Evaluation of Convolutional-Long Short-Term Memory Model in Predicting Surface Displacement of Underground Mines

Surface stability is essential in underground mines health management systems. Unexpected Surface displacement in underground mines could lead to loss of lives, injuries, and economic losses. To reduce or neutralise the adverse effects of surface displacement, it is vital to monitor and accurately predict them and unravel their mechanisms. In recent years, Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM) have proven effective in predicting complex problems. However, CNN neglects the dynamic dependency of the input in the temporal dimension, which affects surface displacement features. The Convolutional-LSTM model can dynamically learn the temporal dependency among input features via the feedback connections in the LSTM to improve accurate captures of surface displacement features. This study focused on evaluating the C-LSTM model in predicting surface displacement of underground mines and assessed the predictive capabilities and generalisation strength of using hybridised ANN models. Geodetic and geotechnical data gathered from a Gold Mine in Ghana was used. The three models were tested on experimental data collected at Monitoring Scan Point 3. It was observed from the prediction output that all the methods could provide applicable and practical results. However, using indicators like root mean square error (RMSE) and correlation coefficient (R) in assessing the output of the prediction, the C-LSTM had the best prediction output. This study contributes to the advancement of accurate and efficient prediction of surface displacement of underground mines, ultimately enhancing and assisting safety operations.

Artificial ground freezing of underground mines in cold regions using thermosyphons with air insulation

Current practice of underground artificial ground freezing(AGF)typically involves huge refrigeration systems of large economic and environmental costs.In this study,a novel AGF technique is proposed deploying available cold wind in cold regions.This is achieved by a static heat transfer device called thermosyphon equipped with an air insulation layer.A refrigeration unit can be optionally integrated to meet additional cooling requirements.The introduction of air insulation isolates the thermosyphon from ground zones where freezing is not needed,resulting in:(1)steering the cooling resources(cold wind or refrigeration)towards zones of interest;and(2)minimizing refrigeration load.This design is demonstrated using well-validated mathematical models from our previous work based on two-phase enthalpy method of the ground coupled with a thermal resistance network for the thermosyphon.Two Canadian mines are considered:the Cigar Lake Mine and the Giant Mine.The results show that our proposed design can speed the freezing time by 30%at the Giant Mine and by two months at the Cigar Lake Mine.Further,a cooling load of 2.4 GWh can be saved at the Cigar Lake Mine.Overall,this study provides mining practitioners with sustainable solutions of underground AGF.

Predicting pillar stability for underground mine using Fisher discriminant analysis and SVM methods

The purpose of this study is to apply some statistical and soft computing methods such as Fisher discriminant analysis （FDA） and support vector machines （SVMs） methodology to the determination of pillar stability for underground mines selected from various coal and stone mines by using some index and mechanical properties, including the width, the height, the ratio of the pillar width to its height, the uniaxial compressive strength of the rock and pillar stress. The study includes four main stages： sampling, testing, modeling and assessment of the model performances. During the modeling stage, two pillar stability prediction models were investigated with FDA and SVMs methodology based on the statistical learning theory. After using 40 sets of measured data in various mines in the world for training and testing, the model was applied to other 6 data for validating the trained proposed models. The prediction results of SVMs were compared with those of FDA as well as the measured field values. The general performance of models developed in this study is close; however, the SVMs exhibit the best performance considering the performance index with the correct classification rate Prs by re-substitution method and Pcv by cross validation method. The results show that the SVMs approach has the potential to be a reliable and practical tool for determination of pillar stability for underground mines.

Safety analysis of Sormeh underground mine to improve sublevel stoping stability

In underground mines,sublevel stoping is used among a variety of different methods for mining an orebody,which creates large underground openings.In this case,the stability of these openings is affected by a number of factors,including the geometrical characteristics of the rock and mining-induced stresses.In this study,a sensitivity analysis was conducted with the numerical,squat pillar,and Mathews stability methods using the Taguchi technique to properly understand the influence of geometric parameters and stress on stope stability according to Sormeh underground mine data.The results show a full factorial analysis is more reliable since stope stability is a complex process.Furthermore,the numerical results indicate that overburden stress has the most impact on stope stability,followed by stope height.However,the results obtained with Mathews and squat pillar methods show that stope height has the greatest impact,followed by overburden stress and span.It appears that these methods overestimate the impact of stope height.Therefore,it is highly recommended that Mathews and squat pillar methods should not be used in high stope that is divided with several sill pillars.Nonetheless,Mathews method cannot accurately predict how the sill pillar impacts the stope stability.In addition,numerical analysis shows that all geometric parameters affect the roof safety factor,whereas the sill pillar has no significant influence on the safety factor of the hanging wall,which is primarily determined by the stope height–span ratio.

