Multi-classifier Combined Anomaly Detection Algorithm Based On Feature Map In Underground Coal Mine

Face recognition based on depth images is widely studied due to its advantages of 3 dimensional information and environment illumination insensitivity. The traditional recognition methods in this field mainly focus on hand-crafted feature design, which cannot achieve satisfactory result. In addition, there is no fixed face feature extraction method. To achieve a better face recognition performance on depth images, this paper proposes a method based on Convolutional Neural Networks(CNN). The experiment performed on database IIITD Kinect suggests that the proposed CNN architecture has better recognition performance than some traditional manual feature extraction methods, such as HOG and LBP.

Underground coal mining causes ‘damage–fracture–instability’ of the overlying rock in the stope, resulting in movement and deformation of the mined rock mass that eventually extends to the surface, posing serious threats to buildings, transportation, pipelines, etc. To study the evolution rules of the overlying rock and disaster-causing mechanisms after coal seam mining, this paper analysed the surface subsidence of the West 1st District of Linsheng Coal Mine in the southern coalfield, Shenyang, China. Continuous models were initially used to analyse seven mining times in the coal mine. The overlying rock and surface displacement evolution rules were analysed, and then discrete models were used to further study the rock mass deformation and damage mechanism and rules for surface subsidence caused by coal mining. The results showed that the mining of inclined coal seams has a smaller impact on the surface in the shallower direction of the coal seam, but a larger impact on the surface in the deeper direction of the coal seam, as the deeper coal seam mining causes longer tensile fractures cutting the surface. The vertical surface displacement polyline caused by the mining of the upper coal seam with smaller inclination angle and shallow burial depth was slightly symmetrical on both sides, and the surface subsidence caused by each mining was small. In such cases the rock strata disturbance, including tensile and shear fractures and the associated large-scale rock mass subsidence, caused by the coal seam mining is restricted to localized areas. When mining the lower coal seam with a large inclination angle and great burial depth, affected by the surface deformation that had occurred, the surface displacement value becomes larger and appeared asymmetrical like a parabola. The subsidence range at the end of the mining area with greaterburial depth was larger, as additional fractures are formed across the upper rock mass above the coal seams, eventually causing larger areas of rock mass subsidence. The discrete model can more realistically simulate the deformation and damage characteristics of the rock mass, and its simulated maximum surface settlement value caused by rock layer rupture and instability was greater than that of the continuous model. This study will provide a theoretical basis for monitoring, early warning, prevention, control and treatment of geological hazards in coal mining subsidence areas.

In order to prevent the accidents in underground coal mines, an index system for safety supervision ability evaluation is established in the present paper. The index system is composed of goal layer, criteria layer (includes five indexes) and sub-criteria layer (includes 19 indexes). Based on the Analytical Hierarchy Process (AHP) method, the weight of each index within the system is determined. The analysis results indicate that the excess production is one of main indexes influencing the safety supervision ability. Subsequently, the causes underlying the excess production in underground coal mines are analyzed based on the 24Model. Moreover, the management measures are put forward. Thereafter, the safety supervision model for coal mine is proposed to prevent excess production. The functions of the proposed model are introduced in detail. Finally, Shanxi province is taken as a practical example to verify the capability of the established model. The application results show that the safety supervision model can realize the accurate monitoring of coal mine production, as well as improve the efficiency of safety supervision. With the model, the safety supervision ability of the branch of coal mine safety administration is effectively enhanced.

Underground coal mining presents significant health and safety challenges, primarily due to the accumulation and emission of hazardous gases. This study investigates the occurrence, impact, and control of such gases at the Yayo Underground Coal Mine, located in the Illubabor Zone of the Oromia Region, Ethiopia. The Yayo mine, being one of the few operational underground coal mines in the country, provides a relevant case for assessing gas-related risks and developing appropriate safety strategies in Ethiopia. Field investigations, interviews, and review of operational records revealed the presence of four major hazardous gases: methane (CH₄), carbon monoxide (CO), carbon dioxide (CO₂), and hydrogen sulfide (H₂S). These gases are released from coal seams, equipment operations, and decomposition of organic matter, and pose serious risks including fire and explosion, oxygen displacement, and respiratory illnesses. The study found that some areas of the mine experience gas concentrations that approach or exceed acceptable exposure limits, endangering both miners' health and mine product. The assessment also highlighted several limitations in the current mitigation measures at Yayo, including inadequate ventilation systems, the absence of real-time gas monitoring devices, limited use of personal protective equipment (PPE), and lack of worker training. These deficiencies significantly increase the risk of gas-related accidents and long-term health effects. To improve safety, the study recommends the implementation of modern mechanical ventilation systems, deployment of portable and fixed gas detectors, and mandatory use of self-contained breathing apparatus (SCBA) in high-risk zones. Furthermore, it advocates for enhanced government regulation, periodic safety audits, and structured training programs for mine workers.

