sensors Article
A Methodology to Monitor Airborne PM10 Dust Particles Using a Small Unmanned Aerial Vehicle Miguel Alvarado 1, *, Felipe Gonzalez 2 , Peter Erskine 1 , David Cliff 3 and Darlene Heuff 4 1 2 3 4
*
Environment Centre, Sustainable Mineral Institute, The University of Queensland, 4072 Brisbane, Australia; [email protected] Science and Engineering Faculty, Queensland University of Technology (QUT), 4000 Brisbane, Australia; [email protected] People Centre, Sustainable Mineral Institute, The University of Queensland, 4072 Brisbane, Australia; [email protected] Advanced Environmental Dynamics Pty Ltd., Ferny Hills, 4055 Queensland, Australia; [email protected] Correspondence: [email protected]; Tel.: +61-7-3346-4065 or +61-7-3346-4086
Academic Editor: Assefa M. Melesse Received: 15 December 2016; Accepted: 4 February 2017; Published: 14 February 2017
Abstract: Throughout the process of coal extraction from surface mines, gases and particles are emitted in the form of fugitive emissions by activities such as hauling, blasting and transportation. As these emissions are diffuse in nature, estimations based upon emission factors and dispersion/advection equations need to be measured directly from the atmosphere. This paper expands upon previous research undertaken to develop a relative methodology to monitor PM10 dust particles produced by mining activities making use of small unmanned aerial vehicles (UAVs). A module sensor using a laser particle counter (OPC-N2 from Alphasense, Great Notley, Essex, UK) was tested. An aerodynamic flow experiment was undertaken to determine the position and length of a sampling probe of the sensing module. Flight tests were conducted in order to demonstrate that the sensor provided data which could be used to calculate the emission rate of a source. Emission rates are a critical variable for further predictive dispersion estimates. First, data collected by the airborne module was verified using a 5.0 m tower in which a TSI DRX 8533 (reference dust monitoring device, TSI, Shoreview, MN, USA) and a duplicate of the module sensor were installed. Second, concentration values collected by the monitoring module attached to the UAV (airborne module) obtaining a percentage error of 1.1%. Finally, emission rates from the source were calculated, with airborne data, obtaining errors as low as 1.2%. These errors are low and indicate that the readings collected with the airborne module are comparable to the TSI DRX and could be used to obtain specific emission factors from fugitive emissions for industrial activities. Keywords: PM10 ; monitoring; blasting; unmanned aerial vehicle (UAV); multi-rotor UAV; optical sensor
1. Introduction The use of Unmanned Aerial Vehicles (UAVs) to collect environmental data has increased exponentially in only a few years. Their capacity to reach places where humans could be at risk or it could become too expensive for small operations, makes them very useful for gas and particulate monitoring in the mining industry. Several authors have published articles where progress on the use of UAVs for industrial use or investigation has been made by using two UAVs simultaneously [1], combining multiple gas sensors and imagery [2,3], using gas tracking algorithms [4–6], using solar energy [7], and making use of microwave sensors [8].
Sensors 2017, 17, 343; doi:10.3390/s17020343
www.mdpi.com/journal/sensors
Sensors Sensors 2017, 2017, 17, 17, 343 343
22 of of 25 25
The methodology applied and tests undertaken for this investigation are a continuation of the methodology applied and undertaken for this investigation are a continuation workThe published by Alvarado et tests al. [9,10]. Improvements to the methodology include:of the work published by Alvarado et al. [9,10]. Improvements to the methodology include: the development development of a new modular sensor, focusing only on the use of a multi-rotor UAV with a tailored of a new modular sensor, on thetouse of a multi-rotor UAV withobtained a tailoredbysampling probe, sampling probe, and the focusing use of a 5only m tower validate the PM10 readings the UAV. The and the use of a 5 m tower to validate the PM readings obtained by the UAV. The emission rates emission rates obtained from a point source created for the flight tests, were evaluated to determine 10 obtained fromof a point source created forused the flight tests, to determine the factors. capability the capability the methodology to be at open pitwere mineevaluated sites to estimate emission In of the methodology be used at open pit sites to estimate emission factors. In the following the following sectionstothese experiments aremine described and their results discussed. sections these experiments are described and their results discussed. 2. Experimental Development 2. Experimental Development 2.1. Dust Monitoring Module 2.1. Dust Monitoring Module Following Alvarado et al. [9,10] a sampling module and data logger transmission system were Following Alvarado et al. [9,10] a sampling module and data logger transmission system were developed using the OPC-N2 particle counter from Alphasense (Great Notley, Essex, UK) [11] with a developed using the OPC-N2 particle counter from Alphasense (Great Notley, Essex, UK) [11] with a minimum response time of 0.2 s. In addition to the particle counter the dust monitoring module is minimum response time of 0.2 s. In addition to the particle counter the dust monitoring module is integrated with a: integrated with a: • • • • • •
Barometer, temperature and and humidity humidity sensor, sensor,BME280 BME280(Bosch, (Bosch,Reutlingen, Reutlingen,Germany); Germany); Barometer, temperature Raspberry board (Raspberry (Raspberry Pi Pi Foundation, Foundation, Cambridge, Cambridge, UK); UK); Raspberry Pi Pi 33 board uBlox LEA6 GPS (uBlox, Thalwil, Switzerland); uBlox LEA6 GPS (uBlox, Thalwil, Switzerland); XBee (XBee, Minnetonka, Minnetonka, MN, MN, USA); USA); XBee 2.4 2.4 GHz GHz Radio Radio (XBee, 3DR Airspeed Sensor (3DR, Berkeley, CA, USA); 3DR Airspeed Sensor (3DR, Berkeley, CA, USA); Lipo battery (3.7 V) (3DR, Berkeley, CA, USA). Lipo battery (3.7 V) (3DR, Berkeley, CA, USA).
The architecture of the dust monitoring module is presented in Figure 1. Information from each sensor (e.g., was sent to to a ground station viavia the the XBee 2.4 (e.g., dust, dust,pressure, pressure,temperature temperatureand andhumidity) humidity) was sent a ground station XBee GHz radio for monitoring, and recorded in a micro-SD card used by the Raspberry Pi. The dust 2.4 GHz radio for monitoring, and recorded in a micro-SD card used by the Raspberry Pi. The monitoring monitoring module module was was programmed programmed to to have a response time of 0.2 s. Time synchronisation was mainly achieved achievedusing usingthe the GPS time stamp of each reading. The name each log file also GPS time stamp of each reading. The name of eachoflog file was also was recorded recorded with the time and date synchronised with the internet using a portable WiFi hot spot. with the time and date synchronised with the internet using a portable WiFi hot spot.
Architecture of dust monitoring module. Figure 1. Architecture
The totalweight weightof of monitoring module 382.0 andfollowing had thedimensions: following The total thethe dustdust monitoring module was ofwas 382.0of g and hadg the dimensions: × 9.0cm. cmThe × 10.0 cm.multi-rotor The small multi-rotor UAV for the investigation was an 14.0 cm × 9.014.0 cm cm × 10.0 small UAV used for theused investigation was an IRIS+ from IRIS+ from 3DR (Berkeley, CA, USA), with an endurance of approximately 10 min. The IRIS+ 3DR (Berkeley, CA, USA), with an endurance of approximately 10 min. The IRIS+ multi-rotor UAV multi-rotor UAVflying wasplatform the selected platform due to robustness, user-friendly controls, was the selected due toflying robustness, user-friendly controls, versatility and ability to be versatility andserviced. ability toThe be IRIS+ modified serviced. TheanIRIS+ was adapted carry an air sampling modified and was and adapted to carry air sampling probeto and an enclosure for the probe and an enclosure dustinmonitoring dust monitoring modulefor as the shown Figure 2. module as shown in Figure 2.
