for oil-type gas composite disasters according to claim 1, wherein the dynamic threshold adjustment algorithm updates the early warning threshold every 15 minutes based on near 72 hours of historical data, real-time environmental temperature and humidity and rock stratum stress change rate.
5. The intelligent sensing and early warning prevention and control system for the oil-type gas composite disasters, which is disclosed in claim 1, is characterized in that the probability of occurrence of the oil-type gas composite disasters is greater than or equal to 80% within 1 hour corresponding to the primary early warning, the probability of occurrence of the oil-type gas composite disasters is 50% -80% within 3 hours corresponding to the secondary early warning, and the probability of occurrence of the oil-type gas composite disasters is 30% -50% within 6 hours corresponding to the tertiary early warning.
6. The intelligent sensing and early warning prevention and control system for the oil-type gas composite disasters, which is disclosed in claim 1, is characterized in that in the hierarchical linkage regulation and control, the primary early warning is used for raising the power of a pumping pump to 80% -100%, the secondary pumping pump is raised to 60% -80%, and the tertiary pumping pump is raised to 40% -60%.
7. The intelligent sensing and early warning prevention and control system for the oil-type gas composite disasters, which is disclosed in claim 1, is characterized in that the total parameter amount of the lightweight mixed model is less than or equal to 500 ten thousand, the inference delay is less than or equal to 0.5 seconds, and the intelligent sensing and early warning prevention and control system is suitable for mine edge computing equipment.
8. The intelligent sensing and early warning prevention and control system for the oil-type gas composite disasters, which is disclosed in claim 1, is characterized in that the centralized intelligent management and control platform supports Web end and mobile end access, and early warning information is distributed through multiple channels of audible and visual alarm, short message and APP push.
9. The intelligent sensing and early warning prevention and control system for the oil-type gas composite disasters, which is disclosed in claim 1, is characterized in that the sensor has dustproof, waterproof and impact resistance, the protection level is more than or equal to IP67, and the measurement error is less than or equal to +/-2%.
10. The intelligent sensing and early warning prevention and control system for the oil-gas composite disaster according to claim 1, wherein the emergency linkage comprises marking the positions of operators and sending evacuation navigation by a personnel positioning system and circularly playing an alarm by an emergency broadcasting system when three-stage early warning is performed, and a power-off system cuts off power supply of non-intrinsically safe equipment in a dangerous area.

Description

Intelligent sensing and early warning prevention and control system for oil type gas-gas composite disaster Technical Field The invention relates to the technical field of coal mine disaster prevention and control, in particular to an oil type gas-gas composite disaster intelligent sensing and early warning prevention and control system. Background In the coal mining process, oil-gas and gas composite disasters become one of key hidden dangers threatening the safe and efficient mining of mines. The oil-type gas has the characteristics of special occurrence and disaster which are obviously different from the gas, namely, the oil-type gas is in small trap distribution, so that the detection difficulty is extremely high, a monitoring blind area is easy to form by a conventional monitoring means, the space distribution is highly non-uniform, the disaster prevention and control lack of pertinence, the accurate prevention and control are difficult to realize, meanwhile, the oil-type gas is inflammable and explosive, the explosion limit range is wide, once leaked, the malignant safety accident is easy to cause, and the hazard degree is far higher than that of a single gas disaster. More importantly, the oil-type gas is stored in the rock stratum in a free state, so that the oil-type gas has the characteristic of instantaneous massive emission, the oil-type gas is further induced to be released in a large scale due to the fact that the oil-type gas is protruded, a disaster amplifying effect is formed after the oil-type gas and the oil-type gas are overlapped, the occurrence probability of the coal mine gas is greatly increased, the risk of the gas explosion accident is exponentially improved due to the inflammable and explosive characteristics of the oil-type gas, and extremely severe requirements are provided for safety management work such as mine ventilation and gas extraction. In an actual mining scene, as the working surface is continuously pushed, the rock stratum within a certain range of the bottom plate can be damaged and destroyed gradually. When the damage depth reaches the layer of the bottom plate shale layer, the shale layer which is originally used as an oil type gas cover layer can lose the sealing capacity, oil type gas can quickly flow into a working surface through a crack channel formed by the damage of the shale layer, and is free and aggregated in the bottom plate rock layer on the top of the coal bed, so that a disaster source with strong concealment and high burst performance is formed, and the life safety of operators and the normal production order of a mine are seriously threatened. The existing system has the defects that firstly, a sensor deployment mode is stiff, a fixed position installation mode is adopted, complex coupling characteristics of composite disasters are difficult to capture, thirdly, an early warning mechanism has intrinsic defects, the traditional system can only passively warn when disaster symptoms are quite obvious, trend prediction cannot be realized, a fixed threshold is difficult to adapt to dynamic changes of mine environments, the error rate and the leakage rate are high and cannot be insufficient, the data processing and fusion capability is weak, the existing system is used for independently analyzing single indexes (such as CH 4 concentration), the gas concentration data and microseismic signals, rock stratum stress, environment temperature and humidity, equipment running states and the like, the data utilization rate is low, the complex coupling characteristics of the composite disasters are difficult to capture, thirdly, the early warning mechanism has intrinsic defects, the traditional system relies on a fixed threshold alarm mode, trend prediction cannot be realized, the fixed threshold is difficult to adapt to the dynamic changes of mine environments, the error rate and the leakage rate cannot be high, the adequate time cannot be obtained for the prevention and control operation, the four-stage extraction precision is not needed, the emergency extraction system is not matched with the emergency extraction system, the emergency extraction and the emergency extraction system is not matched with the emergency extraction system, the emergency control system is not matched with the emergency control system, the emergency extraction and the emergency control system is not well-controlled, and the emergency control system is not matched with the existing system is not to be matched with the emergency, and the emergency control system is difficult to be matched with the emergency control system, and the emergency control system is not to be matched with the emergency-down, and the emergency control system is difficult to be matched, and has the emergency control to be matched with the emergency control to be in the emergency-down, and has to be in a disaster-down. The technical defects cause that the existing prevention and control means cannot effectively cope