• <tr id="yyy80"></tr>
  • <sup id="yyy80"></sup>
  • <tfoot id="yyy80"><noscript id="yyy80"></noscript></tfoot>
  • 99热精品在线国产_美女午夜性视频免费_国产精品国产高清国产av_av欧美777_自拍偷自拍亚洲精品老妇_亚洲熟女精品中文字幕_www日本黄色视频网_国产精品野战在线观看 ?

    Field test and probabilistic analysis of irregular steel debris casualty risks from a person-borne improvised explosive device

    2022-01-05 09:41:30PiotrSielikiMrkStewrtTomszGjewskiMihMlendowskiPiotrPeksHsnAlRifieRoertStudzinskiWojiehSumelk
    Defence Technology 2021年6期

    Piotr W. Sieliki , Mrk G. Stewrt , Tomsz Gjewski , Mih? Mlendowski ,Piotr Peks , Hsn Al-Rifie , Roert Studzi′nski , Wojieh Sumelk

    a Institute of Structural Analysis, Faculty of Civil and Transport Engineering, Poznan University of Technology, 5 Maria Sk?odowska-Curie Square, 60-965,Poznan, Poland

    b Institute of Building Engineering,Faculty of Civil and Transport Engineering,Poznan University of Technology,5 Maria Sk?odowska-Curie Square,60-965,Poznan, Poland

    c Centre for Infrastructure Performance and Reliability, The University of Newcastle, University Drive, Callaghan, NSW, 2308, Australia

    Keywords:Flying fragments Human safety Person-borne improvised explosive device experiment Probabilistic analysis

    ABSTRACT Person-borne improvised explosive devices (PBIEDs) are often used in terrorist attacks in Western countries. This study aims to predict the trajectories of PBIED fragments and the subsequent safety risks for people exposed to this hazard. An explosive field test with a typical PBIED composed of a plastic explosive charge and steel nut enhancements was performed to record initial fragment behaviour,including positions, velocity, and trajectory angles. These data were used to predict the full trajectory of PBIED fragments using a probabilistic analysis. In the probabilistic analyses a probability of fatality or serious injury was computed. Based on the results presented, many practical conclusions can be drawn,for instance, regarding safe evacuation distances if a person were exposed to a suspected PBIED.

    1. Introduction

    Improvised explosive devices (IEDs) have been a weapon of choice for terrorist attacks in Europe, North America, and other Western countries. Recent IED attacks include trains in Madrid(2004);trains and a bus in London(2004);government buildings in Oslo (2011), Stade de France, the Bataclan Theatre, and a caf′e in Paris (2015); an airport and train station in Brussels (2016), and Manchester Arena (2017). Most of these IED attacks involved a person-borne improvised explosive device (PBIED)containing less than 10 kg of explosives [1].This conclusion is not unexpected;an analysis of bomb incidents over the past 15 years found that most IEDs weigh less than 5 kg[2].Most of the attacks were conducted in urban environments where structures are exposed to the blast effects. As a result, there has been substantial research into the structural damage to buildings,bridges,pipelines,trains,and other infrastructure elements from blast pressure and impulse [3,4].However, most terrorist attack casualties in Western countries(post-9/11) have been from primary and secondary injuries [5,6]due to fragmentation and blast overpressure (e.g. lung rupture,brain acceleration, whole-body displacement, or skull fracture)rather than from falling columns, beams, slabs, or other structural components [7,8]. Infrastructure structural components have proven highly resilient to terrorist IED attacks; however, more research is needed to understand casualty risks from bomb fragmentation and blast overpressure hazards,especially from PBIEDs.Understanding of initial fragment mass, velocity, and distribution as well as the fragments’ drag coefficients, kinetic energy, throw range,and density probabilities is essential.For example,fragment mass, count distributions, and velocities are used in SAFER (Safety Assessment for Explosives Risk) and VAPO (Vulnerability Assessment and Protection Option), software developed by the U.S.Department of Defense Explosive Safety Board (DDESB) and the Defense Threat Reduction Agency(DTRA)to predict fragmentation distributions and casualty risks.However,these software programs are best suited to assessing military ordnance safety risks because they utilise predefined munitions parameters and fragment masses that are not easily applied to IEDs. IED fragment density is highly spatial, not symmetric, and heavily reliant on the explosive’s placement and the source material of the fragments.This paper was prompted by Yokohama et al.’s [9] conclusions where they noted that,“IED fragmentation data is unavailable since it is not published in the open literature;”“IED fragmentation is difficult to predict for its stochastic nature;”and“the current modelling capability of the IED fragmentation is less mature.” There are significant uncertainties and variabilities associated with modelling primary fragmentation.The explosive safety distances proposed by the U.S.Department of Defense Explosive Safety Board [10] consider some of these uncertainties when predicting casualty risks from accidental detonation of military ordnances. The DDESB [10] suggests that, when predicting casualty risks from accidental detonation of military ordnances, model accuracy is low at between -95%and +200%. Clearly, there is a need for improved modelling and statistical characterisation of the main variables associated with fragmentation. Characterisation of IED blast and fragmentation uncertainties using stochastic (probabilistic) methods is a logical step to improve risk estimates.

    One of the most common PBIEDs worldwide is the suicide vest.These devices were used in the 2015 Paris attacks and contained no more than 10 kg of triacetone triperoxide (TATP) as the explosive agent. Pachman et al. [11] examined TATP and found an average TNT equivalent from incident overpressure equal to 70%, and for impulse of the positive phase of the blast wave,it reduces to 55%;it may be assumed that a similar effect would be obtained from an approximately 6-8 kg TNT charge. Thus, an explosive charge of modest mass may significantly accelerate IED fragments,increasing the risk of mass casualties.

    The present paper assesses the casualty risks and safe evacuation distances for people exposed to suicide vest PBIEDs.The main aim of this paper is an improved understanding of the primary fragmentation of PBIED vests containing hundreds of steel nuts;its secondary aim is for the knowledge gained herein to be generalisable to other improvised projectiles intended to result in mass casualties.