Sensors 2017, 17, 343 Sensors 2017, 17, 343
3 of 25 3 of 25
Figure 2.2.Physical Physicalconfiguration configuration of multi-rotor the multi-rotor UAVthewith the sampling and dust of the UAV with sampling probe andprobe dust monitoring monitoring module attached. module attached.
2.2. Aerodynamics Experiment and and Analysis Analysis 2.2. Aerodynamics Downwash/Upwash Downwash/Upwash Experiment When using using aa multi-rotor multi-rotorUAV UAVtwo twomain mainissues issues are identified when obtaining a representative When are identified when obtaining a representative air air sample from the atmosphere: difficulty of producing isokinetic flow, and designing an air sample from the atmosphere: difficulty of producing isokinetic flow, and designing an air sampling sampling and determining the bestinposition in the multi-rotor UAV. probe andprobe determining the best position the multi-rotor UAV. To understand understand the the disturbance disturbance produced produced by propellers in the volume volume of air that that will will be be To by the the propellers in the of air sampled, an aerodynamic experiment was designed. Through this experiment, it was possible to sampled, an aerodynamic experiment was designed. Through this experiment, it was possible to observe the thebehavior behaviorof of volume air surrounding the multi-rotor UAV and determine if observe thethe volume of airofsurrounding the multi-rotor UAV and determine if isokinetic isokinetic flow would be possible. Isokinetic flow is an important consideration when monitoring or flow would be possible. Isokinetic flow is an important consideration when monitoring or sampling for samplingbelow for particles 10 µm in diameter, otherwise the resulting value could be under[12]. or particles 10 µm inbelow diameter, otherwise the resulting value could be under or over-predicted over-predicted However, being able toinobtain isokinetic flow can in non-static conditions can be However, being [12]. able to obtain isokinetic flow non-static conditions be extremely difficult [13,14]. extremely difficult [13,14]. Air flow intake needs to be controlled so the air streamlines are unaltered Air flow intake needs to be controlled so the air streamlines are unaltered during the transportation during the transportation of particles the atmosphere the probe’s nozzle. Toseveral achievefactors these of particles from the atmosphere past from the probe’s nozzle. Topast achieve these conditions conditions several factors have to be velocity has to be insidenozzle, and outside the have to be considered: air velocity hasconsidered: to be equalair inside and outside theequal sampling air intake sampling nozzle, air intake has to have the same direction (iso-axial), and has to be placed out of the has to have the same direction (iso-axial), and has to be placed out of the air mixing zone [14–16]. air mixing zone to [14–16]. Other variables to consider for isokinetic measurements with aircrafts are: Other variables consider for isokinetic measurements with aircrafts are: diffusion, sedimentation, diffusion, sedimentation, and turbulent inertial deposition. However, the variables considered will and turbulent inertial deposition. However, the variables considered will vary depending on the type vary depending on the type of aircraft [15–17]. Related literature on the topic mainly focuses on fixed of aircraft [15–17]. Related literature on the topic mainly focuses on fixed wing UAVs which, due wing UAVs due to their characteristics, can provide bettersampling conditions for isokinetic to their flightwhich, characteristics, can flight provide better conditions for isokinetic [16,17]. Factors sampling [16,17]. Factors such as angled flight, static flight or hovering, and constant change in wind such as angled flight, static flight or hovering, and constant change in wind direction, make achieving direction, make achieving isokinetic conditions very challenging for multi-rotor UAVs. Von der isokinetic conditions very challenging for multi-rotor UAVs. Von der Weiden et al. [12], used a different Weiden et al.to[12], used a of different approach dueconditions. to the difficulty of sampling in isokinetic approach due the difficulty sampling in isokinetic They created a correction calculator conditions. They created a correction calculator for air sampling that does not meet criteria for air sampling that does not meet the criteria necessary to be isokinetic and iso-axial. Thethe calculator necessary variation to be isokinetic and iso-axial. The calculator considers in sedimentation, considers in sedimentation, deposition due to inlets, bends,variation contractions, and diffusion deposition due to inlets, bends, contractions, and diffusion factors. However, their approach was factors. However, their approach was only used and tested for the fixed wing UAV specified in only used and tested for the fixed wing UAV specified in their study. their study. For the the aerodynamics aerodynamics experiment, For experiment, the the UAV UAV was was placed placed indoors indoors and and mounted mounted on on aa 2.5 2.5 m m pole. pole. Measurements were taken with a Kestrel 2500 anemometer (Minneapolis, MN, USA) which was Measurements were taken with a Kestrel 2500 anemometer (Minneapolis, MN, USA) which was positioned around the UAV following a grid pattern at different x, y and z positions (Figure 3). With positioned around the UAV following a grid pattern at different x, y and z positions (Figure 3). With the readings readings collected, collected, aa 3D by the the the 3D visualization visualization map map was was generated generated to to observe observe the the airflow airflow produced produced by propeller downwash and upwash (Figure 4). propeller downwash and upwash (Figure 4).
Sensors 2017, 17, 343 Sensors 2017, 17, 343
4 of 25 4 of 25
Set up up of of the the multi-rotor UAV and airspeed airspeed measuring measuring pole Figure 3. Set pole for for aerodynamics aerodynamics test. test.
Figure 4. 4. 3D 3D visualization of the airspeed regions regions produced produced by by the the propellers propellers upwash upwash at at the the Figure visualization of the different different airspeed top and and sides sides of of the the multi-rotor multi-rotor UAV. UAV.Axis Axisunits unitsininmeters. meters. top
It was determined that the downwash produced by the multi-rotor UAV does not generate a It was determined that the downwash produced by the multi-rotor UAV does not generate a uniform pattern by observing the readings and visualisations. The downwash varies by forming a uniform pattern by observing the readings and visualisations. The downwash varies by forming cone which increases its base as distance from the center of the UAV increases (Figure 4). The upper a cone which increases its base as distance from the center of the UAV increases (Figure 4). The side of the UAV on the other hand has a constant air speed flux which drops after a distance of upper side of the UAV on the other hand has a constant air speed flux which drops after a distance approximately 40.0–45.0 cm. This constant air flux is also observed to the sides of the UAV up to a of approximately 40.0–45.0 cm. This constant air flux is also observed to the sides of the UAV up to distance of 40.0–45.0 cm. The constant mixed air boundaries around the multi-rotor UAV helped a distance of 40.0–45.0 cm. The constant mixed air boundaries around the multi-rotor UAV helped define the position of the probe on the side of the UAV (horizontally) or on the top of it (vertically). define the position of the probe on the side of the UAV (horizontally) or on the top of it (vertically). A horizontal probe would increase the capacity of the UAV to collect an isokinetic sample when the A horizontal probe would increase the capacity of the UAV to collect an isokinetic sample when the wind direction is in line with the direction of the probe. However, when conducting filed tests, wind wind direction is in line with the direction of the probe. However, when conducting filed tests, wind direction varied up to 68° during the testing period (10 min approximately) for three out of four direction varied up to 68◦ during the testing period (10 min approximately) for three out of four flights flights conducted. Such ample variation in the intake angle is a considerable change of the ideal conducted. Such ample variation in the intake angle is a considerable change of the ideal conditions conditions required for isokinetic sampling. To overcome this issue, it was decided to use the required for isokinetic sampling. To overcome this issue, it was decided to use the vertical probe on vertical probe on top of the UAV at a distance of 47.5 cm from the center of the UAV (Figure 2). The use of a canopy on the top also increases its capacity to retain small volumes of the air targeted. This
Sensors 2017, 17, 343
5 of 25
Sensors 2017, 17, 343
5 of 25
top of 2017, the UAV Sensors 17, 343at a distance of 47.5 cm from the center of the UAV (Figure 2). The use of a canopy 5 of 25 on the top also increases its capacity to retain small volumes of the air targeted. This can be useful in can be useful in high wind speeds. A vertical probe has the same exposure when taking air samples can be useful in high wind speeds. A vertical probe has the same exposure when taking air samples high wind of speeds. A verticalof probe has the same exposure taking air samples regardless of the regardless the orientation the multi-rotor UAV or the when wind flow. regardless of the or multi-rotor orientation ofthe theorientation multi-rotorof UAV the wind UAV flow. or the wind flow.