With the rapid development of Internet of Things technology, intelligent signal detection has become a highly challenging key issue. Traditional manual feature extraction methods are no longer able to cope with complex and ever-changing practical environments. In response to this issue, this study proposes an intelligent signal detection model for the Internet of Things based on deep learning. This model adopts a convolutional neural network structure, which can automatically learn local pattern features in signals, thereby improving its adaptability to complex environments. Research a large-scale wireless signal dataset constructed based on real IoT scenarios, and adopt reasonable data preprocessing and augmentation strategies. Through scientific model training and multi-dimensional evaluation index settings, a comprehensive evaluation of the proposed model was conducted on the test set. The experimental results show that the model has achieved excellent performance over traditional methods in key indicators such as accuracy, recall, and F1 score, especially in complex signal recognition scenarios.

Investigation of sudden faults instability induced by coal mining

Regional surface subsidence induced by underground coal mining is very common and is a serious environmental problem in China. The mining subsidence not only causes damages to surface buildings but also change the pattern of surface drainage in a densely urbanized area. Monitoring and Analyzing the spatial distribution of the endangered surface may be helpful for land-use planning and for land reclamation. Interferometry SAR (InSAR) can be used to effectively monitor the succession of the spatial extent and the magnitude of subsidence in coal mining areas. In this paper, the multi-temporal D-InSAR method was applied for the generation of deformation map in coal mining area. With the "Interferometry Coherence Estimation Minimum Span Tree (ICEMST)" model, the optimized SAR images combination for D-InSAR processing for long term surface subsidence monitoring were predicted. With the estimation of ICEMST, several scenes SAR SLC data with time spanned more than half a year and spatial baseline more than 400m long were combined for D-InSAR processing to study the succession of land subsidence induced by underground coal mining and groundwater exploration in the test site, Kailuan coal mine area, a typical mining industrial area in north China, which has 125 years coal mining history. After being processed with the conventional "2 Pass" differential InSAR method, the deformation caused by underground coal mining in the line of sight (LOS) was transformed into vertical subsidence map. The experiment shows that the short time span is more suitable for D-InSAR application in mining areas than the long time span, for the lower correlation due to densely growing vegetation, seasonal changing factors and large water plashes. The time decorrelation, spatial decorrelation and the D-InSAR error resources were analyzed and discussed, and the Connor Reflectors method integrated with D-InSAR and GPS are presented, which will be a key practical technology for information obtain in digital mine.

This study compared conditions, practices, and attitudes at underground bituminous coal mines having low injury incidence rates with those found at mines having high injury incidence rates. Several characteristics common to many of the low incidence rate mines that differentiate them from those having high incidence rates were identified. (1) Training programs: adequate and relevant training materials; qualified instructors; restricted classroom size to encourage student participation; and tailored to meet individual miner needs. (2) Management/labor relations tend to have a positive impact upon a mine's accident and injury experience when: both management and labor have a positive attitude toward safety and health; open lines of communication permit management and labor to jointly reconcile problems affecting safety and health; representatives of labor become actively involved in issues concerning safety, health and production; and management and labor identify and accept their joint responsibility for correcting unsafe conditions and practices. (3) Safety and health conditions are improved when: standard operating procedures are established, understood, and implemented; management equitably enforces established policies concerning absenteeism, job assignments, and standard operating procedures; formal safety and health programs are communicated to all employees and subsequently implemented by management and labor; safety department has top management support in terms of funds, manpower, and the authority necessary to implement the safety and health program; mine plans are thoroughly reviewed by management, labor, and MSHA to insure that such plans incorporate measures to adequately control the physical environment of a coal mine; and MSHA inspection activity is most effective when the inspectors encourage increased cooperative interaction between themselves, mine management, and labor.