Identifcation of large-scale goaf instability in underground mine using particle swarm optimization and support vector machine

An approach which combines particle swarm optimization and support vector machine(PSO–SVM)is proposed to forecast large-scale goaf instability(LSGI).Firstly,influencing factors of goaf safety are analyzed,and following parameters were selected as evaluation indexes in the LSGI:uniaxial compressive strength(UCS)of rock,elastic modulus(E)of rock,rock quality designation(RQD),area ration of pillar(Sp),the ratio of width to height of the pillar(w/h),depth of ore body(H),volume of goaf(V),dip of ore body(a)and area of goaf(Sg).Then LSGI forecasting model by PSO-SVM was established according to the influencing factors.The performance of hybrid model(PSO+SVM=PSO–SVM)has been compared with the grid search method of support vector machine(GSM–SVM)model.The actual data of 40 goafs are applied to research the forecasting ability of the proposed method,and two cases of underground mine are also validated by the proposed model.The results indicated that the heuristic algorithm of PSO can speed up the SVM parameter optimization search,and the predictive ability of the PSO–SVM model with the RBF kernel function is acceptable and robust,which might hold a high potential to become a useful tool in goaf risky prediction research.

Development of underground mine monitoring and communication system integrated ZigBee and GIS

An automated underground mine monitoring and communication system based on the integration of new technologies is introduced to promote safety and health,operational management and cost-effectiveness.The proposed system integration considering Wireless Sensor Network(WSN) assisted Geographic Information System(GIS) enables to monitor and control underground mining applications from surface office.Based on the capabilities of WSNs,ZigBee network is adapted for near real-time monitoring,ventilation system control and emergency communication in underground mine.ZigBee nodes were developed to sense environmental attributes such as temperature,humidity and gases concentration;switching ON and OFF ventilation fans;and texting emergency messages.A trigger action plan for monitored attributes above normal and threshold value limits is programmed in the surface GIS management server.It is designed to turn the auxiliary fans on remotely or automatically in orange condition and sending evacuation messages for underground miners in unsafe(red) condition.Multi-users operation and 3D visualisations are other successful achievements of the proposed system for the underground monitoring and communication.

Sensor deployment strategy for chain-type wireless underground mine sensor network

Wireless sensor networks (WSNs) are very important for monitoring underground mine safety. Sensor node deployment affects the performances of WSNs. In our study, a chain-type wireless underground mine sensor network (CWUMSN) is first pre- sented. A CWUMSN can monitor the environment and locate miners in underground mines. The lowest density deployment strate- gies of cluster head nodes are discussed theoretically. We prove that the lifetime of CWUMSN with a non-uniform deployment strategy is longer than with a uniform deployment strategy. Secondly, we present the algorithm of non-uniform lowest density de- ployment of cluster head nodes. Next, we propose a dynamic choice algorithm of cluster head nodes for CWUMSN which can im- prove the adaptability of networks. Our experiments of CWUMSN with both non-uniform lowest density and uniform lowest den- sity deployments are simulated. The results show that the lifetime of CWUMSN with non-uniform lowest density deployment is almost 2.5 times as long as that of the uniform lowest density deployment. This work provides a new deployment strategy for wire- less underground mine sensor networks and then effectively promotes the application of wireless sensor networks to underground mines.

Determination of velocity correction factors for real-time air velocity monitoring in underground mines

When there are installations of air velocity sensors in the mining industry for real-time airflow monitoring, a problem exists with how the monitored air velocity at a fixed location corresponds to the average air velocity, which is used to determine the volume flow rate of air in an entry with the cross-sectional area. Correction factors have been practically employed to convert a measured centerline air velocity to the average air velocity. However, studies on the recommended correction factors of the sensor-measured air velocity to the average air velocity at cross sections are still lacking. A comprehensive airflow measurement was made at the Safety Research Coal Mine, Bruceton, PA, using three measuring methods including single-point reading, moving traverse, and fixed-point traverse. The air velocity distribution at each measuring station was analyzed using an air velocity contour map generated with Surfer~. The correction factors at each measuring station for both the centerline and the sensor location were calculated and are discussed.