    2. Materials and methods

    2.1. Experiment setup

    The primary objective of this experiment was to measure the initial velocity, trajectory angle, final range, and kinetic energy of fragments from a PBIED.Due to security reasons,it is not possible to show all details of the assembly and detonation of a PBIED.Nevertheless, the concept and geometry presented herein may be viewed as representative of some PBIEDs used in the 2015 November attacks in Paris. In this study, 4.43 g M8 stainless steel hex nuts were used as enhancements. Hex nuts were chosen for representing the flying projectiles because they are one of the most used in PBIED;mainly because they are easy to get by trade without raising suspicions. The setup of the device was built on the experiences of special forces, which often investigates such explosive devices. Thus, hex nuts arrangement and quantity are close to the real-world cases. The nuts were layered in front of the explosive charge in an open-faced sandwich, as shown in Fig.1. The layer of nuts comprised nine rows and 24 columns (a total of 216 nuts in one layer) measuring 120 mm × 310 mm, see Fig. 1(b). The PBIED-shown at A in Fig. 2(a)-consisted of the suicide vest containing charge of plastic explosive number four (PE4), a similarly-sized layer of clay a few centimetres thick,a rigid,bulletproof layer of Kevlar, and the layer of 216 hex nut projectiles weighing a total of 957 g.The PBIED was fixed to wooden stakes at a height of 0.9 m above the ground level,as shown in Fig.2(b).Due to the sensitivity of the data the mass of the IED was classified.

    The full experimental site is shown in Fig. 3. One of the objectives of this paper is to assess the trajectory of the debris originating from the PBIED;another objective is to evaluate the effect of PBIED fragment distribution on the car, shown at D in Figs. 3 and 4.Fragment distribution on the plywood panels(B,C,and E)remains a topic for future research. Plywood panel B is situated perpendicular to the camera view in order to create the background for camera for observing horizontally flying nuts.The car,D,is located 10.8 m from the PBIED, see Figs. 2(a) and Fig. 3.

    An important aspect of this test was to measure the velocity and trajectory of the hex nut enhancements during the first 10 ms of flight after acceleration by the explosive charge.Fragment velocity data was collected through video recording and later postprocessing of the data.The magnitude of velocity was estimated for 74 fragments using a high-speed video in combination with velocity screens and witness screens.

    Fig. 2. PBIED experiment set up: (a) back view and (b) front view.

    Fig.1. Elements of the PBEID used in the experiments: (a) the set-up plan and (b) the layer of 216 hex nuts.

    Fig. 3. Full (side) view of the experiment site.

    Fig. 4. Small-size car used in the experiment.

    2.2. Fragment loading measurements

    Several factors influence the reliability of the measurements of highly dynamic experiments.Verification that the values recorded are physically correct was allowed for in test preparations.Based on the previous experiences of the Poznan research group, it was crucial to define the location of the high-speed camera to obtain an accurate measurement of velocity of objects that could be traveling faster than the speed of sound. The goal was to measure 10 to 20 individual nuts, each with a different velocity, flying at different horizontal and vertical angles.

    The test scheme is presented in Fig. 5. It was assumed that the projectiles would be launched at point A and fly to Cn. The highspeed camera recording system was capable of following the objects in their first 10 m of the flight-from the detonation position, A, to the car, D. This trajectory is presented in Fig. 5, see red(rising) and green(falling) curves.

    Another important description of the trajectory of each nut is the vertical angle between the trajectory and horizontal direction-the angle θxin Fig.5.The perfect measurement assumes that a separated nut followed by the measurement equipment can be found at the final XMAXdistance,Cn.The objective of this paper is to predict the trajectory of all nuts during full flight trajectory using only initial launch conditions. This prediction can be obtained based on a classical system of equations of motion:

    where x and y are the coordinates of the plane of interest;t is time;Cdxand Cdyare the drag coefficients in x and y directions, respectively; Axand Ayare the nut area projections in x and y directions,respectively;ρ is air density; g is gravity; and m is fragment mass.The drag coefficients Cdxand Cdywere assumed at 0.8 for horizontal and 1.5 for vertical ascending and descending movements. For further consideration of the trajectory assessment, the crosssectional areas (Ax, Ay) of the nut are necessary. Because of the uniform geometry of projectiles, the authors assumed that the projected areas would be measured in two major orientations.Hence, the vertical and horizontal cross-sectional areas of a nut were equal Ax= 105.5 mm2and Ay= 129.3 mm2, respectively. In general,the shapes of projectiles may be highly irregular and thus,determining its drag coefficients and projected areas may be difficult to estimate,see Moxnes et al.[12]and Kljuno and Catovic[13].

    Fig. 5. Range and locations of the primary measurement areas.

    Fig. 6. Sequential time steps from 0 to 26.4 ms after detonation: blast wave movement (yellow dashed lines) and hex nut fragments in flight (red squares).

    3. Results and discussion

    The explosive field test was conducted in Poland close to the city of Poznan.Fig.6 shows time-lapse images of the detonation of the PBIED-the blast wave(shock front)and the fragment locations.As documented, the blast wave caused by the detonation overtakes the flying nuts in the first few milliseconds,see Fig.6(b).The blast wave front is marked with a yellow dashed line. Due to the resolution of the video,the blast wave velocity and how it decreases in time may be observed, compare Fig. 6(b) through Fig. 6(e). The average wave velocity reached 470 m/s,with an estimated incident overpressure value of 15-20 kPa at a distance of 8 m from the PBIED, Fig. 6(d).

    Fig. 6 shows the projectile cloud area; for clarity, nuts are marked by red squares. Each nut located within the recording has own location in the experiment based on velocity, trajectory, and time.During the experiment,side boundaries were used,consisting of a rigid box around the vest shown as G in Fig.1 (a), to catch all projectiles with trajectories greater than θy= 2.5°from a perfect horizontal trajectory or azimuth angle. Thus, it may be stated that the fragments are analysed in two-dimensional plane,i.e.XY plane,see Fig. 5.

    Acceleration of debris increases in the first few metres, and velocity reaches its maximum magnitude in the next 5-8 m,so the initial velocity was measured as the average velocity over that 5-8 m. The accuracy of the in-plane distance measurements from the high-speed video is about ±0.05 m, leading to an in-plane accuracy of velocity of about±0.5%.In addition,if an individual nut’s flight trajectory is 2.5°horizontally from the perfect(camera-view plane) trajectory, the out-of-plane error of distance while measuring velocity is approximately 0.01 m (10.8 m-0.8 m × cos 2.5°).Based on the slowest and the fastest fragment measurements,velocity measurements are accurate out-of-plane to less than±1%.Taking into account both in-plane and out-of-plane directions,the overall velocity measurement error is equal to 1.1%.

    The trajectories of 74 nuts were recorded during the field experiment. Two high-speed cameras recorded the velocity and distance measurements from different angles. The primary record was made using a Phantom v711 camera-as used in Sielicki et al.[14,32,34],-located 50 m from the PBIED and perpendicular to the virtual axis A-C,see Figs.3 and 6.A Phantom Miro320S was used to record the nuts that directly hit the car or the wooden witness obstacles, see camera F, Fig. 3.