2.3. Calcined Alumina Correction Factor 2.3. Calcined Alumina Correction Factor A correction factor was calculated for the dust source so the data collected by the sensing A correction calculated for for the the dust source soofthe collected by sensing module correction factorwas was calculated source sodata the data for collected bywas the calcined sensing module could befactor used for further estimates. Thedust source dust chosen thethe test could be used for further estimates. The source of dust chosen for the test was calcined alumina, which module could be used for further estimates. The source of dust chosen for the test was calcined alumina, which is less hygroscopic than talcum powder, used in previous tests [9]. Being less is less hygroscopic than talcum powder, used in previous tests [9]. Being less hygroscopic helped keep alumina, which is less hygroscopic than talcum powder, used in previous tests [9]. Being less hygroscopic helped keep the optical sensors free of dust adhering to the components and the optical sensors free of dust adhering to the components and accumulating inside the protective hygroscopic helped keep the optical sensors free of dust adhering to the components and accumulating inside the protective casing. In addition, according to the particle size distribution of casing. In addition, to the particle calcined alumina used for the accumulating inside according the protective casing. In size addition, to the particle size distribution of the calcined alumina used for the experiment, thedistribution dustaccording had a of PMthe 10 content of approximately 30%. experiment, the dust had a PM content of approximately 30%. This value is ten times higher than the calcined alumina used for the experiment, the dust had a PM 10 content of approximately 30%. 10 This value is ten times higher than the average PM10 content found in talcum powder. A higher PM10 the content found in talcum powder. Acontent higher PM10The content allowed testing theofsensor Thisaverage value isPM ten10times higher than the average PM 10 found in talcum powder. A higher PM10 10 content allowed testing the sensor with small amounts of dust. highest concentration PM with small amounts of dust. The highest concentration of PM measured during the experiment was content allowed the sensorwas withofsmall of dust. The3 highest PM10 10 mg/m measured duringtesting the experiment 17.32amounts mg/m3 (19.42 of TSP).concentration An OPC-N2 of optical 3 (19.42 mg/m3 of TSP). An OPC-N2 optical 3 (19.42 3 of TSP). of 17.32 mg/m particle counter and a TSI DRX 8533 measured during the experiment was of 17.32 mg/m mg/m An OPC-N2 optical particle counter and a TSI DRX 8533 [18] were collocated with a gravimetric sampler, [18] an were collocated with ancalculate AirChek2000 Four,a factor PA, USA), to calculate particle counter anda gravimetric a TSI PA, DRXsampler, 8533 to [18] were collocated with gravimetric sampler, an AirChek2000 (Eighty Four, USA), their(Eighty correction (Figures 5 andtheir 6). correction factor (Figures 5OPC-N2 andPA, 6). Concentrations from the OPC-N2 and the TSI DRX AirChek2000 (Eighty Four, USA), calculate their correction factor (Figures 5 obtained andfiles 6). Concentrations from the and thetoTSI DRX were obtained directly from thewere logging directly from the logging files produced by the interface software provided by the manufacturers. Tests Concentrations from the OPC-N2 and the TSI DRX were obtained directly from the logging files produced by the interface software provided by the manufacturers. Tests were done in a dust were doneofby inapproximately athe dustinterface chamber of approximately m3 .manufacturers. Tests were done in a dust produced software by the chamber 0.104 m3. provided0.104 chamber of approximately 0.104 m3.
Figure 5. Set up of monitoring devices and dust chamber for calcined alumina calibration tests. Figure 5. Set up of monitoring devices and dust chamber for calcined alumina calibration tests. Figure 5. Set up of monitoring devices and dust chamber for calcined alumina calibration tests.
Figure 6. Layout of all monitoring equipment and sensors used to obtain the particle correction Figure 6. Layout of all monitoring equipment and sensors used to obtain the particle correction factor. Figure 6. Layout of all monitoring equipment and sensors used to obtain the particle correction factor. factor.
The tests consisted of inducing dust into the chamber with the aid of an axial fan. Three tests tests consisted inducing dustofinto the3 aid axial fan. Three wereThe conducted using a of regular supply dustthe at chamber intervals with lasting minofforanthe first, 6 min fortests the were conducted using a regular supply of dust at intervals lasting 3 min for the first, 6 min for the second and 9 min for the third test. In addition to the tests where dust was supplied, two blank tests second and 9 min for the third test. In addition to the tests where dust was supplied, two blank tests
Sensors 2017, 17, 343
6 of 25
The tests consisted of inducing dust into the chamber with the aid of an axial fan. Three tests were conducted using a regular supply of dust at intervals lasting 3 min for the first, 6 min for the Sensors 2017, 17, 343 6 of 25 second and 9 min for the third test. In addition to the tests where dust was supplied, two blank tests 3 were were conducted conducted to to know know background backgrounddust dustlevels, levels,resulting resultingin inan anaverage averageconcentration concentrationofof4.82 4.82µg/m µg/m3 of data waswas collected, lineallineal regression analysis was used to calculate correction of PM PM1010.. Once Onceallallthethe data collected, regression analysis was used to the calculate the factors between correction factorsdevices. between devices. The with the the AirChek2000 were used used as The filters The samples samples collected collected with AirChek2000 were as the the main main reference. reference. The filters used used by the sampler were sent to a certified laboratory (ALS Environmental, New Castle, Australia) by the sampler were sent to a certified laboratory (ALS Environmental, New Castle, Australia) for for analysis of calcined alumina PM10 concentrations. The correction factors calculated analysis and anddetermination determination of calcined alumina PM10 concentrations. The correction factors are shown in 1. Figure 7 shows the resulting predicted values for the values AirChek2000 use of calculated areTable shown in Table 1. Figure 7 shows the resulting predicted for themaking AirChek2000 the particle factor calculated withcalculated the correlation the TSIbetween DRX and the airDRX sampler. making use correction of the particle correction factor with between the correlation the TSI and 2 ) of 0.92 for the correlation indicates a good fit between the optical A coefficient of determination (R the air sampler. A coefficient of determination (R2) of 0.92 for the correlation indicates a good fit sensor and air sampler. between thethe optical sensor and the air sampler. Table 1. Correction factors calculated with the analysis of the data from the dust chamber. Table 1. Correction factors calculated with the analysis of the data from the dust chamber.