Location estimation of autonomous driving robot and 3D tunnel mapping in underground mines using pattern matched LiDAR sequential images

In this study,a machine vision-based pattern matching technique was applied to estimate the location of an autonomous driving robot and perform 3D tunnel mapping in an underground mine environment.The autonomous driving robot continuously detects the wall of the tunnel in the horizontal direction using the light detection and ranging(Li DAR)sensor and performs pattern matching by recognizing the shape of the tunnel wall.The proposed method was designed to measure the heading of the robot by fusion with the inertial measurement units sensor according to the pattern matching accuracy;it is combined with the encoder sensor to estimate the location of the robot.In addition,when the robot is driving,the vertical direction of the underground mine is scanned through the vertical Li DAR sensor and stacked to create a 3D map of the underground mine.The performance of the proposed method was superior to that of previous studies;the mean absolute error achieved was 0.08 m for the X-Y axes.A root mean square error of 0.05 m^(2)was achieved by comparing the tunnel section maps that were created by the autonomous driving robot to those of manual surveying.

High Efficiency Dust Scrubbers for Continuous Miner in Underground Mines

Based on the experimental data by a full scale test model and the relevant existing achievements, four new concepts of group dust scrubbers were designed in the paper. The new dust scrubbers consist of two to four common dust separators in series: wet fan, wet Venturi and foam filter bed. Wetting and foaming agents are used to increase the efficiency of dust separation. High efficiency is not required for each part of the group dust scrubbers, but the whole system has a high working reliability. All parts of the group dust scrubbers have the most suitable separating efficiency for different size and concentration of dust particles in airflow, according to their technical features. Four group dust scrubbers have a high efficiency from 94.4% to 99.7% for separating respirable dust at a rational cost.

A General Channel Model for Visible Light Communications in Underground Mines

In underground mines, visible light communication(VLC) system is a promising method to realize effective communication,which supports communication and illumination at the same time. Therefore, adequate research of underlying physical propagation phenomenon should be carried out to realize VLC system in underground mines. To design VLC system and evaluate its performance, accurate and efficient channel models, including large-scale fading and scattering characteristics, are needed to be established. However,the characteristics of the underlying VLC channels about fading and scattering have not been sufficiently investigated yet. In this paper, a path loss channel model, based on the recursive model, is proposed precisely. Its path loss exponent is determined by three different trajectories, which is studied in the mining roadway and working face environment. Besides, the shadowing effect for VLC has been modelled by a Bimodal Gaussian distribution in underground mines. Considering the number of transmitters in line-of-sight(Lo S) as well as non-line-of-sight(NLo S) scenarios,our simulation illustrates the fact that, as the curve fitting technique is employed, the path loss displays a linear behavior in log-domain.The path loss expression is derived, it is related to the distance. Finally, root mean square(RMS) delay spread and Mie scattering in underground mines are analyzed.

Design of a Mobile Mechanism for Missing Miner Search Robots in Underground Mines

A mobile mechanism with four tracked-units for a missing miner search robot (MMSR) is presented, with a design based on the terrain features and atrocious environment of an underground mine. Its structure and working prin- ciple is discussed. The four tracked-units are controlled independently and driven cooperatively. By means of two DC motors being controlled respectively, one tracked-unit can accomplish two types of driving mode: tracked travel and in- tegral unit legged rotation (IULR), forming a track-legged compound function mechanism. Its capabilities of surmount- ing obstacles and its toppling stability in underground mines have also been analyzed. The results show that the mobile mechanism can directly surmount an obstacle of the height less than the length of one tracked-unit and get across a raceway with a span less than the length of one tracked-unit by using tracked travel and IULR. Its unstable slope angle is 51.3°. Toppling stability is determined by its structural size, moving direction and slope angle. IULR of four tracked-units can adjust the robot’s posture and then enhance toppling stability or assist in surmounting obstacles. Its track-legged compound function mechanism makes it suitable for working in underground mines.