    During the initial measurement of object trajectories, the number of frames recording these movements is crucial. With a recording speed of 2700 frames per second, Camera F recorded objects on 129 frames due to processing speed limitations.All flying objects were identified in each frame. A similar situation was observed on the primary camera, where the recording speed was 19000 fps, and more than 1100 frames were used. Automatic procedures for monitoring a moving object were applied and implemented in Scilab code with Phantom PCC 3.1 camera control software.

    The final data for all 74 measured nuts are presented in Table 2.Note that a negative value of θxwas obtained if the debris impacted the ground close to the PBIED. Also, some fragments with short landing distances were impossible to measure due to the obscuring effects of the fireball and dust cloud directly after detonation. The peak value recorded was 50°and for instance two pieces of projectile were found only 1.0 m and 2.5 m from the PBIED, thus yielding angle values of-43°and-21°,respectively.Results show that velocity decreased as angle increased-from 500 m/s to 200 m/s.

    Experimental data obtained from the field test established boundary conditions for the system’s equations of motion. Eq. (1)was used for each fragment to estimate trajectories with particular interest in final resting place and the velocity magnitude just before its contact with the ground, and the kinetic energy was computed.All those data are presented in Table 3. The trajectories computed from Eq. (1) are presented in Fig. 7, where the kinetic energy is shown in colour-black,blue,and grey-for the particular position of the fragment.The energetic criteria are used to assist in the analysis of the harmfulness of the fragments. Three levels of kinetic energy were assigned based on the kinetic energy of the fragment at particular moment.Injury due to the blast wave itself is not analysed in this study;however, this is an important aspect of low mass charge damage, as shown by Gajewski and Sielicki[15,30].

    In Fig. 7, fragments drawn in black show the positions of particles with high kinetic energy-greater than 300 J-consistent with typical military bullets (9 mm and greater calibres). As shown by Kerampran et al.[33];a 9×19 mm Parabellum bullet,used widely in military handguns, with a mass of approximately 8 g obtains a velocity of 355 m/s, which results in kinetic energy of approximately 500 J.

    Dots in light blue represent positions in which the kinetic energy is moderate-between 20 J and 300 J-which corresponds to the energy of 0.22LR calibre bullets.

    The grey dots represent positions with low kinetic energy-below 20 J-which corresponds to pre-charged pneumatic rifle bullet.In Fig.7,to underline the boundaries between moderate and low kinetic energy, white circles were added.

    Fragments with high kinetic energy (black) are highly dangerous to humans. These fragments encountering a human body will cause severe tissue damage, and are most likely pass through the body. Fragments with moderate kinetic energy (light blue) will cause less damage. They are likely to perforate the skin only without going through the body. Fragments with low kinetic energy(grey)are unlikely to perforate skin but will mark the skin,and severe bleeding is possible.

    The harmfulness of the fragments accelerated by the PBIED should be assessed based on two criteria. The first criterion is related to close-range harm-up to about 20 m-where the fragments have the greatest kinetic energy, and their trajectory is approximately horizontal (a low or negative vertical trajectory angle θx),or their path height is not greater than 2 m(equivalent to about maximum human height). According to this criterion, only the first row of people would be hit with a fragment, but these individuals would experience severe injuries and/or death. People in the second row could be injured if fragments pass through the first row, with or without encountering a first-row body; the second row injuries resulting from fragments that have already impacted others would likely be minor due to fragment energy decrease. In our study, the primary kinetic energy of close range fragments was above 300 J, seeTable 3.

    The second safety criterion is related to fragments falling after reaching their maximum heights,as illustrated by the green curve in Fig.5.The kinetic energy of fragments based on landing distances is shown in Fig. 8. Landing distances were between 120 m and 300 m,with major values greater than 200 m.The kinetic energy of all individual fragments was less than 70 J, with more than 90%below 20 J. Fragments landing at greater distances would have relatively low harmfulness;more individuals would be affected but injuries sustained would be less severe.

    Fig. 8. Kinetic energy of fragments at landing distances.

    4. Probabilistic data assessment

    4.1. Trajectory angle and fragment density

    Trajectory angle and fragment density may be treated as a bivariate normal distribution [10,17]; and [18] for vertical (θx) and horizontal (θy) trajectory angles:

    where μxand μyare the mean angles of θxand θyand where σxand σyare the standard deviations of angles θxand θy,respectively.If the trajectory angles are equally spread with no bias, then the mean angle is μx=μy=0o.It might be expected that variability of θxand θywill be similar. Pope[17] also notes that, for a PBIED,“there is a strong bias for low values of θxand θyto be selected,” suggesting that the variability of angle trajectory will be relatively low.

    A statistical analysis of the angle trajectories from 74 fragments(Table 2) reveals μx= 10.7°and σx= 18.8°. The skewness of data towards positive angles may be due to the detonator’s location near the bottom of the explosive mass.If the detonator had been placed more centrally,it is more likely that about half the fragments would have a positive angle trajectory,and the other half a negative angle trajectory (μx≈0°). If the data is limited to include only the 20 negative angles (from -43°to -0.9°)and the lowest 20 positive angles’ trajectories (from -0.65°to 6.02°), then μx= -3.4°and σx=8.7°,and as expected,the mean moves much closer to 0°and variability decreases. The statistics for angle trajectory are clearly variable and highly dependent on the characteristics of the PBIED composition and circumstances of detonation.

    Fig.9 shows that the normal distribution provides a reasonable fit for the data from the 74 fragments described in Table 2. More data is needed to increase confidence in the best probabilistic model fit.

    Statistics for the horizontal trajectory angle(θy)are not available because the test set-up only allowed fragments to deviate within±2.5°.Nonetheless,as noted above,it is expected that variability of θxand θywill be similar. As a starting point for analysis, two scenarios are considered:

    Fig. 9. Histogram of trajectory angles and normal probability distribution.

    1. PBIED used in current test:μx=μy= 10.7°and σx=σy= 18.8°

    2. Generic IED detonated symmetrically: μx= μy= 0.0°and σx=σy= 18.8°

    4.2. Probability of hit

    The projected impact area (Ap) is assumed to be rectangular in shape with an aspect ratio (ratio of height to width) of ahw; this allows the height h and width w to be estimated.It is then assumed that the projected area is centred at the same height of the PBIED used in the test (0.9 m), a schematic is shown in Fig.10. Note that though the fragments have an initial spatial distribution before detonation based on the layout of the hex nuts in a 120 mm×310 mm layer,for convenience,a“point source”ejection is assumed [17].

