2018 Vol. 38, No. 3
Display Method:
2018, 38(3): 473-484.
doi: 10.11883/bzycj-2017-0230
Read OL
PDF 
Cited by
Abstract:
As one of the major damage modes of an aircraft fuel tank, the hydrodynamic ram may cause a catastrophic destroy to the fuel tank structure. In order to analyze the multi-fragment impact on the fuel tank under actual combat circumstance, a validated full-filled fuel tank model under single fragment impact is firstly established. The impact situations of multi-fragment hit at the same time, double fragments hit with different spaces, and double fragments hit at different times are studied by using the finite element analysis software of ANSYS/LS-DYNA. The following parameters are carefully analyzed, including the pressure inside the fuel tank, the fragment velocity decay, the total energy absorbed by the water and the deformation of the fuel tank walls. The results show that, the peak pressure of the fluid in the tank is derived from the shock wave which is produced by the fragment; the superposition effect of pressure is enhanced under multi-fragment impact condition; the tank walls significantly deform with the growth of the number of incident fragments. Moreover, the residual velocity of the incident fragment will be increased by the latish one.
As one of the major damage modes of an aircraft fuel tank, the hydrodynamic ram may cause a catastrophic destroy to the fuel tank structure. In order to analyze the multi-fragment impact on the fuel tank under actual combat circumstance, a validated full-filled fuel tank model under single fragment impact is firstly established. The impact situations of multi-fragment hit at the same time, double fragments hit with different spaces, and double fragments hit at different times are studied by using the finite element analysis software of ANSYS/LS-DYNA. The following parameters are carefully analyzed, including the pressure inside the fuel tank, the fragment velocity decay, the total energy absorbed by the water and the deformation of the fuel tank walls. The results show that, the peak pressure of the fluid in the tank is derived from the shock wave which is produced by the fragment; the superposition effect of pressure is enhanced under multi-fragment impact condition; the tank walls significantly deform with the growth of the number of incident fragments. Moreover, the residual velocity of the incident fragment will be increased by the latish one.
2018, 38(3): 485-490.
doi: 10.11883/bzycj-2017-0379
Abstract:
To understand the detonation reaction process of the TATB-based JB-9014 explosive, experimental measurements of the detonation wave profile of solid explosives using photon Doppler velocimetry (PDV) were performed. Planar detonations were produced by impacting the explosive with a sapphire flyer in a gas gun. Particle velocity wave profiles were measured at the explosive/window interface using PDV. LiF windows with very thin vapor deposited aluminum mirrors were used, and the time resolution of the measuring system was less than 2 ns. The interface velocity histories were derived to determine the reaction zone length and reaction time. The results show that the reaction time of JB-9014 is (0.26±0.02) μs, and the corresponding reaction zone length is (1.5±0.2) mm. The pressure at the end of chemical reaction is 27.3 GPa, the pressure at von Neumann spike is 40.3 GPa.
To understand the detonation reaction process of the TATB-based JB-9014 explosive, experimental measurements of the detonation wave profile of solid explosives using photon Doppler velocimetry (PDV) were performed. Planar detonations were produced by impacting the explosive with a sapphire flyer in a gas gun. Particle velocity wave profiles were measured at the explosive/window interface using PDV. LiF windows with very thin vapor deposited aluminum mirrors were used, and the time resolution of the measuring system was less than 2 ns. The interface velocity histories were derived to determine the reaction zone length and reaction time. The results show that the reaction time of JB-9014 is (0.26±0.02) μs, and the corresponding reaction zone length is (1.5±0.2) mm. The pressure at the end of chemical reaction is 27.3 GPa, the pressure at von Neumann spike is 40.3 GPa.
2018, 38(3): 491-500.
doi: 10.11883/bzycj-2017-0255
Abstract:
The dynamic propagation mechanics of prefabricated crack and its effect on joint were studied for the impact fracture process of L-shaped beam-column specimens with prefabricated cracks. The drop testing was carried out on the specimens with cracks at different distances l far from node core area by using the digital dynamic caustic system and the variations of crack propagation trajectory, velocity and dynamic stress-intensity factor were obtained. The results show that the increase of l leads to the angle between the penetration point of the extended crack at the edge of the beam and the prefabricated crack increases gradually and the degree of crack curving becomes larger. With the increase of l, the oscillation of crack propagation velocity increases, the average velocity decreases and the crack propagation time is gradually prolonged. When the value of l is 2 mm, the fracture pattern is mode Ⅰ. As the value of l increases, the crack tip is increased by the shear stress, the dynamic stress intensity factor of mode Ⅰ decreases and the dynamic stress intensity factor of mode Ⅱ gradually increases. The fracture pattern gradually becomes Ⅰ-Ⅱ complex mode.