A considerable number of coal mines employed room and pillar mining in the last century in northern China, where the goaf remained stable for a period of time; however, with the increased exposure of coal pillars, their collapse may gradually increase. The stability assessment of these old rooms and pillar goafs is challenging due to their concealment, irregular mining patterns, and the long passage of time. The methodology developed in this study, based on “space-sky-earth” remote sensing such as InSAR to trace historical deformation, the UAV observation of current surface damage, and comparison of mining spaces, can rapidly detect on a large scale the collapse of old goafs and the trend of damage. This study is conducted with an example of a coal mine in Yulin, Northern China, where obtained quantitative surface deformation values were integrated with qualitative surface damage interpretation results, followed by a yearly analysis of the overlying rock movement in accordance with the underground coal mining process. The results show that from 2007 to 2021, corresponding surface deformation and damage occurred following mining progress. However, the room and pillar goaf areas had not undergone any surface deformation, nor had there been incidents of landslides or ground fissures; therefore, it was speculated that no roof collapse had occurred in this region. The surface deformation and damage associated with underground coal mining are complex and influenced by the coal seam occurrence, mining methods, strata lithology, terrain slope, temporal evolution, and anthropogenic modifications. These phenomena are representative of the coal mining area, and this methodology can provide a reference for similar endeavors.

In view of the problems existing in the present transportation ways of coal mine, a kind of new auxiliary haulage system has been designed, which can achieve the going there and back transportation only by using one winch in the multi-slope and undulating roadway. And according to the requirements of safety production of coal mine, the non-wheeled supply mine car without anti-car device and the safety interlock system of wheeled supply mine car have been designed. Meanwhile, finite element analysis software is adopted to check and analyze the locking pawl strength of the safety interlock system with wheeled supply mine car. The analyzing result shows that the locking pawl can completely stand the impact loading when the mine car sliding down. The haulage costs and the occurrence of security incidents have been reduced when the designed system is applied to coal mine. Engineering practice shows that the good results have been achieved.

Strata control plays a critical role in ensuring the safety and stability of undergr ound coal mining operations, especially in India’s increasingly complex geo-mining environments. With the growing shift from conventional to mechanised mining methods, the industry faces intensified challenges related to stress redistribution, reduced pillar stability, and frequent occurrences of roof falls. Strata control is a major challenge in underground coal mining and needs special attention throughout the mining operations. Since ancient times, safety has been given top priority in underground coal mines, where accidents related to strata control, like the fall of the roof and sides, have occurred frequently. These accidents and fatality figures have caused concern for the Indian coal mining industry and have drawn attention to focus on advanced technologies for safe extraction in mines. This paper investigates the primary challenges associated with strata control in underground coal mines and emphasises the importance of real-time monitoring and advanced instrumentation in mitigating these risks. It explores the limitations of traditional support systems, the complexities of predicting rock mass behaviour under dynamic conditions, and the inadequacies of conventional monitoring tools in harsh underground conditions. Furthermore, the study advocates for the adoption of integrated, data-driven approaches involving IoT-enabled sensors, wireless monitoring networks, and AI-based predictive analytics to enhance strata management. Major Findings: The findings underscore the urgent need for research and development in durable instrumentation, improved data interpretation systems, and the development of adaptive support technologies to ensure safer, more efficient, and sustainable underground mining practices in India.

From the viewpoint of safety in underground coal mining, the most suitable mining panel is the one with minimum geological structures, the right machinery, and equipment selection, trained employee, and proficient stope management. Since the ground parameters are uncontrollable and inherent uncertainties exist, a high percent of risk will usually accompany the underground coal mining activities. The main purpose of this study is to present a geological–geotechnical risk assessment model for identification of high risk-prone areas in underground coal mines using an integrated GIS-geostatistics system. Tabas as the first mechanized and largest underground coal mine in Iran was selected as a case study in this study. Gas content of coal seam, Coal Mine Roof Rating (CMRR), initial in situ stress state, fault throw, and orientation were selected as hazard/risk factors. For estimating the amount of coal seam gas content, CMRR and initial in situ stress in unsampled areas and providing the prediction maps, geostatistics module in ArcGIS was used. Rock engineering system–interaction matrix method was used for attribute weight assignment. Next, the attribute layers were weighted, rated, and overlaid to create a final map of geohazards risk. The analysis results of final risk map indicate that about 45% of under study area is prone to high to very high geohazards risk. Comparison of the results with experiences obtained during the early part of the mine and mined-out panels showed generally good agreement with promising ideas. This highlights the potential application of the GIS-based approach for hazards detection and geohazards risk assessment in underground coal mines.