    The serious injury or fatality zone for a PBIED with a few kilograms of explosive is likely to be no more than about 20-30 m because at that range,the velocity of the fragments will be high and with a relatively close spatial density.At these short distances,it is appropriate to use a line-of-sight approach because,at high speeds,the fragments will travel in approximately a straight line over short distances of<100 m,as shown in Fig.6(b)and Fig.7.where R is the distance (range)from the PBIED to the individual (see Fig.10).

    The average number of incapacitating fragments,N,at a distance R hitting the projected area of the exposed person located directly in front of the PBIED is expressed thus:

    where n is the total number of fragments(n=216)and the bivariate fragment density distribution p(θx,θy) is given by Eq. (2).

    Seven fragments measured at the field test impacted the car D.The projected area of the car is 4.024 m2, and an aspect ratio of ahw= 0.33 is assumed. The dispersion angles from the test are μx=μy=10.7°and σx=σy=18.8°.With these observed parameters and R = 10.8 m, Equation (3) yields N(R) = 7.6 fragments colliding with the car. If ahw= 0.25 or 0.5, then N(R) of the colliding fragments is 7.6 and 7.8,respectively;therefore,the predicted number of colliding fragments is not sensitive to the aspect ratio of the target. A predicted value of N(R) = 7.6 fragments compares well with the observed value of seven colliding fragments and provides some validation of the probabilistic modelling of fragment distribution.

    Fig. 10. Schematic illustration of projected impact area, Ap. The trajectory angles needed to affect the projected area are as follows:

    The probability of being hit, Phit, by at least one fragment is deduced from the Poisson distribution by the following equation:

    4.3. Casualty risk assessment

    The projected area of an exposed person is Ap=0.26 m2[19].If it assumed that the aspect ratio for Apis a height to width of 2,then an exposed person is modelled as a rectangle of height h=0.72 m and width w = 0.36 m (0.36 m × 0.72 m = 0.26 m2).

    The density of fragments is not uniformly spread over the surface area of half a hemisphere centred on the exploding device.Fragment density and probability of being hit will be highest if the person stands close to and in front of the PBIED.Those values will reduce as the distance between the person and the IED increases,and the angle from the front of the device increases, i.e. as θyincreases to 90°. In this case, the average number of incapacitating fragments N at a distance R hitting the projected area of the exposed person is estimated from Eq. (3), and the probability of being hit by a fragment is calculated from Eq.(4)for each horizontal increment(θyi)until θy=±90°.These values are then used to infer the average hit probability for a person anywhere along a circumference with radius R from the PBIED. The probabilities computed according to Eq. (4)are presented in Fig.11.

    4.4. Initial fragment velocity

    Fig.12. Comparison of several probabilistic models with data histogram.

    The statistics of initial fragment velocity, vo, obtained from launch data from Table 2 are μv= 405 m/s and σv= 110 m/s. The variability of initial fragment velocity is high, with a coefficient of variation (COV) of 0.27. Fig.12 shows the normal, lognormal, Weibull, gamma and Gumbel probability distributions fitted to the experimental data. The comparisons show that the normal distribution provides a good fit to the data, although the lower tail of velocities appears under-predicted.The Kolmogorov-Smirnov test found that all probability models were not rejected at the 5% significance level.

    Note that the normally distributed random variables used in the present paper are truncated at zero to avoid negative values.Moreover, it should be noted that a probabilistic analysis can also consider non-standard probability distributions (e.g., bi-modal)and it may also allow for truncation of parameter values if there is a physical limit on that value.A benefit of Monte-Carlo simulation is that it can readily incorporate such non-ideal distributions into a probabilistic analysis.

    Fig.11. Probability of being hit by at least one fragment based on location (σx = σy = 18.8°).

    Most research on the initial velocity of fragments is based on cased explosives such as military munitions. Under such circumstances, Gurney equations are used, where the Gurney constant(√——--2E) is given for different types of high explosives. The Gurney constant for plastic explosive C4, which is equivalent to the PE4 used here[20,21]is 2530 m/s[22].Gurney equations are provided for cylindrical and spherical charges, thus applicable to most missiles,bombs,and mortars[22].However,a PBIED typically does not conform to these shape configurations. In our case, the PBIED is attached to stiff supports, which may be model as an infinitely tamped sandwich,see Fig.1(a).A Gurney analysis for this idealised configuration leads to an initial velocity of the hex nut fragments as follows [23]:

    where C is the explosive charge mass(kg)and M is the total mass of fragment enhancements(kg).For instance,if C=1 kg and M=1 kg,then initial fragment velocity is equal about 2400 m/s.If the PBIED is model as an open-faced sandwich with no tamping, then the initial fragment velocity is reduced to about 1400 m/s [23]. These calculated model velocities are significantly higher than those observed from the test, where the mean velocity is 405 m/s. This difference can be explained by the layer structure of the PBIED;because the fragments and the explosive were separated by a layer of clay and bulletproof material and not in direct contact,the crosssectional area of the 216 fragments was smaller than the explosive charge. Pope [17] graphically presents observations regarding“adaptation of Gurney analysis to handle fragment throw-out.”,with particular interest in its application to real events.

    4.5. Final fragment velocity

    The flight of fragments will be affected by air resistance and gravity, leading to the following form of fragment velocity as an approximation to the equations of motion given by Eq. (1) [19]:

    where k=0.002 for supersonic velocities(vo>335 m/s)and 0.0014 for subsonic velocities,vois the initial velocity of the fragment,m is the mass of the fragment (kg), x is the distance travelled by the fragment(m),and cdis the drag coefficient.The drag coefficient is a function of fragment shape, and as previously discussed, the drag coefficients of the nuts are in the range of 0.8-1.5. Hence, in the probabilistic analysis to follow,cdis taken as a uniform distribution[0.8,1.5].If cdis taken as 1.15(mean value),then the final fragment velocity at 20 m decreases by approximately 18% and 25% for subsonic and supersonic velocities, respectively.

    4.6. Injury models

    The injury from penetrating fragments“is of particular interest to the military, and much of the data are from military sources”[19]. The military are more concerned with injuries that will incapacitate but not kill a soldier.However,an incapacitation model has been adapted to estimate fatal and serious injuries by the probit equation:

    Table 1 Random variables for casualty risks.

    where v is the velocity of the fragment(m/s) and m is the mass of the fragment (kg). The probability of serious injury or fatality exceeds 90% if the impact velocity exceeds 190 m/s.