Device Device
TSI DRX TSI DRX Correction Factor R2 Correction Factor
OPC N2 OPC N2 (airborne module)0.342 (airborne module) TSI DRX TSI DRX NA
0.342 NA
0.62 NA
R2 0.62 NA
AirChek2000 AirChek2000 Correction Factor Correction Factor 1.362 1.362 1.312 1.312
2 RR2
0.73 0.73 0.92 0.92
NA = Not Applicable. NA = Not Applicable.
Figure 7. Correlation obtained obtained to to calculate calculate the the correction correction factor factor for for the the TSI TSI DRX against the 7. Correlation gravimetric sampler. sampler.
In Figure Figure 8, 8, the the process process to to obtain obtain the the correction correction factors factors is is presented presented as as aa flow flow diagram. diagram. This This In diagram also summarises the other two experiments conducted to validate the readings of the dust diagram also summarises the other two experiments conducted to validate the readings of the dust monitoring sensors. sensors. These are described described in in the the following following sections. sections. monitoring These experiments experiments are 2.4. Variable Wind Speed Modeling
additiontotothe the previous a variable air speed experiment was designed the In addition previous testtest a variable air speed experiment was designed to studyto thestudy response response of the OPC-N2 with the probe attached using different wind speeds. For this experiment of the OPC-N2 with the probe attached using different wind speeds. For this experiment a wind tunnela windbuilt tunnel was built to produce laminar and control the air(Figure speed changes (Figure was to produce a laminar flowaand have flow control ofhave the air speedof changes 9). The test area 9).the Thewind test tunnel area ofwas the40.0 wind was 40.0 cross in whichtothe (attached to the of cmtunnel cross section, in cm which thesection, probe (attached theprobe OPC-N2), a secondary OPC-N2),without a secondary OPC-N2 probe, and TSIconductive DRX (making use were of theplaced conductive OPC-N2 a probe, and thewithout TSI DRXa (making usethe of the tubing), at the tubing), were(15.0 placed thebase sameofheight (15.0for cmuniformity. from base of test area) for uniformity. same height cm at from test area) Four individual experiments were conducted to generate a variable wind speed model. Each experiment used a different wind speed: 1.11 m/s, 1.67 m/s, 3.06 m/s and 3.89 m/s. The air flow was produced with fans and measured with a SkyWatch ATMOS anemometer (Sudbury, UK). Dust particles were supplied using calcined alumina at an approximate rate of 0.015 g every 10 min. The concentration read by the OPC-N2 was logged and reported by the software interface provided by Alphasense Ltd. (Great Notley, Essex, UK).
Sensors2017, 2017,17, 17,343 343 Sensors
25 77ofof25
Figure8.8.Flow Flowdiagram diagramsummarizing summarizingtests testsundertaken undertakentotovalidate validatereadings readingscollected collectedwith withthe thedust dustmonitoring monitoringsensors. sensors. Figure
Sensors 2017, 17, 343 Sensors 2017, 17, 343
8 of 25
8 of 25
(a)
(b)
Figure 9. Wind tunnel and equipment used for the variable wind speed tests, (a) overall view of the Figure 9. Wind tunnel and equipment used for the variable wind speed tests, (a) overall view of the wind tunnel and equipment used, and (b) test area set with the anemometer, probe, OPC-N2 and TSI wind tunnel and equipment used; and (b) test area set with the anemometer, probe, OPC-N2 and DRX. TSI DRX.
Two equations were generated out of this experiment, one for the OPC-N2 connected to the Four individual experiments were conducted generate with a variable Each probe and one equation for the OPC-N2 which wastocollocated the TSIwind DRXspeed duringmodel. the flight experiment used a different wind speed: 1.11 m/s, 1.67 m/s, 3.06 m/s and 3.89 m/s. The air flow was tests. The resulting coefficients for the airborne module system equation and for the collocated produced fans and measured(1) with a SkyWatch ATMOS anemometer (Sudbury, UK). Dust particles modulewith are shown in Equations and (2) and Table 2: were supplied using calcined alumina at an approximate rate of 0.015 g every 10 min. The concentration CUAV = (2.118 × CR) – (1.175 × R.H.) – (1.261 × T) + (1.822 × U) + 102.082 (1) read by the OPC-N2 was logged and reported by the software interface provided by Alphasense Ltd. CColl = (0.197 × CR) – (0.623 × R.H.) – (0.641 × T) + (1.125 × U) + 57.292 (2) (Great Notley, Essex, UK). Two equations were generated out of this experiment, one for the OPC-N2 connected to the probe and one equation for the OPC-N2for which wasofcollocated with themonitoring TSI DRX sensors duringwith the their flight tests. Table 2. Resulting coefficients variables equations for dust P-value coefficients and confidence The resulting forlevel. the airborne module system equation and for the collocated module are shown in Equations (1) and (2) and Table 2: Sensor
Variable
Coefficients
p-Value
2.5 × 10−223
× R.H.) − (1.261 × T) +102.082 (1.822 × U) + 102.082 CUAV = (2.118 × CR ) − (1.175Intercept Raw Conc. PM10 (CR, µg/m3)
2.118
10−235 CColl = (0.197 × CR ) − (0.623Relative × R.H.) − (0.641 (1.125 × U)4.2+×57.292 Humidity (R.H.×%)T) +−1.175 Airborne module (CUAV)
Temperature (T, °C)
−1.261
Air speed (U, m/s)
1.822
Intercept
57.292
(1)
~0.0
(2)
2.1 × 10−231 2.8 × 10−159
Table 2. Resulting coefficients for variables of equations for dust monitoring sensors with their p-value NA 0.55 R2 and confidence level. Sensor
Variable
Raw Conc. PM10 (CR, µg/m3Coefficients ) 0.197
Intercept Relative Humidity (R.H. %)102.082 −0.623 Collocated module sensor (CColl) PM10 (CR , µg/m3 ) 2.118 Raw Conc. Temperature (T, °C) −0.641 Relative Humidity (R.H. %) −1.175
Airborne module (CUAV )
Temperature Air (T, ◦speed C) (U, m/s) Air speed (U, m/s) 2
−1.261 1.125 1.822
R2
0.55
R
NA = Not Applicable.