The dynamic propagation mechanics of prefabricated crack and its effect on joint were studied for the impact fracture process of L-shaped beam-column specimens with prefabricated cracks. The drop testing was carried out on the specimens with cracks at different distances l far from node core area by using the digital dynamic caustic system and the variations of crack propagation trajectory, velocity and dynamic stress-intensity factor were obtained. The results show that the increase of l leads to the angle between the penetration point of the extended crack at the edge of the beam and the prefabricated crack increases gradually and the degree of crack curving becomes larger. With the increase of l, the oscillation of crack propagation velocity increases, the average velocity decreases and the crack propagation time is gradually prolonged. When the value of l is 2 mm, the fracture pattern is mode Ⅰ. As the value of l increases, the crack tip is increased by the shear stress, the dynamic stress intensity factor of mode Ⅰ decreases and the dynamic stress intensity factor of mode Ⅱ gradually increases. The fracture pattern gradually becomes Ⅰ-Ⅱ complex mode.
2018, 38(3): 501-508.
doi: 10.11883/bzycj-2016-0272
Abstract:
Effects of the three-point initiation control parameters on the formation of an explosively-formed projectile (EFP) with fins were studied, including initiation diameter and initiation synchronization error. The Mach wave parameters (pressure and Mach wave domain area) at the metal liner were calculated theoretically at different initiation diameters. The fin formation and stable-flight velocity of EFP caused by the synchronization error at different initiation diameters were investigated by employing the LS-DYNA software. Results indicate that the Mach wave domain area decreases, but the pressure, final velocity and length-to-diameter ratio of the EFP increase, as the initiation diameter increases. It is noted that a better empennage formation can be approached when the initiation synchronization errors at the initiation diameters of 30, 40, 50 mm are less than 50, 100, 150 ns, respectively. In addition, the lateral velocity of the EFP is lower and the flight stability is higher when the delay time of the middle detonation point is about half the maximum delay time of the three initiation points.
Effects of the three-point initiation control parameters on the formation of an explosively-formed projectile (EFP) with fins were studied, including initiation diameter and initiation synchronization error. The Mach wave parameters (pressure and Mach wave domain area) at the metal liner were calculated theoretically at different initiation diameters. The fin formation and stable-flight velocity of EFP caused by the synchronization error at different initiation diameters were investigated by employing the LS-DYNA software. Results indicate that the Mach wave domain area decreases, but the pressure, final velocity and length-to-diameter ratio of the EFP increase, as the initiation diameter increases. It is noted that a better empennage formation can be approached when the initiation synchronization errors at the initiation diameters of 30, 40, 50 mm are less than 50, 100, 150 ns, respectively. In addition, the lateral velocity of the EFP is lower and the flight stability is higher when the delay time of the middle detonation point is about half the maximum delay time of the three initiation points.
2018, 38(3): 509-516.
doi: 10.11883/bzycj-2017-0251
Abstract:
Due to its wide shock frequency and high amplitude, an explosion shock signal is commonly characterized by the phenomenon of signal zero drift in its measuring shock tests. In this study, we examined the origin of this zero drift phenomenon, compared the advantages and disadvantages of various methods used to rectify it, and proposed an improved version of the empirical mode decomposition (EMD) method with the window function filter in the frequency domain. The result shows that our improved EMD method can make up for the defects suffered by the current methods and provide data for explosion shock signal processing with low distortion.
Due to its wide shock frequency and high amplitude, an explosion shock signal is commonly characterized by the phenomenon of signal zero drift in its measuring shock tests. In this study, we examined the origin of this zero drift phenomenon, compared the advantages and disadvantages of various methods used to rectify it, and proposed an improved version of the empirical mode decomposition (EMD) method with the window function filter in the frequency domain. The result shows that our improved EMD method can make up for the defects suffered by the current methods and provide data for explosion shock signal processing with low distortion.