    Kinetic energy-based calculations and other probabilistic analyses [19,24] may be used with other injury criteria defined by NATO and the U.S. Department of Defense to predict injuries-for example, incapacitation may occur if the kinetic energy exceeds 79 J(DOD,2009,[25].In the instant matter,probabilistic analysis is used to derive the probability of fatal and serious injuries for an individual:

    · directly in front of the PBIED (θx=θy= 0°)

    · anywhere within a 90°zone (θy= ±45°)

    · anywhere within a 180°zone (θy= ±90°)

    at distances from 5 m to 20 m from the PBIED. The random variables are shown in Table 1.

    Results from the Monte Carlo simulation are shown in Fig.13 for 10000 simulation runs.As expected,casualty risks are significantly higher if the person stands directly in front of the PBIED, and the odds reduce if a person is located somewhere else, i.e. θy> 0°.Schematics of the location of victims from the 2017 Manchester Arena bombing [26] suggest that most victims were within 10-20 m of the PBIED. The casualty risks in Fig.13 also show that this range is the zone of high risk, and the risk diminishes beyond 20 m.For example,at 30 m,the casualty risks are about 5%for the 90°zone.

    Risks may also be compared with risk acceptance criteria. For example, the UN recommends that, to be considered safe, evacuation distances ensure that the density of hazardous fragments(79 J)is less than 1 per 56 m2; this is approximately equivalent to a 1%probability of a person being struck by a lethal fragment[22].If this is the criterion,then an outdoor safe evacuation distance based on a 180°zone is 52 m; this would, however, increase to 69 m if the distance were directly in front of the PBIED. If μx= μy= 0.0°and σx=σy= 18.8°, the safety distance changes to 46 m and 83 m for the 180°zone and directly in front, respectively. In these cases, a conservative safe evacuation distance for the PBIED considered in the present paper would be approximately 85 m. This is a reasonable estimate given that the United States National Counter Terrorism Center [27] recommends an outdoor safe evacuation distance of 259 m for a 2.3 kg pipe bomb.

    4.7. Sensitivity analysis

    Fig.13. Probability of fatality or serious injury (σx = σy = 18.8°).

    Fig.14. Change in casualty risks for person directly in front of a PBIED (σx = σy = 18.8°).

    A sensitivity study is carried out to assess the relative impact of the variability and uncertainty of the model parameters on casualty risks. This was achieved by running the Monte Carlo simulation analysis with each parameter in turn modelled deterministically while all other parameters given in Fig.14 are modelled probabilistically. In Fig.14, the sensitivity study results are shown as the percentage point-change in casualty risks for a person standing directly in front of a PBIED at a distance of 5 m-20 m. Results are insensitive to drag coefficient variability, and thus, it is concluded that drag coefficient has a negligible influence on the results.Variability of initial fragment velocity changes the casualty risks by less than 1.4% in absolute terms. Clearly, if variability of fragment trajectory angle is omitted from the probabilistic analysis (i.e.,σx=σy=0°),then casualty risk is near 100% when μx=μy=0.0°.

    5. Conclusions

    Suicide vest PBIEDs are an important and current topic. An explosive field test for a typical PBIED containing steel fragment(hex nuts) enhancements was used to determine the initial velocities, initial trajectory angles, and kinetic energy of fragments.Those data were then used as boundary conditions to predict the trajectory of suicide vest PBIED fragments according to the classical system of equations of motion.An analysis of fragment harmfulness over various distance from the PBIED detonation. Two crucial aspects for safety were observed:severe harmfulness occurs at close range but with fewer individuals injured, and due to fragments losing velocity, at longer distances, lower harmfulness occurs but more individuals are injured.

    Moreover,a probabilistic analysis was then conducted to predict the probability of a person being hit by at least one fragment, a probability of fatality or serious injury, and safe evacuation distances.It was found that fatality risks are 65-95%within 5 m of the PBIED and fall to less than 1% for distances of about 85 m.

    Declaration of competing interest

    The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

    Acknowledgements

    This work was supported by the Poland National Center for Research and Development, under the grant DOB-BIO10/01/02/2019 within the Defence and Security Programme.

    Appendix A

    Table 2 Actual measurements of 74 flying nut fragments in the first 0-50 ms after detonation (in descending order of initial incident angle).

    Appendix B

    Table 3 Prediction of the final fragment range and kinetic energy at ground impact.

    Table 3 (continued)