0.66
2.7 × 10−84 ~0.0p-Value −79 5.6 2.5 × 10× 10−223
1.7 ~0.0 × 10−72
4.2 × 10−235
−74 2.0 2.1 × 10× 10−231
NA2.8 × 10
−159
NA
Intercept 57.292 2.7 × 10−84 The R2 for both equations using a multivariate linear regression was above~0.0 0.5. The airborne Raw Conc. PM10 (CR , µg/m3 ) 0.197 moduleCollocated had a coefficient of 0.55 and the collocated module sensor of 0.66. Coefficients module Relative Humidity (R.H. %) −0.623 5.6 ×with 10−79a P-value (CColl ) is regarded as statistically
A Methodology to Monitor Airborne PM10 Dust Particles Using a Small Unmanned Aerial Vehicle Miguel Alvarado 1, *, Felipe Gonzalez 2 , Peter Erskine 1 , David Cliff 3 and Darlene Heuff 4 1 2 3 4
*
Environment Centre, Sustainable Mineral Institute, The University of Queensland, 4072 Brisbane, Australia; [email protected] Science and Engineering Faculty, Queensland University of Technology (QUT), 4000 Brisbane, Australia; [email protected] People Centre, Sustainable Mineral Institute, The University of Queensland, 4072 Brisbane, Australia; [email protected] Advanced Environmental Dynamics Pty Ltd., Ferny Hills, 4055 Queensland, Australia; [email protected] Correspondence: [email protected]; Tel.: +61-7-3346-4065 or +61-7-3346-4086
Academic Editor: Assefa M. Melesse Received: 15 December 2016; Accepted: 4 February 2017; Published: 14 February 2017
Abstract: Throughout the process of coal extraction from surface mines, gases and particles are emitted in the form of fugitive emissions by activities such as hauling, blasting and transportation. As these emissions are diffuse in nature, estimations based upon emission factors and dispersion/advection equations need to be measured directly from the atmosphere. This paper expands upon previous research undertaken to develop a relative methodology to monitor PM10 dust particles produced by mining activities making use of small unmanned aerial vehicles (UAVs). A module sensor using a laser particle counter (OPC-N2 from Alphasense, Great Notley, Essex, UK) was tested. An aerodynamic flow experiment was undertaken to determine the position and length of a sampling probe of the sensing module. Flight tests were conducted in order to demonstrate that the sensor provided data which could be used to calculate the emission rate of a source. Emission rates are a critical variable for further predictive dispersion estimates. First, data collected by the airborne module was verified using a 5.0 m tower in which a TSI DRX 8533 (reference dust monitoring device, TSI, Shoreview, MN, USA) and a duplicate of the module sensor were installed. Second, concentration values collected by the monitoring module attached to the UAV (airborne module) obtaining a percentage error of 1.1%. Finally, emission rates from the source were calculated, with airborne data, obtaining errors as low as 1.2%. These errors are low and indicate that the readings collected with the airborne module are comparable to the TSI DRX and could be used to obtain specific emission factors from fugitive emissions for industrial activities. Keywords: PM10 ; monitoring; blasting; unmanned aerial vehicle (UAV); multi-rotor UAV; optical sensor
1. Introduction The use of Unmanned Aerial Vehicles (UAVs) to collect environmental data has increased exponentially in only a few years. Their capacity to reach places where humans could be at risk or it could become too expensive for small operations, makes them very useful for gas and particulate monitoring in the mining industry. Several authors have published articles where progress on the use of UAVs for industrial use or investigation has been made by using two UAVs simultaneously [1], combining multiple gas sensors and imagery [2,3], using gas tracking algorithms [4–6], using solar energy [7], and making use of microwave sensors [8].
Sensors 2017, 17, 343; doi:10.3390/s17020343
www.mdpi.com/journal/sensors
Sensors Sensors 2017, 2017, 17, 17, 343 343
22 of of 25 25
The methodology applied and tests undertaken for this investigation are a continuation of the methodology applied and undertaken for this investigation are a continuation workThe published by Alvarado et tests al. [9,10]. Improvements to the methodology include:of the work published by Alvarado et al. [9,10]. Improvements to the methodology include: the development development of a new modular sensor, focusing only on the use of a multi-rotor UAV with a tailored of a new modular sensor, on thetouse of a multi-rotor UAV withobtained a tailoredbysampling probe, sampling probe, and the focusing use of a 5only m tower validate the PM10 readings the UAV. The and the use of a 5 m tower to validate the PM readings obtained by the UAV. The emission rates emission rates obtained from a point source created for the flight tests, were evaluated to determine 10 obtained fromof a point source created forused the flight tests, to determine the factors. capability the capability the methodology to be at open pitwere mineevaluated sites to estimate emission In of the methodology be used at open pit sites to estimate emission factors. In the following the following sectionstothese experiments aremine described and their results discussed. sections these experiments are described and their results discussed. 2. Experimental Development 2. Experimental Development 2.1. Dust Monitoring Module 2.1. Dust Monitoring Module Following Alvarado et al. [9,10] a sampling module and data logger transmission system were Following Alvarado et al. [9,10] a sampling module and data logger transmission system were developed using the OPC-N2 particle counter from Alphasense (Great Notley, Essex, UK) [11] with a developed using the OPC-N2 particle counter from Alphasense (Great Notley, Essex, UK) [11] with a minimum response time of 0.2 s. In addition to the particle counter the dust monitoring module is minimum response time of 0.2 s. In addition to the particle counter the dust monitoring module is integrated with a: integrated with a: • • • • • •
Barometer, temperature and and humidity humidity sensor, sensor,BME280 BME280(Bosch, (Bosch,Reutlingen, Reutlingen,Germany); Germany); Barometer, temperature Raspberry board (Raspberry (Raspberry Pi Pi Foundation, Foundation, Cambridge, Cambridge, UK); UK); Raspberry Pi Pi 33 board uBlox LEA6 GPS (uBlox, Thalwil, Switzerland); uBlox LEA6 GPS (uBlox, Thalwil, Switzerland); XBee (XBee, Minnetonka, Minnetonka, MN, MN, USA); USA); XBee 2.4 2.4 GHz GHz Radio Radio (XBee, 3DR Airspeed Sensor (3DR, Berkeley, CA, USA); 3DR Airspeed Sensor (3DR, Berkeley, CA, USA); Lipo battery (3.7 V) (3DR, Berkeley, CA, USA). Lipo battery (3.7 V) (3DR, Berkeley, CA, USA).
The architecture of the dust monitoring module is presented in Figure 1. Information from each sensor (e.g., was sent to to a ground station viavia the the XBee 2.4 (e.g., dust, dust,pressure, pressure,temperature temperatureand andhumidity) humidity) was sent a ground station XBee GHz radio for monitoring, and recorded in a micro-SD card used by the Raspberry Pi. The dust 2.4 GHz radio for monitoring, and recorded in a micro-SD card used by the Raspberry Pi. The monitoring monitoring module module was was programmed programmed to to have a response time of 0.2 s. Time synchronisation was mainly achieved achievedusing usingthe the GPS time stamp of each reading. The name each log file also GPS time stamp of each reading. The name of eachoflog file was also was recorded recorded with the time and date synchronised with the internet using a portable WiFi hot spot. with the time and date synchronised with the internet using a portable WiFi hot spot.
Architecture of dust monitoring module. Figure 1. Architecture
The totalweight weightof of monitoring module 382.0 andfollowing had thedimensions: following The total thethe dustdust monitoring module was ofwas 382.0of g and hadg the dimensions: × 9.0cm. cmThe × 10.0 cm.multi-rotor The small multi-rotor UAV for the investigation was an 14.0 cm × 9.014.0 cm cm × 10.0 small UAV used for theused investigation was an IRIS+ from IRIS+ from 3DR (Berkeley, CA, USA), with an endurance of approximately 10 min. The IRIS+ 3DR (Berkeley, CA, USA), with an endurance of approximately 10 min. The IRIS+ multi-rotor UAV multi-rotor UAVflying wasplatform the selected platform due to robustness, user-friendly controls, was the selected due toflying robustness, user-friendly controls, versatility and ability to be versatility andserviced. ability toThe be IRIS+ modified serviced. TheanIRIS+ was adapted carry an air sampling modified and was and adapted to carry air sampling probeto and an enclosure for the probe and an enclosure dustinmonitoring dust monitoring modulefor as the shown Figure 2. module as shown in Figure 2.
