Introduction
The National Institute for Occupational Safety and Health (NIOSH) Dust Control Handbook (1) defines mine dust as small solid particles created by the breaking up of the larger particles. When suspended in air, these particles will become hazardous to workers’ health depending on their size. Particles ranging from 2 to 60 µm remain suspended in the air for long periods and may be inhaled by miners working underground. Dust particles < 10 µm are considered respirable. ISO 7708-1995 and the American Conference of Governmental Industrial Hygienist (ACGIH) determined the respirable criterion for dust, D50 to be 4 µm. Furthermore, the Mine Safety and Health Administration (MSHA) mandates a permissible exposure limit for respirable dust of 1.5 mg/m3, averaged over an 8 h shift, to prevent adverse health effects among miners.
Mine workers continue to suffer from overexposures to respirable dust, with debilitating health consequences that eventually lead to death. Colinet et al. (2) found the prevalence of coal workers’ pneumoconiosis (CWP) to be increasing between 2000 and 2006, showing CWP in almost 8 % of underground coal miners who had more than 25 years of mining experience. The U.S. National Academies of Engineering (3) report on “Monitoring and Sampling Approaches to Assess Underground Coal Mine Dust Exposures” documents that, since 2002 and especially since 2010, there have been significant increases in the prevalence of CWP in U.S. miners who have 15 or more years’ tenure in coal.
MSHA requires measurement of respirable dust by an approved continuous personal dust monitor (CPDM) for all underground miners exposed to respirable dust. Currently, there is one available commercial instrument that has been approved as a CPDM, Thermo ScientificTM PDM37001, which incorporates a real-time particulate monitor to measure respirable dust mass concentration, shift exposure, and accumulated exposure in real-time (4).
1 The mention of specific product names and manufacturers does not imply endorsement by the authors or the publisher.
It uses a tapered element oscillating microbalance (TEOM) to measure continuously the mass of collected particles. Airflow enters the PDM through a cyclone, drawn into a heated flow tube to be heated, then flows into mass transducer where the particulate matter is deposited onto the TEOM. After this, the air flows through temperature, relative humidity, and differential pressure sensors, and finally, through the pump to exit the system (5). The PDM3700 maintains a constant volumetric flow rate of about 2 l/min and reports the sample volumes in volumetric terms based on ambient temperature as measured near the cyclone.
PDM3700 samples, analyzes and calculates the mass-based concentration of the respirable dust. The mass calculation is based on the change in frequency of the TEOM. Figure 1 shows an extract of PDM data for one shift.
The blue line shows the dust mass concentration (Mass1), expressed in mg/-m3. Mass1 is normally additive, i. e. it should show a steady increase of dust mass over time. If the PDM gets shaken or bumped, as indicated by erratic “tilt” readings (green graph), dust mass may be shaken off the TEOM and lost. The PDM software then corrects the Mass1 reading to revert to the last good mass reading and continues counting, ignoring the lost dust mass. Plotted in red is the cumulative dust concentration, end-of-shift (EOS) forecast. In this case, the miner was exposed to a high dust concentration during the first 40 min of his/her shift, as indicated by the steep rise of the blue curve between 22:00 and 23:00. This coincides combined with the peak of the EOS estimate shown as the red curve. The tilt sensor indicates erratic data between 1:30 and 4:30 am, which resulted in several corrections of the dust mass, indicated by brief reductions in Mass1 (blue line) that were recovered a short while later.
Due to the end-of-shift averaging, the blue and red lines will converge at the end of the shift as both data streams mark the EOS dust concentration at that point. The slight divergence of about 0.1 mg/-m3 signifies an error in the recording that is acceptable given the systematic inaccuracies in the measurement process. Volkwein et al. (6), showed that the PDM had a 95 % confidence that the individual PDM measurements were within ± 25 % of the reference measurements.
Despite having a near real-time dust exposure vs. time record from the PDM3700, the instrument cannot track miners’ working locations along with their associated dust exposure levels. Location monitoring and tracking are necessary to identify sources and areas of excessive dust exposure, to predict hazardous exposure levels and to implement mitigating engineering and administrative controls before a miner is exposed.
The objective of this study is to improve the functionality of the PDM by adding geolocation information to the respirable dust exposure monitoring data. The location information can be extracted from a given miner tracking system such as Leaky-Feeder-based trackers, RFID tracking systems, or GPS where available. Dust exposure locations will be obtained by combining location and dust concentrations, and these locations will be visualized using geographic mapping tools. This information may then be used to design engineering and administrative controls to reduce or eliminate respirable dust exposure among miners, plant, and factory workers.