    50天的宝宝边吃奶边哭怎么回事| 母亲3免费完整高清在线观看| 成人国产一区最新在线观看| 啦啦啦视频在线资源免费观看| 免费观看人在逋| 97人妻天天添夜夜摸| 日韩欧美在线二视频 | 久久久久久免费高清国产稀缺| 色综合欧美亚洲国产小说| 欧美在线一区亚洲| 日韩欧美一区二区三区在线观看 | 国产野战对白在线观看| 国产激情欧美一区二区| 一区二区三区精品91| 成年动漫av网址| 热99国产精品久久久久久7| 精品国产一区二区久久| 国产精品一区二区在线观看99| 精品国产亚洲在线| 免费在线观看日本一区| 在线观看日韩欧美| 91成年电影在线观看| 国产在线观看jvid| 欧美中文综合在线视频| 高清毛片免费观看视频网站 | 国产精品一区二区精品视频观看| 日韩欧美在线二视频 | 国产成人免费无遮挡视频| 久久99一区二区三区| 黑人巨大精品欧美一区二区mp4| 欧美日韩亚洲国产一区二区在线观看 | 丰满迷人的少妇在线观看| а√天堂www在线а√下载 | 日本黄色日本黄色录像| 大型黄色视频在线免费观看| 亚洲中文日韩欧美视频| 最新美女视频免费是黄的| 老司机午夜福利在线观看视频| 男女免费视频国产| 亚洲,欧美精品.| 国产精品综合久久久久久久免费 | 午夜激情av网站| 热re99久久国产66热| av天堂久久9| 大陆偷拍与自拍| 欧美激情 高清一区二区三区| 久久久精品免费免费高清| 可以免费在线观看a视频的电影网站| 极品教师在线免费播放| 久久国产精品男人的天堂亚洲| 水蜜桃什么品种好| 黄色a级毛片大全视频| 精品无人区乱码1区二区| 交换朋友夫妻互换小说| 人人妻人人爽人人添夜夜欢视频| 亚洲成人免费av在线播放| 亚洲自偷自拍图片 自拍| 亚洲欧美日韩高清在线视频| 中文字幕人妻熟女乱码| 免费高清在线观看日韩| 日韩 欧美 亚洲 中文字幕| 黄网站色视频无遮挡免费观看| 欧美精品av麻豆av| 亚洲熟女精品中文字幕| 90打野战视频偷拍视频| 国产淫语在线视频| 亚洲成人免费av在线播放| 下体分泌物呈黄色| 91大片在线观看| 亚洲成国产人片在线观看| 国产区一区二久久| 99久久国产精品久久久| 视频区图区小说| 69精品国产乱码久久久| 成人亚洲精品一区在线观看| 老熟女久久久| 黑人巨大精品欧美一区二区蜜桃| 女人被躁到高潮嗷嗷叫费观| 久久久久久免费高清国产稀缺| 99国产精品99久久久久| 国产av又大| 99精国产麻豆久久婷婷| 女性生殖器流出的白浆| 欧美激情极品国产一区二区三区| a级片在线免费高清观看视频| 热re99久久精品国产66热6| 国产精品99久久99久久久不卡| 18禁黄网站禁片午夜丰满| 最近最新免费中文字幕在线| 啪啪无遮挡十八禁网站| 中国美女看黄片| 免费看十八禁软件| 国产一区二区三区在线臀色熟女 | 成人精品一区二区免费| 国产精品欧美亚洲77777| 国产精品欧美亚洲77777| 天天操日日干夜夜撸| 少妇被粗大的猛进出69影院| 日韩欧美国产一区二区入口| 久久香蕉精品热| 色婷婷久久久亚洲欧美| 亚洲色图综合在线观看| 欧美精品av麻豆av| 亚洲一区高清亚洲精品| 99香蕉大伊视频| 国产欧美日韩综合在线一区二区| 一级片免费观看大全| a级毛片在线看网站| 成在线人永久免费视频| 久热这里只有精品99| 伊人久久大香线蕉亚洲五| 麻豆乱淫一区二区| 亚洲一区二区三区不卡视频| 免费在线观看日本一区| 久久国产精品人妻蜜桃| 亚洲在线自拍视频| 久久久久久久久免费视频了| 国产高清videossex| 最近最新中文字幕大全电影3 | 成人特级黄色片久久久久久久| 在线十欧美十亚洲十日本专区| 午夜福利影视在线免费观看| e午夜精品久久久久久久| 亚洲九九香蕉| 黄色女人牲交| av欧美777| 久久中文看片网| 19禁男女啪啪无遮挡网站| 亚洲精品一卡2卡三卡4卡5卡| 久久 成人 亚洲| 欧美黄色淫秽网站| 男女下面插进去视频免费观看| 欧美久久黑人一区二区| av在线播放免费不卡| 色在线成人网| 亚洲全国av大片| 国产精品影院久久| 国产亚洲精品第一综合不卡| 国产成人欧美| 一二三四在线观看免费中文在| 好看av亚洲va欧美ⅴa在| 高潮久久久久久久久久久不卡| 老司机福利观看| 色婷婷av一区二区三区视频| 亚洲一卡2卡3卡4卡5卡精品中文| 亚洲色图 男人天堂 中文字幕| 国产精品自产拍在线观看55亚洲 | 在线观看午夜福利视频| 丝袜人妻中文字幕| 中文欧美无线码| 免费观看a级毛片全部| 黄色片一级片一级黄色片| 波多野结衣一区麻豆| 天天添夜夜摸| 人人妻人人爽人人添夜夜欢视频| 五月开心婷婷网| 亚洲中文av在线| 色94色欧美一区二区| 国产一区有黄有色的免费视频| 一级黄色大片毛片| 老司机靠b影院| 99久久人妻综合| 亚洲熟妇中文字幕五十中出 | 十八禁高潮呻吟视频| 久久国产乱子伦精品免费另类| 波多野结衣av一区二区av| tube8黄色片| 天天躁狠狠躁夜夜躁狠狠躁| 色在线成人网| 亚洲国产精品sss在线观看 | www.精华液| 夜夜躁狠狠躁天天躁| 国产亚洲精品一区二区www | 久久中文看片网| 午夜视频精品福利| 999久久久国产精品视频| 国产一区二区三区视频了| 韩国精品一区二区三区| 亚洲国产精品合色在线| 日韩免费高清中文字幕av| 国产激情欧美一区二区| 中文字幕av电影在线播放| 国产1区2区3区精品| 一级毛片高清免费大全| 久久精品国产综合久久久| 黑人巨大精品欧美一区二区mp4| 手机成人av网站| 精品欧美一区二区三区在线| av网站在线播放免费| 久久人妻熟女aⅴ| videos熟女内射| 无限看片的www在线观看| 国产精品欧美亚洲77777| 色综合欧美亚洲国产小说| 最近最新中文字幕大全免费视频| 久久国产精品大桥未久av| 精品久久久久久久毛片微露脸| 日韩三级视频一区二区三区| 久久久久精品人妻al黑| 男女免费视频国产| tocl精华| 亚洲欧美一区二区三区黑人| 俄罗斯特黄特色一大片| 久久久久久久久久久久大奶| 欧美乱妇无乱码| 免费在线观看黄色视频的| 国产精品久久电影中文字幕 | 国产精品一区二区免费欧美| 成人18禁高潮啪啪吃奶动态图| 男女高潮啪啪啪动态图| 午夜91福利影院| 建设人人有责人人尽责人人享有的| av天堂在线播放| 久久精品aⅴ一区二区三区四区| 精品亚洲成国产av| 国产深夜福利视频在线观看| 国产视频一区二区在线看| 一边摸一边抽搐一进一出视频| 国产亚洲精品久久久久久毛片 | av超薄肉色丝袜交足视频| 大香蕉久久成人网| 国产麻豆69| 成熟少妇高潮喷水视频| 人妻丰满熟妇av一区二区三区 | 欧美性长视频在线观看| 国产精品久久久久久精品古装| 男女午夜视频在线观看| 一a级毛片在线观看| 黄片小视频在线播放| 久久精品熟女亚洲av麻豆精品| 女人高潮潮喷娇喘18禁视频| 亚洲,欧美精品.