Sensors 2017, 17, 343 Sensors 2017, 17, 343
3 of 25 3 of 25
Figure 2.2.Physical Physicalconfiguration configuration of multi-rotor the multi-rotor UAVthewith the sampling and dust of the UAV with sampling probe andprobe dust monitoring monitoring module attached. module attached.
2.2. Aerodynamics Experiment and and Analysis Analysis 2.2. Aerodynamics Downwash/Upwash Downwash/Upwash Experiment When using using aa multi-rotor multi-rotorUAV UAVtwo twomain mainissues issues are identified when obtaining a representative When are identified when obtaining a representative air air sample from the atmosphere: difficulty of producing isokinetic flow, and designing an air sample from the atmosphere: difficulty of producing isokinetic flow, and designing an air sampling sampling and determining the bestinposition in the multi-rotor UAV. probe andprobe determining the best position the multi-rotor UAV. To understand understand the the disturbance disturbance produced produced by propellers in the volume volume of air that that will will be be To by the the propellers in the of air sampled, an aerodynamic experiment was designed. Through this experiment, it was possible to sampled, an aerodynamic experiment was designed. Through this experiment, it was possible to observe the thebehavior behaviorof of volume air surrounding the multi-rotor UAV and determine if observe thethe volume of airofsurrounding the multi-rotor UAV and determine if isokinetic isokinetic flow would be possible. Isokinetic flow is an important consideration when monitoring or flow would be possible. Isokinetic flow is an important consideration when monitoring or sampling for samplingbelow for particles 10 µm in diameter, otherwise the resulting value could be under[12]. or particles 10 µm inbelow diameter, otherwise the resulting value could be under or over-predicted over-predicted However, being able toinobtain isokinetic flow can in non-static conditions can be However, being [12]. able to obtain isokinetic flow non-static conditions be extremely difficult [13,14]. extremely difficult [13,14]. Air flow intake needs to be controlled so the air streamlines are unaltered Air flow intake needs to be controlled so the air streamlines are unaltered during the transportation during the transportation of particles the atmosphere the probe’s nozzle. Toseveral achievefactors these of particles from the atmosphere past from the probe’s nozzle. Topast achieve these conditions conditions several factors have to be velocity has to be insidenozzle, and outside the have to be considered: air velocity hasconsidered: to be equalair inside and outside theequal sampling air intake sampling nozzle, air intake has to have the same direction (iso-axial), and has to be placed out of the has to have the same direction (iso-axial), and has to be placed out of the air mixing zone [14–16]. air mixing zone to [14–16]. Other variables to consider for isokinetic measurements with aircrafts are: Other variables consider for isokinetic measurements with aircrafts are: diffusion, sedimentation, diffusion, sedimentation, and turbulent inertial deposition. However, the variables considered will and turbulent inertial deposition. However, the variables considered will vary depending on the type vary depending on the type of aircraft [15–17]. Related literature on the topic mainly focuses on fixed of aircraft [15–17]. Related literature on the topic mainly focuses on fixed wing UAVs which, due wing UAVs due to their characteristics, can provide bettersampling conditions for isokinetic to their flightwhich, characteristics, can flight provide better conditions for isokinetic [16,17]. Factors sampling [16,17]. Factors such as angled flight, static flight or hovering, and constant change in wind such as angled flight, static flight or hovering, and constant change in wind direction, make achieving direction, make achieving isokinetic conditions very challenging for multi-rotor UAVs. Von der isokinetic conditions very challenging for multi-rotor UAVs. Von der Weiden et al. [12], used a different Weiden et al.to[12], used a of different approach dueconditions. to the difficulty of sampling in isokinetic approach due the difficulty sampling in isokinetic They created a correction calculator conditions. They created a correction calculator for air sampling that does not meet criteria for air sampling that does not meet the criteria necessary to be isokinetic and iso-axial. Thethe calculator necessary variation to be isokinetic and iso-axial. The calculator considers in sedimentation, considers in sedimentation, deposition due to inlets, bends,variation contractions, and diffusion deposition due to inlets, bends, contractions, and diffusion factors. However, their approach was factors. However, their approach was only used and tested for the fixed wing UAV specified in only used and tested for the fixed wing UAV specified in their study. their study. For the the aerodynamics aerodynamics experiment, For experiment, the the UAV UAV was was placed placed indoors indoors and and mounted mounted on on aa 2.5 2.5 m m pole. pole. Measurements were taken with a Kestrel 2500 anemometer (Minneapolis, MN, USA) which was Measurements were taken with a Kestrel 2500 anemometer (Minneapolis, MN, USA) which was positioned around the UAV following a grid pattern at different x, y and z positions (Figure 3). With positioned around the UAV following a grid pattern at different x, y and z positions (Figure 3). With the readings readings collected, collected, aa 3D by the the the 3D visualization visualization map map was was generated generated to to observe observe the the airflow airflow produced produced by propeller downwash and upwash (Figure 4). propeller downwash and upwash (Figure 4).
Sensors 2017, 17, 343 Sensors 2017, 17, 343
4 of 25 4 of 25
Set up up of of the the multi-rotor UAV and airspeed airspeed measuring measuring pole Figure 3. Set pole for for aerodynamics aerodynamics test. test.
Figure 4. 4. 3D 3D visualization of the airspeed regions regions produced produced by by the the propellers propellers upwash upwash at at the the Figure visualization of the different different airspeed top and and sides sides of of the the multi-rotor multi-rotor UAV. UAV.Axis Axisunits unitsininmeters. meters. top
It was determined that the downwash produced by the multi-rotor UAV does not generate a It was determined that the downwash produced by the multi-rotor UAV does not generate a uniform pattern by observing the readings and visualisations. The downwash varies by forming a uniform pattern by observing the readings and visualisations. The downwash varies by forming cone which increases its base as distance from the center of the UAV increases (Figure 4). The upper a cone which increases its base as distance from the center of the UAV increases (Figure 4). The side of the UAV on the other hand has a constant air speed flux which drops after a distance of upper side of the UAV on the other hand has a constant air speed flux which drops after a distance approximately 40.0–45.0 cm. This constant air flux is also observed to the sides of the UAV up to a of approximately 40.0–45.0 cm. This constant air flux is also observed to the sides of the UAV up to distance of 40.0–45.0 cm. The constant mixed air boundaries around the multi-rotor UAV helped a distance of 40.0–45.0 cm. The constant mixed air boundaries around the multi-rotor UAV helped define the position of the probe on the side of the UAV (horizontally) or on the top of it (vertically). define the position of the probe on the side of the UAV (horizontally) or on the top of it (vertically). A horizontal probe would increase the capacity of the UAV to collect an isokinetic sample when the A horizontal probe would increase the capacity of the UAV to collect an isokinetic sample when the wind direction is in line with the direction of the probe. However, when conducting filed tests, wind wind direction is in line with the direction of the probe. However, when conducting filed tests, wind direction varied up to 68° during the testing period (10 min approximately) for three out of four direction varied up to 68◦ during the testing period (10 min approximately) for three out of four flights flights conducted. Such ample variation in the intake angle is a considerable change of the ideal conducted. Such ample variation in the intake angle is a considerable change of the ideal conditions conditions required for isokinetic sampling. To overcome this issue, it was decided to use the required for isokinetic sampling. To overcome this issue, it was decided to use the vertical probe on vertical probe on top of the UAV at a distance of 47.5 cm from the center of the UAV (Figure 2). The use of a canopy on the top also increases its capacity to retain small volumes of the air targeted. This
Sensors 2017, 17, 343
5 of 25
Sensors 2017, 17, 343
5 of 25
top of 2017, the UAV Sensors 17, 343at a distance of 47.5 cm from the center of the UAV (Figure 2). The use of a canopy 5 of 25 on the top also increases its capacity to retain small volumes of the air targeted. This can be useful in can be useful in high wind speeds. A vertical probe has the same exposure when taking air samples can be useful in high wind speeds. A vertical probe has the same exposure when taking air samples high wind of speeds. A verticalof probe has the same exposure taking air samples regardless of the regardless the orientation the multi-rotor UAV or the when wind flow. regardless of the or multi-rotor orientation ofthe theorientation multi-rotorof UAV the wind UAV flow. or the wind flow.