In this paper, the authors present the proof-of-concept for the proposed continuous dust monitoring with the localization system. For this purpose, researchers use the PDM and location data collected in the Edgar Experimental Mine of the Colorado School of Mines (CSM), Golden, CO/USA.
Methodology
The methodology used to develop a visualization tool for continuous dust monitoring with the localization system is shown in Figure 2.
In the first step, dust concentrations and location data are collected using PDM and the tracking system. In the second step, the time series of location and dust concentration values are combined. In the third step, spline interpolation technique is used to interpolate the collected dust concentration values to a raster surface along the data collection route. In the last step, the result of spline interpolation is used to create 3-D heat map visualization of dust concentration values.
Data collection
The researchers at the CSM have acquired a PDM3700 personal dust monitor from Thermo Fisher Scientific. They experimented with PDM in the Edgar Experimental Mine to demonstrate the capabilities of the proposed system. Since the geolocation capability of the system is still in the development stage, location information is collected manually. The authors marked the locations and the time while PDM records dust data in the mine. The locations where the PDM recorded dust readings in the Edgar Mine are acquired by matching the time readings of the PDM and manual location tracking. The route of PDM data collection in Edgar Mine is shown in Figure 3.
Pre-processing of PDM data
For each minute interval, the PDM records 30 min average dust concentrations for the prior 30 min interval, the cumulative dust concentration to that point in time, and an estimated end-of-shift projected concentration. The mass calculation is based on the change in the frequency of the tapered element. Using the frequency values, equation 1 determines the dust concentration values. K0 is the instrument calibration constant.
The K0 value obtained using equation 1 is used in the re-calculation of the Mass Total values. This calculation also overrules the PDM’s internal principle of only recording dust mass values if the difference is greater than 0.01 mg. Since the re-calculation considers the small changes in Mass Frequency values, more sensitive Mass Total values are obtained (Equation 2).
The PDM calculates Mass Total values each minute. From this data, equation 3 determines dust concentrations per minute:
Dust concentration values are shown in Figure 4. 30 min average and cumulative concentration values are obtained from the PDM, while the concentration per minute values are calculated using equation 3.
The graph shows that the dust concentrations are usually low in Edgar Mine, as evident from the dust data recorded during the first hour of testing. During the second hour of testing, researchers released compressed air in various locations to create artificial dust clouds. Recorded dust concentration peaks as indicated by the 30 min average and cumulative concentration values.
Spline interpolation
To create a heat map, researchers interpolated the PDM dust concentration values over the data collection route using spline interpolation. Spline estimates values using a mathematical function that minimizes the overall surface curvature. This results in a smooth surface that passes exactly through the input points (7).
A standard spline interpolation produces a raster surface over a specific geographic space. However, dust concentration events should be analyzed in terms of areas inside the network structure since, in underground mines, dust clouds can only occur along the network of mine entries. Therefore, the interpolation of dust concentration values should be limited to the grid of cells that belong to the Edgar mine drifts. This approach prevents the interpolation calculations outside the mine entries. Examples where spline interpolation with cell restrictions were used include travel-to-work analysis in sustainable urban development (8) and visualization of the depth of water channels (9). In this study, the Edgar mine drift network is obtained by creating a polygon feature that follows the data collection route (Figure 5).
In this study, researchers used the “Spatial Analyst” toolbox available in the ArcGIS software to apply spline interpolation on dust concentration values (10). In spline interpolation, the choice of output cell size is important since this parameter controls how smooth the results will be. For this study, after obtaining interpolation results for several cell sizes, a cell size of 0.15 m resulted in the most effective smoothing of dust concentration values. Figure 6 shows the result of spline interpolation. These results represent the end of shift dust exposure of a person if she stays in that location for a full 8 h shift.
The result of the spline indicates low dust concentration for most of the Edgar Mine. Several locations show medium levels of dust concentrations and a few locations exhibit dust concentrations above 1.5 mg/-m3. Location mapping clearly documents several, dictinct, high-dust areas. In comparison, the PDM-only, time series data in Figure 4 only reveals a single peak of dust concentrations.