| 国产男女超爽视频在线观看| 亚洲av欧美aⅴ国产| 国产亚洲欧美在线一区二区| 日韩熟女老妇一区二区性免费视频| 狠狠婷婷综合久久久久久88av| 母亲3免费完整高清在线观看| 国产精品成人在线| 黄色片一级片一级黄色片| 中文字幕色久视频| 中文字幕色久视频| 欧美日韩一级在线毛片| 成年女人毛片免费观看观看9 | 色尼玛亚洲综合影院| xxx96com| 亚洲成人手机| 女性被躁到高潮视频| 777久久人妻少妇嫩草av网站| 精品国产亚洲在线| 欧美+亚洲+日韩+国产| 在线观看一区二区三区激情| 精品亚洲成a人片在线观看| 精品视频人人做人人爽| 高清视频免费观看一区二区| 精品久久久久久久毛片微露脸| 国产成人精品无人区| 十八禁人妻一区二区| 精品熟女少妇八av免费久了| 18禁裸乳无遮挡动漫免费视频| 日日夜夜操网爽| 亚洲欧美一区二区三区久久| 国产精品久久久久成人av| 悠悠久久av| 两个人免费观看高清视频| 51午夜福利影视在线观看| 精品久久久久久电影网| 看免费av毛片| a级毛片黄视频| 妹子高潮喷水视频| 自线自在国产av| av视频免费观看在线观看| 午夜福利欧美成人| 一级a爱片免费观看的视频| 久久久水蜜桃国产精品网| 一区二区三区精品91| 国产精品久久电影中文字幕 | 欧美 亚洲 国产 日韩一| 亚洲色图av天堂| 久久久久久免费高清国产稀缺| 国产99白浆流出| 国产亚洲精品第一综合不卡| 久久香蕉国产精品| 亚洲精品自拍成人| 欧美成人免费av一区二区三区 | 男男h啪啪无遮挡| 亚洲成人免费电影在线观看| 老熟女久久久| 丰满的人妻完整版| 精品国产一区二区三区久久久樱花| 成人手机av| 成人18禁高潮啪啪吃奶动态图| 欧美 亚洲 国产 日韩一| 国产乱人伦免费视频| 下体分泌物呈黄色| 久久这里只有精品19| 欧美国产精品一级二级三级| 午夜91福利影院| 悠悠久久av| 亚洲国产精品sss在线观看 | av电影中文网址| 国产精品久久久久久精品古装| 精品一区二区三卡| 亚洲精品国产精品久久久不卡| 夜夜爽天天搞| 久久久久久亚洲精品国产蜜桃av| 精品国产乱子伦一区二区三区| 手机成人av网站| 久久久精品区二区三区| 欧美午夜高清在线| 亚洲精华国产精华精| 午夜成年电影在线免费观看| 欧美 亚洲 国产 日韩一| 一边摸一边做爽爽视频免费| 天天躁狠狠躁夜夜躁狠狠躁| 一二三四社区在线视频社区8| 国产精品久久久久久人妻精品电影| 亚洲人成电影免费在线| 高清在线国产一区| 亚洲一区二区三区不卡视频| 国产精品久久久久久精品古装| 精品视频人人做人人爽| 国产一区二区三区在线臀色熟女 | 国产精品一区二区在线观看99| 两性夫妻黄色片| 日日夜夜操网爽| 12—13女人毛片做爰片一| 十八禁网站免费在线| 国产色视频综合| 亚洲欧美日韩高清在线视频| 超碰97精品在线观看| 高清视频免费观看一区二区| 日韩熟女老妇一区二区性免费视频| 99精国产麻豆久久婷婷| 成人黄色视频免费在线看| 50天的宝宝边吃奶边哭怎么回事| 在线观看www视频免费| 精品人妻熟女毛片av久久网站| 欧美精品人与动牲交sv欧美| 性少妇av在线| 久久人人爽av亚洲精品天堂| 亚洲国产精品合色在线| www.自偷自拍.com| 露出奶头的视频| 国产精品秋霞免费鲁丝片| 黑人操中国人逼视频| 国产野战对白在线观看| videosex国产| 亚洲人成伊人成综合网2020| 99久久99久久久精品蜜桃| 久久久国产欧美日韩av| 捣出白浆h1v1| 超色免费av| 高潮久久久久久久久久久不卡| 12—13女人毛片做爰片一| 在线天堂中文资源库| 精品一区二区三区四区五区乱码| 午夜91福利影院| 久久久久久人人人人人| 丁香欧美五月| 丰满的人妻完整版| 母亲3免费完整高清在线观看| www.熟女人妻精品国产| 日本黄色视频三级网站网址 | 91国产中文字幕| 欧美成人免费av一区二区三区 | 动漫黄色视频在线观看| 午夜福利一区二区在线看| 精品福利永久在线观看| 婷婷丁香在线五月| 不卡av一区二区三区| 日韩成人在线观看一区二区三区| 国产精品自产拍在线观看55亚洲 | 亚洲精品在线美女| 欧美黑人欧美精品刺激| 国产av又大| 一级毛片高清免费大全| av天堂在线播放| 亚洲av熟女| 午夜成年电影在线免费观看| 精品久久久精品久久久| 自线自在国产av| 久久久久久久午夜电影 | 中文欧美无线码| 免费一级毛片在线播放高清视频 | 国产亚洲欧美在线一区二区| 免费观看a级毛片全部| 国产精品一区二区免费欧美| 黄网站色视频无遮挡免费观看| 精品国产美女av久久久久小说| 久久久久久久久久久久大奶| 亚洲国产中文字幕在线视频| 美国免费a级毛片| 村上凉子中文字幕在线| 欧美午夜高清在线| 国产一区二区三区综合在线观看| 如日韩欧美国产精品一区二区三区| 国产极品粉嫩免费观看在线| 亚洲欧洲精品一区二区精品久久久| 国产精品免费视频内射| 在线永久观看黄色视频| 免费看十八禁软件| 丰满迷人的少妇在线观看| 久久香蕉精品热| 人人妻人人爽人人添夜夜欢视频| 久久精品91无色码中文字幕| 搡老乐熟女国产| 国产无遮挡羞羞视频在线观看| 男女床上黄色一级片免费看| 午夜精品在线福利| 啦啦啦在线免费观看视频4| 