2018, 38(3): 517-524.
doi: 10.11883/bzycj-2017-0361
Abstract:
This paper presents a bidirectional-load split Hopkinson compression bar (BSHCB), namely, adding other symmetrical incident wave on the basis of traditional split Hpkinson pressure bar (SHPB), the two incident waves would load the specimen synchronously and symmetrically. The data process equations are approached according to the one-dimensional stress wave theory. According to the numerical simulation analyses, it is concluded that the proposed data process equations can determine the engineering stress, engineering strain and engineering strain rate of the tested material in bidirectional-load split Hopkinson compression test. Moreover, compared with the traditional SHPB test, the specimens in bidirectional-load split Hopkinson compression test can reach constant stress state in shorter time, and the strain rate can be improved.
This paper presents a bidirectional-load split Hopkinson compression bar (BSHCB), namely, adding other symmetrical incident wave on the basis of traditional split Hpkinson pressure bar (SHPB), the two incident waves would load the specimen synchronously and symmetrically. The data process equations are approached according to the one-dimensional stress wave theory. According to the numerical simulation analyses, it is concluded that the proposed data process equations can determine the engineering stress, engineering strain and engineering strain rate of the tested material in bidirectional-load split Hopkinson compression test. Moreover, compared with the traditional SHPB test, the specimens in bidirectional-load split Hopkinson compression test can reach constant stress state in shorter time, and the strain rate can be improved.
2018, 38(3): 525-533.
doi: 10.11883/bzycj-2016-0269
Abstract:
Early-stage deformation of water drops under Weber numbers ranging from 2 100 to 2 700 is investigated by experimental, numerical and theoretical methods, to reveal the influences of primary flow parameters on drop deformation as well as the mechanism behind them. Images of the drop deformation with noteworthy differences under different test conditions are captured with high-speed photography technique, demonstrating that though the Weber numbers are similar, drop deformation can be largely affected by the involved primary flow parameters, such as gas velocity, gas density and drop diameter. By substituting the liquid drop with a rigid sphere body, the gas flow field is numerically simulated, and the aerodynamic forces acting on sphere surface are distilled based on which the drop deformation is theoretically computed. The results show a good agreement between the theoretical and experimental deformation trends. The early-stage deformation of the drop is found to be in coherence with the flow separation and vortex distribution characteristics of the gas flow. Evolution of the gas flow field can be divided into a transient separation developing period and a following globally steady period. The pressure distribution exerted by the gas flow and the radial acceleration induced by it exhibit large differences in the two periods. The characteristic time of the separation development relative to the drop deformation, which can be represented by the square root of gas-liquid density ratio, is found to be a dominant parameter determining the drop deformation pattern in early stage of aero-breakup. A higher gas density leads to a higher occupation of the separation developing period in the whole drop deformation process, and the drop tends to develop a single ridge on its rear surface; on the contrary, multiple ridges with similar amplitude are more likely to happen when the gas density is lower, reflecting the characteristics of the outer flow in the globally stable period.
Early-stage deformation of water drops under Weber numbers ranging from 2 100 to 2 700 is investigated by experimental, numerical and theoretical methods, to reveal the influences of primary flow parameters on drop deformation as well as the mechanism behind them. Images of the drop deformation with noteworthy differences under different test conditions are captured with high-speed photography technique, demonstrating that though the Weber numbers are similar, drop deformation can be largely affected by the involved primary flow parameters, such as gas velocity, gas density and drop diameter. By substituting the liquid drop with a rigid sphere body, the gas flow field is numerically simulated, and the aerodynamic forces acting on sphere surface are distilled based on which the drop deformation is theoretically computed. The results show a good agreement between the theoretical and experimental deformation trends. The early-stage deformation of the drop is found to be in coherence with the flow separation and vortex distribution characteristics of the gas flow. Evolution of the gas flow field can be divided into a transient separation developing period and a following globally steady period. The pressure distribution exerted by the gas flow and the radial acceleration induced by it exhibit large differences in the two periods. The characteristic time of the separation development relative to the drop deformation, which can be represented by the square root of gas-liquid density ratio, is found to be a dominant parameter determining the drop deformation pattern in early stage of aero-breakup. A higher gas density leads to a higher occupation of the separation developing period in the whole drop deformation process, and the drop tends to develop a single ridge on its rear surface; on the contrary, multiple ridges with similar amplitude are more likely to happen when the gas density is lower, reflecting the characteristics of the outer flow in the globally stable period.