Heat map generation
Spline interpolation results are commonly visualized in 3-D since it provides a more intuitive understanding of the results. Figure 7 visualized the raster surface in 3-D using the end of shift dust exposure values obtained from spline interpolation as elevation values. Elevations are amplified for better visualization of the higher dust concentrations.
Discussion
Researchers at the Colorado School of Mines experimented with personal dust monitor (PDM) data tied to location tracking to provide a proof-of-concept for continuous dust monitoring with a localization system. The method works with any miner tracking system, including RFID, GPS, Leeaky Feeder or even manual location tracking. Combining PDM dust output with coordinates using data collection time provides enough data to show the capabilities of the system.
PDM only provides cumulative dust concentration values or a 30 min moving average of dust concentrations which are not enough to use for identifying hot spots in the mine. Therefore, the researchers calculate dust concentrations for each minute using the mass frequency values as the basis of the calculation. This calculation provides dust concentration values that can be related to specific locations in the mine. However, the error rate may increase as the time interval for the dust concentration value decreases. Further investigation of the error propagation is required to understand the accuracy of the calculated values.
Dust concentration time series values were interpolated over the data collection route in Edgar using spline interpolation. The analysis showed that the dust concentrations in the Edgar Mine are consistently low. Researcher artificially created several hot spots with high dust levels. PDM output alone indicated only a single hot spot, but spline interpolation and location analysis confirmed a number of distinct locations with high dust concentrations. Results were visualized in 3-D as a heat map to provide a clear representation of the hot spot locations.
Conclusions
Measurement of respirable dust by a PDM is required for all underground miners exposed to respirable dust. The device measures respirable dust mass concentration, shift exposure and accumulated exposure in real-time. However, miners’ locations are not tracked along with their associated dust exposure levels. Location tracking is necessary to identify sources of excessive dust exposure and to implement mitigating engineering and administrative controls before a miner is exposed. To properly manage respirable dust exposures, operators need a tool that identifies hazardous dust sources by location as well as concentration.
The proposed system uses spline interpolation tool on minute by minute dust concentration values calculated based on the mass frequencies obtained by the PDM. The heat map is generated by using the raster surface obtained by spline interpolation. The heat maps obtained with this methodology provides a clear understanding of where the excessive dust exposures occur.
References / Quellenverzeichnis
References / Quellenverzeichnis
(1) Cecala, A. B. (2019): Dust control handbook for industrial minerals mining and processing 2nd edition. NIOSH. Pittsburgh PA.
(2) Colinet, J. F.; Rider, J. P.; Listak, J. M.; Organiscak, J. A.; Wolfe, A. L. (2010): Best practices for dust control in coal mining. U.S. Department of Health and Human Services, Public Health Service, Centers for Disease Control and Prevention, National Institute for Occupational Safety and Health DHHS Publication No. 2010-110 (IC 9517), pp 1 – 76.
(3) National Academies (2018): Monitoring and sampling approaches to assess underground coal mine dust exposures. National Academies Press Washington DC, DOI: 10.17226/2511.
(4) Halterman, A. (2018): Comparison of respirable mass concentrations measured by a personal dust monitor and a personal DTA RAM to gravimetric measurement. Annals of Work Exposures and Health, Vol. 62, No. 1, pp. 62 – 71.
(5) ThermoScientific, Operator’s Manual for PDM3700 Personal Dust Monitor, 25 March 2016: assets.thermofisher.com/TFSAssets/LSG/manuals/EPM-manual-PDM3700.pdf, Accessed October 2019.
(6) Volkwein, J. (2004): Performance of a new personal respirable dust monitor for mine use. US Department of Health and Human Services, Public Health Services, Center for Disease Control, National Institute for Occupational Safety and Health.
(7) Childs, C. (2004): Interpolating surfaces in ArcGIS spatial analyst. ArcUser, July-September, 3235 (569), pp 32 – 35.
(8) Wolny, A.; Ogryzek, M.; Zrobek, R. (2017): Challenges, opportunities and barriers to sustainable transport development in functional urban areas. Environmental Engineering: Proceedings of the International Conference on Environmental Engineering, 10, pp. 1 – 9.
(9) ESRI (2018): How to use spline with barriers to visualize the depth of water channels. Accessed from support.esri.com/en/technical-article/000014496, on July 7th 2020.
(10) ESRI (2019): ArcGIS Desktop: Release 10.7.1. Redlands, CA: Environmental Systems Research Institute.