There are three opencast Coal Mine pits in the area, namely Pandharkawda OC, Marki OC, and Mini Marki OC. The Opencast Pit at Marki has an area of 44 ha and will attain a maximum depth of 50 mbgl with open pit mine life of 10 years. Pandharkawda Opencast is having an area of 40 ha with maximum depth of 45 mbgl having opencast mine life of 9 years. Mini Marki OC is the third small Opencast Pit which will start production in 28th Year i.e., after the closure of Pandharkawda Underground. It has a small area of 7.25 ha with life span of 3 Year (28th to 30th Year). Thus, the total opencast mining area is 91.25 ha. In the area, two underground (UG) patches are proposed namely Marki and Pandharkawda UG which will commence production after completion of opencast workings. These sections will attain a maximum depth of 150 mbgl. Life span of Marki UG will be 20 Years (10th to 27th Year) and that of Pandharkawda UG will be 18 Years (9th to 27th Year). Coal Mine is producing coal @ 0.30MTPA peak since 2017. The Coal Mine was operated for 3 years from 2011 to 2013 and again resumed production since 2017. Hence, the opencast mine Pandharkawda OC, Marki OC will last its operation till first 9 years, and underground Coal Mine will continue from 9th year onwards till 27th year. Total planned mine life of Marki Mangli-I Coal Mine is around 20 years whereas Mini Marki OC will start its operation from 27th years onwards till 30th year. Study for prediction of groundwater withdrawal for subjected Coal Mine was carried out pre-monsoon and post-monsoon period in year 2016 to quantify the groundwater withdrawal from coal mining operations. The predicted quantity of groundwater withdrawal from Coal Mine is used by mine management to create groundwater recharge structure to compensate the groundwater discharge during mine operations.

Purpose – The high-velocity wind caused by a methane-air explosion is one of the important hazardous effects in explosion events of coal mines, and, however, until now it has not been received much attention from scientific works. The paper aims to discuss this issue. Design/methodology/approach – In consideration of the difficulties in observing particle velocities of high-velocity flows, this work presented a study to reveal the regularity during a methane-air explosion happening in the tunnel of coal mine through the numerical analysis approach. Findings – The strong wind caused by a methane-air explosion is a significant hazard and can cause damage in the accidents of methane-air explosion in underground coal mines, especially at structural opening, according to this work. Obtained results show that maximum particle velocity of the high-velocity wind occurs in the outside region of the premixed area, with a peak value of 400∼500 m/s, and the peak velocity of the high-velocity wind decreases exponentially with distance beyond the premixed area. Originality/value – The objective of this work was to examine the effect of wind caused by a methane-air explosion in a tunnel. Other information, such as shock wave and flame and temperature distribution, has been reported in the previous literatures. However, in the accidents of methane-air explosion in underground coal mines, some phenomena (structural opening is destroyed badly) can not be understood by using shock wave and flame and temperature distribution caused by the explosion. The strong wind caused by a methane-air explosion is another significant hazard and can cause damage in the methane-air explosion accidents in underground coal mines, especially at structural opening, according to this work.

Induced seismicity is common in underground mining, particularly in mines that cave by design, such as longwall coal mining. Seismic monitoring is one of the few remote sensing technologies that provide an understanding of a mine's response to extraction. However, its effectiveness usually depends on understanding event locations in relation to the mine workings. A significant source of uncertainty in event locations is the velocity model: a description of the speed at which seismic energy propagates. A nondestructive, repeatable compressed load column seismic source (CLCSS) was developed for velocity model calibration in underground coal mines. This paper describes the CLCSS and its application at a longwall coal mine. Signals from the CLCSS were detected up to 950 m epicentrally (1,100 m hypocentrally) from the source. Using the ground-truthed signals from the seismic source, we estimate a velocity model for locating events from a surface geophone array. Model performance is evaluated by relocating mining-induced events with well-constrained locations. A three-layer model with P-wave velocities ranging from 3.6 km/s to 4.4 km/s and a VP/VS ratio of 2.1 performed best with location errors of approximately 100 m, which is sufficient for many applications of seismic monitoring in coal mines. INTRODUCTION Coal bursts—violent dynamic failures which cause damage to mine openings—were first documented in European coal mines well before World War I (Guan et al., 2009) and continue to threaten underground coal miners around the world. For example, approximately 280 significant bursts occurred in U.S. coal mines between 1983 and 2017, seven of which resulted in fatalities (Mark, 2018). In China, over 200 coal mines have reported bursts cumulatively resulting in over 1,000 injuries and 100 fatalities in the past 10 years (Rong et al., 2022). Poland's Upper Silesian Coal Basin has experienced over 100 significant events, some of which resulted in injuries and fatalities (Mutke et al., 2015). Several other countries have also reported coal mine bursts, including Japan, Australia, India, France, South Africa, Czechoslovakia, Canada, Germany, and Russia (Lama and Bodziony, 1998). Although bursting mechanisms are not well understood and vary considerably, studies have identified risk factors including depth of cover or rapid changes in topographic relief, thick brittle strata near the coal seam, inadequately designed pillars, multi-seam interactions, and a variety of other mining and geological factors (Mark and Guana, 2016).