Diesel engine exhaust exposures in two underground mines

Exposure to diesel engine exhaust(DE) is a major concern in underground mines. It has been linked to cardiopulmonary diseases and is classified as a human carcinogen. The goal of this study is to assess DE exposures in workers at two underground gold mines, to compare exposure levels within and between the mines, and to compare different methods of measuring DE exposures, namely respirable combustible dust(RCD), elemental carbon(EC) and total carbon(TC). Ambient and personal breathing zone(PBZ) measurements were taken. Side-by-side monitoring of RCD and of the respirable fraction of EC and TC(EC_Rand TC_R) was carried out in the workers' breathing zone during full-shift work.Regarding ambient measurements, in addition to EC_R, TC_Rand RCD, a submicron aerosol fraction(less than 1 mm) of EC and TC was also sampled(EC_1and TC_1). Average ambient results of 240 mg/m^3 in RCD, 150 mg/m^3 in EC_Rand 210 mg/m^3 in TC_Rare obtained. Average PBZ results of 190 mg/m^3 in RCD,84 mg/m^3 in EC3Rand 150 mg/min TC_Rare obtained. Very good correlation is found between EC_Rand EC_1 with a Pearson correlation coefficient of 0.99(p < 0.01) calculated between the two logtransformed concentrations. No differences are reported between EC_Rand EC_1, nor between TC_Rand TC_1, since ratios are equal to 1.04, close to 1, in both cases. Highest exposures are reported for loadhaul-dump(LHD) and jumbo drill operators and conventional miners. Significant exposure differences are reported between mines for truck and LHD operators(p < 0.01). The average TC_R/EC_Rratio is 1.6 for PBZ results, and 1.3 for ambient results. The variability observed in the TC_R/EC_Rratio shows that interferences from non-diesel related organic carbon can skew the interpretation of results when relying only on TC data.

Energy optimal routing for long chain-type wireless sensor networks in underground mines

Wireless sensor networks are useful complements to existing monitoring systems in underground mines. They play an important role of enhancing and improving coverage and flexibility of safety monitoring systems.Regions prone to danger and environments after disasters in underground mines require saving and balancing energy consumption of nodes to prolong the lifespan of networks.Based on the structure of a tunnel,we present a Long Chain-type Wireless Sensor Network（LC-WSN）to monitor the safety of underground mine tunnels.We define the optimal transmission distance and the range of the key region and present an Energy Optimal Routing（EOR）algorithm for LC-WSN to balance the energy consumption of nodes and maximize the lifespan of networks.EOR constructs routing paths based on an optimal transmission distance and uses an energy balancing strategy in the key region.Simulation results show that the EOR algorithm extends the lifespan of a network,balances the energy consumption of nodes in the key region and effectively limits the length of routing paths,compared with similar algorithms.

Application of low-cost particulate matter sensors for air quality monitoring and exposure assessment in underground mines:A review

Exposure to mining-induced particulate matter(PM)including coal dust and diesel particulate matter(DPM)causes severe respirat-ory diseases such as coal workers’pneumoconiosis(CWP)and lung cancer.Limited spatiotemporal resolution of current PM monitors causes miners to be exposed to unknown PM concentrations,with increased overexposure risk.Low-cost PM sensors offer a potential solution to this challenge with their capability in characterizing PM concentrations with high spatiotemporal resolution.However,their application in underground mines has not been explored.With the aim of examining the potential application of low-cost sensors in underground mines,a critical review of the present status of PM sensor research is conducted.The working principles of present PM monitors and low-cost sensors are com-pared.Sensor error sources are identified,and comprehensive calibration processes are presented to correct them.Evaluation protocols are pro-posed to evaluate sensor performance prior to deployment,and the potential application of low-cost sensors is discussed.

A hybrid tubular standing support for underground mines:Compressive behaviour

This paper presents the development of an innovative standing support for underground mines.The main feature of this standing support is its exterior container,a combination of polyvinyl chloride(PVC)with large rupture strain and fibre-reinforced polymer(FRP)with high strength-to-weight ratio.To demonstrate the advantages of this cementitious grout filled PVC-FRP tubular(PFT)standing support,a series of compression tests were conducted.Test variables included the strength of cementitious grout infill material and the thickness of FRP jacket.Compression tests were also conducted on cementitious grout-filled PVC tubular(PT)support and cementitious grout-filled FRP tubular(FT)support.These tests showed that PFT support presents a typical strain-hardening behaviour together with an outstanding axial deformation ability(>20%of the overall height of the support).In addition,the maximum compressive strength of PFT support is much higher than that of the corresponding PT support and FT support.Furthermore,using thicker FRP jacket or high strength cementitious grout material can enhance the load carrying capacity of PFT support.These comparative results indicated that the high performance of PFT support is mainly attributed to the combination of confining constituents(i.e.PVC and FRP)and infill material.