2.3. Calcined Alumina Correction Factor 2.3. Calcined Alumina Correction Factor A correction factor was calculated for the dust source so the data collected by the sensing A correction calculated for for the the dust source soofthe collected by sensing module correction factorwas was calculated source sodata the data for collected bywas the calcined sensing module could befactor used for further estimates. Thedust source dust chosen thethe test could be used for further estimates. The source of dust chosen for the test was calcined alumina, which module could be used for further estimates. The source of dust chosen for the test was calcined alumina, which is less hygroscopic than talcum powder, used in previous tests [9]. Being less is less hygroscopic than talcum powder, used in previous tests [9]. Being less hygroscopic helped keep alumina, which is less hygroscopic than talcum powder, used in previous tests [9]. Being less hygroscopic helped keep the optical sensors free of dust adhering to the components and the optical sensors free of dust adhering to the components and accumulating inside the protective hygroscopic helped keep the optical sensors free of dust adhering to the components and accumulating inside the protective casing. In addition, according to the particle size distribution of casing. In addition, to the particle calcined alumina used for the accumulating inside according the protective casing. In size addition, to the particle size distribution of the calcined alumina used for the experiment, thedistribution dustaccording had a of PMthe 10 content of approximately 30%. experiment, the dust had a PM content of approximately 30%. This value is ten times higher than the calcined alumina used for the experiment, the dust had a PM 10 content of approximately 30%. 10 This value is ten times higher than the average PM10 content found in talcum powder. A higher PM10 the content found in talcum powder. Acontent higher PM10The content allowed testing theofsensor Thisaverage value isPM ten10times higher than the average PM 10 found in talcum powder. A higher PM10 10 content allowed testing the sensor with small amounts of dust. highest concentration PM with small amounts of dust. The highest concentration of PM measured during the experiment was content allowed the sensorwas withofsmall of dust. The3 highest PM10 10 mg/m measured duringtesting the experiment 17.32amounts mg/m3 (19.42 of TSP).concentration An OPC-N2 of optical 3 (19.42 mg/m3 of TSP). An OPC-N2 optical 3 (19.42 3 of TSP). of 17.32 mg/m particle counter and a TSI DRX 8533 measured during the experiment was of 17.32 mg/m mg/m An OPC-N2 optical particle counter and a TSI DRX 8533 [18] were collocated with a gravimetric sampler, [18] an were collocated with ancalculate AirChek2000 Four,a factor PA, USA), to calculate particle counter anda gravimetric a TSI PA, DRXsampler, 8533 to [18] were collocated with gravimetric sampler, an AirChek2000 (Eighty Four, USA), their(Eighty correction (Figures 5 andtheir 6). correction factor (Figures 5OPC-N2 andPA, 6). Concentrations from the OPC-N2 and the TSI DRX AirChek2000 (Eighty Four, USA), calculate their correction factor (Figures 5 obtained andfiles 6). Concentrations from the and thetoTSI DRX were obtained directly from thewere logging directly from the logging files produced by the interface software provided by the manufacturers. Tests Concentrations from the OPC-N2 and the TSI DRX were obtained directly from the logging files produced by the interface software provided by the manufacturers. Tests were done in a dust were doneofby inapproximately athe dustinterface chamber of approximately m3 .manufacturers. Tests were done in a dust produced software by the chamber 0.104 m3. provided0.104 chamber of approximately 0.104 m3.
Figure 5. Set up of monitoring devices and dust chamber for calcined alumina calibration tests. Figure 5. Set up of monitoring devices and dust chamber for calcined alumina calibration tests. Figure 5. Set up of monitoring devices and dust chamber for calcined alumina calibration tests.
Figure 6. Layout of all monitoring equipment and sensors used to obtain the particle correction Figure 6. Layout of all monitoring equipment and sensors used to obtain the particle correction factor. Figure 6. Layout of all monitoring equipment and sensors used to obtain the particle correction factor. factor.
The tests consisted of inducing dust into the chamber with the aid of an axial fan. Three tests tests consisted inducing dustofinto the3 aid axial fan. Three wereThe conducted using a of regular supply dustthe at chamber intervals with lasting minofforanthe first, 6 min fortests the were conducted using a regular supply of dust at intervals lasting 3 min for the first, 6 min for the second and 9 min for the third test. In addition to the tests where dust was supplied, two blank tests second and 9 min for the third test. In addition to the tests where dust was supplied, two blank tests
Sensors 2017, 17, 343
6 of 25
The tests consisted of inducing dust into the chamber with the aid of an axial fan. Three tests were conducted using a regular supply of dust at intervals lasting 3 min for the first, 6 min for the Sensors 2017, 17, 343 6 of 25 second and 9 min for the third test. In addition to the tests where dust was supplied, two blank tests 3 were were conducted conducted to to know know background backgrounddust dustlevels, levels,resulting resultingin inan anaverage averageconcentration concentrationofof4.82 4.82µg/m µg/m3 of data waswas collected, lineallineal regression analysis was used to calculate correction of PM PM1010.. Once Onceallallthethe data collected, regression analysis was used to the calculate the factors between correction factorsdevices. between devices. The with the the AirChek2000 were used used as The filters The samples samples collected collected with AirChek2000 were as the the main main reference. reference. The filters used used by the sampler were sent to a certified laboratory (ALS Environmental, New Castle, Australia) by the sampler were sent to a certified laboratory (ALS Environmental, New Castle, Australia) for for analysis of calcined alumina PM10 concentrations. The correction factors calculated analysis and anddetermination determination of calcined alumina PM10 concentrations. The correction factors are shown in 1. Figure 7 shows the resulting predicted values for the values AirChek2000 use of calculated areTable shown in Table 1. Figure 7 shows the resulting predicted for themaking AirChek2000 the particle factor calculated withcalculated the correlation the TSIbetween DRX and the airDRX sampler. making use correction of the particle correction factor with between the correlation the TSI and 2 ) of 0.92 for the correlation indicates a good fit between the optical A coefficient of determination (R the air sampler. A coefficient of determination (R2) of 0.92 for the correlation indicates a good fit sensor and air sampler. between thethe optical sensor and the air sampler. Table 1. Correction factors calculated with the analysis of the data from the dust chamber. Table 1. Correction factors calculated with the analysis of the data from the dust chamber.