老汉色∧v一级毛片| 757午夜福利合集在线观看| 精品人妻1区二区| 国产精品自产拍在线观看55亚洲 | 老司机在亚洲福利影院| 十八禁网站免费在线| 精品一区二区三区四区五区乱码| 又黄又粗又硬又大视频| 亚洲av美国av| 极品人妻少妇av视频| 老熟妇乱子伦视频在线观看| 亚洲男人天堂网一区| 十分钟在线观看高清视频www| 777久久人妻少妇嫩草av网站| 午夜福利乱码中文字幕| 男人的好看免费观看在线视频 | 国产真人三级小视频在线观看| 久久人妻熟女aⅴ| 色尼玛亚洲综合影院| 一二三四在线观看免费中文在| 国产精品香港三级国产av潘金莲| 69av精品久久久久久| 大片电影免费在线观看免费| 建设人人有责人人尽责人人享有的| 18禁国产床啪视频网站| 天堂动漫精品| 色婷婷av一区二区三区视频| 日韩免费高清中文字幕av| 日本欧美视频一区| 母亲3免费完整高清在线观看| 国产日韩一区二区三区精品不卡| 老汉色∧v一级毛片| 久久久久久久午夜电影 | a在线观看视频网站| xxxhd国产人妻xxx| 亚洲美女黄片视频| 亚洲黑人精品在线| 夜夜躁狠狠躁天天躁| 欧美av亚洲av综合av国产av| 国产成人影院久久av| 亚洲成人手机| 久久天堂一区二区三区四区| bbb黄色大片| 香蕉久久夜色| 久久久精品免费免费高清| 大香蕉久久成人网| 黄色毛片三级朝国网站| 国产日韩欧美亚洲二区| 母亲3免费完整高清在线观看| 国产aⅴ精品一区二区三区波| 少妇粗大呻吟视频| 午夜福利免费观看在线| 亚洲第一欧美日韩一区二区三区| 国产精品1区2区在线观看. | 黄频高清免费视频| 亚洲精品国产色婷婷电影| 欧美 日韩 精品 国产| 国产亚洲精品久久久久5区| 黄色丝袜av网址大全| 好男人电影高清在线观看| 亚洲aⅴ乱码一区二区在线播放 | 国产免费av片在线观看野外av| 免费在线观看黄色视频的| 又黄又爽又免费观看的视频| 精品久久久久久久毛片微露脸| 欧美精品av麻豆av| 嫩草影视91久久| 国产精品久久视频播放| 91精品国产国语对白视频| 亚洲精品久久成人aⅴ小说| 国产成人av教育| 国产视频一区二区在线看| 久久国产精品大桥未久av| 国产在线精品亚洲第一网站| 婷婷精品国产亚洲av在线 | 午夜日韩欧美国产| 一级a爱片免费观看的视频| 国产色视频综合| 岛国在线观看网站| 亚洲熟妇熟女久久| 欧美日韩亚洲国产一区二区在线观看 | 国产成人啪精品午夜网站| 午夜影院日韩av| 午夜精品在线福利| 在线十欧美十亚洲十日本专区| 岛国毛片在线播放| 99久久综合精品五月天人人| 亚洲精品中文字幕在线视频| 国产成人精品久久二区二区91| 少妇猛男粗大的猛烈进出视频| 久久久精品免费免费高清| 黄色片一级片一级黄色片| 国产精品一区二区精品视频观看| 亚洲av日韩精品久久久久久密| 国产aⅴ精品一区二区三区波| 日韩成人在线观看一区二区三区| 国产av又大| 亚洲精品成人av观看孕妇| 18禁美女被吸乳视频| 9色porny在线观看| 99re在线观看精品视频| 欧美日韩一级在线毛片| 91在线观看av| 乱人伦中国视频| 成年人黄色毛片网站| 岛国在线观看网站| 日韩熟女老妇一区二区性免费视频| 91老司机精品| 悠悠久久av| 免费看十八禁软件| 老司机深夜福利视频在线观看| 亚洲视频免费观看视频| 久久精品国产99精品国产亚洲性色 | 十八禁高潮呻吟视频| 亚洲av熟女| 亚洲欧美一区二区三区久久| 一本一本久久a久久精品综合妖精| 欧美一级毛片孕妇| 好男人电影高清在线观看| 18在线观看网站| 国产免费男女视频| 美女高潮到喷水免费观看| 正在播放国产对白刺激| 国产成人系列免费观看| 女警被强在线播放| 久久久精品区二区三区| 亚洲aⅴ乱码一区二区在线播放 | 久久天堂一区二区三区四区| 老司机午夜福利在线观看视频| 久久婷婷成人综合色麻豆| 亚洲黑人精品在线| 一本大道久久a久久精品| 女人久久www免费人成看片| 免费观看精品视频网站| 丰满的人妻完整版| 婷婷丁香在线五月| 亚洲美女黄片视频| bbb黄色大片| 中文字幕制服av| 午夜精品在线福利| 女同久久另类99精品国产91| 国产精品久久视频播放| 欧美不卡视频在线免费观看 | 啦啦啦在线免费观看视频4| 热99久久久久精品小说推荐| 国产男女内射视频| 啦啦啦免费观看视频1| 久久久精品国产亚洲av高清涩受| 亚洲熟女毛片儿| 国产精品久久视频播放| 亚洲自偷自拍图片 自拍| 久久久国产精品麻豆| 最新美女视频免费是黄的| 他把我摸到了高潮在线观看| 欧美日韩亚洲国产一区二区在线观看 | 97人妻天天添夜夜摸| 性色av乱码一区二区三区2| 日韩免费高清中文字幕av| 极品教师在线免费播放| 99精品在免费线老司机午夜| 99精品欧美一区二区三区四区| 亚洲精品国产色婷婷电影| 久久亚洲精品不卡| 熟女少妇亚洲综合色aaa.| 国产日韩欧美亚洲二区| 女警被强在线播放| 中文字幕制服av| 欧美人与性动交α欧美精品济南到| 老汉色∧v一级毛片| 男人操女人黄网站| 12—13女人毛片做爰片一| 国产熟女午夜一区二区三区| 黄色 视频免费看| 国产精品偷伦视频观看了| 九色亚洲精品在线播放| 亚洲av熟女| 不卡av一区二区三区| 香蕉丝袜av| 黑人操中国人逼视频| 国产精品一区二区在线观看99| 亚洲av成人一区二区三| 成年人免费黄色播放视频| 中文字幕av电影在线播放| 亚洲精品粉嫩美女一区| 国产精品电影一区二区三区 | 亚洲精品乱久久久久久| 女人高潮潮喷娇喘18禁视频| av片东京热男人的天堂| 亚洲伊人色综图| 精品一区二区三区四区五区乱码| 天天躁夜夜躁狠狠躁躁| 欧美色视频一区免费| 女人被躁到高潮嗷嗷叫费观| 国产熟女午夜一区二区三区| ponron亚洲| 99香蕉大伊视频| 午夜视频精品福利| 18禁黄网站禁片午夜丰满| 午夜精品国产一区二区电影| 日本wwww免费看| 极品少妇高潮喷水抽搐| 国产区一区二久久| 欧美av亚洲av综合av国产av| 久久精品国产99精品国产亚洲性色 |