Self-Driving Algorithm and Location Estimation Method forSmall Environmental Monitoring Robot in Underground Mines

In underground mine environments where various hazards exist,such as tunnel collapse,toxic gases,the application of autonomous robots can improve the stability of exploration and efficiently perform repetitive exploratory operations.In this study,we developed a small autonomous driving robot for unmanned environmental monitoring in underground mines.The developed autonomous driving robot controls the steering according to the distance to the tunnel wall measured using the light detection and ranging sensor mounted on the robot to estimate its location by simultaneously considering the measured values of the inertial measurement unit and encoder sensors.In addition,the robot autonomously drives through the underground mine and performs environmental monitoring using the temperature/humidity,gas,and particle sensors mounted on the robot.As a result of testing the performance of the developed robot at an amethyst mine in Korea,the robot was found to be able to autonomously drive through tunnel sections with∼28 m length,∼2.5 m height,and∼3 m width successfully.The average error of location estimation was approximately 0.16 m.Using environmental monitoring sensors,temperature of 15–17◦C,humidity of 42%–43%,oxygen concentration of 15.6%–15.7%,and particle concentration of 0.008–0.38 mg/m3 were measured in the experimental area,and no harmful gases were detected.In addition,an environmental monitoring map could be created using the measured values of the robot’s location coordinates and environmental factors recorded during autonomous driving.

Preparedness of Swedish EMS Personnel for Major Incidents in Underground Mines

The purpose of this study was to survey the EMS （emergency medical services） personnel preparedness for major incidents in the underground mining industry in Sweden. Every year, a high number of incidents, workplace accidents and fires are reported from the Swedish mining industry. Taking care of patients located in an underground mine represents a challenge to EMS personnel. Today, knowledge about EMS personnel preparedness for major incidents in the mining industry is limited. The study design was a cross-sectional survey. The questionnaires were distributed to EMS personnel working in ambulance stations geographically located near an underground mine. Thirteen ambulance stations were included and 137 of 258 personnel answered. Demographic data were analyzed using descriptive statistics. Differences among groups were analyzed with the Chi-Squared test, continuity correction and t-test. Results showed about half of the participants reported that they do not feel prepared to work in a major incident in an underground mine. The majority wished to receive educational training to enhance their preparedness. The most commonly requested type of education was practical drills on the scene, in an underground mine. The reported preparedness was significantly higher among the participants who had received some kind of education, or had authentic experience of a mission in an underground mine than those who did not. This study reveals shortcomings in the preparedness of EMS personnel. The perceived low preparedness of EMS personnel may affect their ability to work in a major incident in the mining industry. Study findings may be used in planning the future education, including practical drills, of EMS personnel.

Automatic drawing of the geologic profile of an underground mine based on COMGIS

This paper introduces a method of building a prototype system of geologic profile auto-drawing.A.NET development platform and integrated environment was used along with a component based design,a B/S system model,and XML techniques.Knowledge rules for creating geologic profiles and generating virtual drilling data from existing bore data and expert,hand-drawn geologic profiles were acquired. Then a prototype system was established by utilizing the known knowledge rules,topological relationships, and semantic relationships among strata.This system has a friendly human-computer interface and can meet requirements of mutual queries between attribute and spatial data.The generated profile map is editable.This study provides a new powerful tool for underground mine work.

A review of the health effects and exposure-responsible relationship of diesel particulate matter for underground mines

The increasing use of diesel-powered equipment in confined spaces(underground mines) has the potential to over expose underground miners under the threat of diesel particulate matter(DPM). Miners in underground mines can be exposed to DPM concentrations far more than works in other industries. A great number of animal and epidemiological studies have shown that both short-term and long-term DPM exposure have adverse health effect. Based on reviews of related studies, especially some recent evidence, this paper investigated the long and short-term health effects based on animal studies and epidemiological studies. The exposure-response relationship studies were also explored and compared to the current DPM regulation or standards in some countries. This paper found that the DPM health effect studies specifically for miners are not sufficient to draw solid conclusions, and a recommendation limit of DPM concentration can be put in place for better protection of miners from DPM health risk. Current animal studies lack the use of species that have similar lung functions as human for understanding the cancer mode of action in human. And finally, the DPM health hazard will continue to be a challenging topic before the mode of action and reliable exposure-response relationship are established.