Device Device
TSI DRX TSI DRX Correction Factor R2 Correction Factor
OPC N2 OPC N2 (airborne module)0.342 (airborne module) TSI DRX TSI DRX NA
0.342 NA
0.62 NA
R2 0.62 NA
AirChek2000 AirChek2000 Correction Factor Correction Factor 1.362 1.362 1.312 1.312
2 RR2
0.73 0.73 0.92 0.92
NA = Not Applicable. NA = Not Applicable.
Figure 7. Correlation obtained obtained to to calculate calculate the the correction correction factor factor for for the the TSI TSI DRX against the 7. Correlation gravimetric sampler. sampler.
In Figure Figure 8, 8, the the process process to to obtain obtain the the correction correction factors factors is is presented presented as as aa flow flow diagram. diagram. This This In diagram also summarises the other two experiments conducted to validate the readings of the dust diagram also summarises the other two experiments conducted to validate the readings of the dust monitoring sensors. sensors. These are described described in in the the following following sections. sections. monitoring These experiments experiments are 2.4. Variable Wind Speed Modeling
additiontotothe the previous a variable air speed experiment was designed the In addition previous testtest a variable air speed experiment was designed to studyto thestudy response response of the OPC-N2 with the probe attached using different wind speeds. For this experiment of the OPC-N2 with the probe attached using different wind speeds. For this experiment a wind tunnela windbuilt tunnel was built to produce laminar and control the air(Figure speed changes (Figure was to produce a laminar flowaand have flow control ofhave the air speedof changes 9). The test area 9).the Thewind test tunnel area ofwas the40.0 wind was 40.0 cross in whichtothe (attached to the of cmtunnel cross section, in cm which thesection, probe (attached theprobe OPC-N2), a secondary OPC-N2),without a secondary OPC-N2 probe, and TSIconductive DRX (making use were of theplaced conductive OPC-N2 a probe, and thewithout TSI DRXa (making usethe of the tubing), at the tubing), were(15.0 placed thebase sameofheight (15.0for cmuniformity. from base of test area) for uniformity. same height cm at from test area) Four individual experiments were conducted to generate a variable wind speed model. Each experiment used a different wind speed: 1.11 m/s, 1.67 m/s, 3.06 m/s and 3.89 m/s. The air flow was produced with fans and measured with a SkyWatch ATMOS anemometer (Sudbury, UK). Dust particles were supplied using calcined alumina at an approximate rate of 0.015 g every 10 min. The concentration read by the OPC-N2 was logged and reported by the software interface provided by Alphasense Ltd. (Great Notley, Essex, UK).
Sensors2017, 2017,17, 17,343 343 Sensors
25 77ofof25
Figure8.8.Flow Flowdiagram diagramsummarizing summarizingtests testsundertaken undertakentotovalidate validatereadings readingscollected collectedwith withthe thedust dustmonitoring monitoringsensors. sensors. Figure
Sensors 2017, 17, 343 Sensors 2017, 17, 343
8 of 25
8 of 25
(a)
(b)
Figure 9. Wind tunnel and equipment used for the variable wind speed tests, (a) overall view of the Figure 9. Wind tunnel and equipment used for the variable wind speed tests, (a) overall view of the wind tunnel and equipment used, and (b) test area set with the anemometer, probe, OPC-N2 and TSI wind tunnel and equipment used; and (b) test area set with the anemometer, probe, OPC-N2 and DRX. TSI DRX.
Two equations were generated out of this experiment, one for the OPC-N2 connected to the Four individual experiments were conducted generate with a variable Each probe and one equation for the OPC-N2 which wastocollocated the TSIwind DRXspeed duringmodel. the flight experiment used a different wind speed: 1.11 m/s, 1.67 m/s, 3.06 m/s and 3.89 m/s. The air flow was tests. The resulting coefficients for the airborne module system equation and for the collocated produced fans and measured(1) with a SkyWatch ATMOS anemometer (Sudbury, UK). Dust particles modulewith are shown in Equations and (2) and Table 2: were supplied using calcined alumina at an approximate rate of 0.015 g every 10 min. The concentration CUAV = (2.118 × CR) – (1.175 × R.H.) – (1.261 × T) + (1.822 × U) + 102.082 (1) read by the OPC-N2 was logged and reported by the software interface provided by Alphasense Ltd. CColl = (0.197 × CR) – (0.623 × R.H.) – (0.641 × T) + (1.125 × U) + 57.292 (2) (Great Notley, Essex, UK). Two equations were generated out of this experiment, one for the OPC-N2 connected to the probe and one equation for the OPC-N2for which wasofcollocated with themonitoring TSI DRX sensors duringwith the their flight tests. Table 2. Resulting coefficients variables equations for dust P-value coefficients and confidence The resulting forlevel. the airborne module system equation and for the collocated module are shown in Equations (1) and (2) and Table 2: Sensor
Variable
Coefficients
p-Value
2.5 × 10−223
× R.H.) − (1.261 × T) +102.082 (1.822 × U) + 102.082 CUAV = (2.118 × CR ) − (1.175Intercept Raw Conc. PM10 (CR, µg/m3)
2.118
10−235 CColl = (0.197 × CR ) − (0.623Relative × R.H.) − (0.641 (1.125 × U)4.2+×57.292 Humidity (R.H.×%)T) +−1.175 Airborne module (CUAV)
Temperature (T, °C)
−1.261
Air speed (U, m/s)
1.822
Intercept
57.292
(1)
~0.0
(2)
2.1 × 10−231 2.8 × 10−159
Table 2. Resulting coefficients for variables of equations for dust monitoring sensors with their p-value NA 0.55 R2 and confidence level. Sensor
Variable
Raw Conc. PM10 (CR, µg/m3Coefficients ) 0.197
Intercept Relative Humidity (R.H. %)102.082 −0.623 Collocated module sensor (CColl) PM10 (CR , µg/m3 ) 2.118 Raw Conc. Temperature (T, °C) −0.641 Relative Humidity (R.H. %) −1.175
Airborne module (CUAV )
Temperature Air (T, ◦speed C) (U, m/s) Air speed (U, m/s) 2
−1.261 1.125 1.822
R2
0.55
R
NA = Not Applicable.
0.66
2.7 × 10−84 ~0.0p-Value −79 5.6 2.5 × 10× 10−223
1.7 ~0.0 × 10−72
4.2 × 10−235
−74 2.0 2.1 × 10× 10−231
NA2.8 × 10
−159
NA
Intercept 57.292 2.7 × 10−84 The R2 for both equations using a multivariate linear regression was above~0.0 0.5. The airborne Raw Conc. PM10 (CR , µg/m3 ) 0.197 moduleCollocated had a coefficient of 0.55 and the collocated module sensor of 0.66. Coefficients module Relative Humidity (R.H. %) −0.623 5.6 ×with 10−79a P-value (CColl ) is regarded as statistically