2018, 38(3): 534-540.
doi: 10.11883/bzycj-2016-0291
Abstract:
In the present work we investigated the cook-off and combustion explosion properties of 76 L fuel tanks with -10# diesel, fire-resistant diesel (FRD), fire-resistant-explosion-suppression diesel (FED) by performing experiments on the surface temperatures of the fire, the sizes of the jet fire, the internal temperatures of fuel tanks and the evaporation rate of the diesel using a video-camera, a high-speed camera, an infrared thermal imager, a thermocouples and an electronic balance. The results show that explosion might occur when the fuel tanks were filled with -10# diesel and a jet fire would occur when the fuel tanks were filled with FRD or FED in the same cook-off conditions. The highest surface temperatures, the height of the jet fire, the vapor temperature and the fuel temperature of FRD were reduced by 31.39%, 75.34%, 39.05% and 57.32% respectively as compared with -10# diesel. The highest surface temperatures, the height of the jetfire, the evaporation rate of the diesel, the vapor temperature and the fuel temperature of FED were reduced by 24.67%, 61.11%, 14.29%, 7.54% and 7.54% respectively as compared with FRD, indicating that FED was more effective than the other diesel in preventing temperature increase of the jet fire, size growth of the fireball and reducinge the diesel's evaporation rate.
In the present work we investigated the cook-off and combustion explosion properties of 76 L fuel tanks with -10# diesel, fire-resistant diesel (FRD), fire-resistant-explosion-suppression diesel (FED) by performing experiments on the surface temperatures of the fire, the sizes of the jet fire, the internal temperatures of fuel tanks and the evaporation rate of the diesel using a video-camera, a high-speed camera, an infrared thermal imager, a thermocouples and an electronic balance. The results show that explosion might occur when the fuel tanks were filled with -10# diesel and a jet fire would occur when the fuel tanks were filled with FRD or FED in the same cook-off conditions. The highest surface temperatures, the height of the jet fire, the vapor temperature and the fuel temperature of FRD were reduced by 31.39%, 75.34%, 39.05% and 57.32% respectively as compared with -10# diesel. The highest surface temperatures, the height of the jetfire, the evaporation rate of the diesel, the vapor temperature and the fuel temperature of FED were reduced by 24.67%, 61.11%, 14.29%, 7.54% and 7.54% respectively as compared with FRD, indicating that FED was more effective than the other diesel in preventing temperature increase of the jet fire, size growth of the fireball and reducinge the diesel's evaporation rate.
2018, 38(3): 541-548.
doi: 10.11883/bzycj-2016-0262
Abstract:
In order to simulate the thermal ignition process of gasoline-air in continuous hot wall, the chemical kinetic model, hydrodynamic model and radiation heat transfer model were coupled to establish a unified model of gasoline-airthermal ignition. According to the working condition of the experiment, the occurrence process of gasoline-air thermal ignition was simulated under the conditions of high temperature induced by continuous hot wall in confined space. Flow field evolution characteristics of the temperature and the pressure were analyzed. The variation curves of temperature, pressure, flow velocity, turbulent velocity and group concentration were obtained at different positions. By simulation, it is found that there are three stages in the process of gasoline-air thermal ignition, namely, the initial heating stage, the heating intermediate stage and the thermal ignition stage. The main reason for the existence of different stages is that the leading roles of chemical reaction and flow are different.
In order to simulate the thermal ignition process of gasoline-air in continuous hot wall, the chemical kinetic model, hydrodynamic model and radiation heat transfer model were coupled to establish a unified model of gasoline-airthermal ignition. According to the working condition of the experiment, the occurrence process of gasoline-air thermal ignition was simulated under the conditions of high temperature induced by continuous hot wall in confined space. Flow field evolution characteristics of the temperature and the pressure were analyzed. The variation curves of temperature, pressure, flow velocity, turbulent velocity and group concentration were obtained at different positions. By simulation, it is found that there are three stages in the process of gasoline-air thermal ignition, namely, the initial heating stage, the heating intermediate stage and the thermal ignition stage. The main reason for the existence of different stages is that the leading roles of chemical reaction and flow are different.
2018, 38(3): 549-554.
doi: 10.11883/bzycj-2016-0318
Abstract:
In this study we investigated the property of the shock wave propagating into the LiF window using the pole-curve method of two-dimensional unsteady flow theory and the hydrokinetical software LS-DYNA. The result derived from the pole-curve method shows that, when the one dimensional planar shock wave moves from the wedge-shaped explosive to the LiF window, the moving direction of the particles changes with the angle of about 3.3°, whereas the change angle actually obtained by the LS-DYNA simulation is 2.77°~3.03°. This disparity may be due to the fact that the rarefaction wave is ignored in the pole-curve method.
In this study we investigated the property of the shock wave propagating into the LiF window using the pole-curve method of two-dimensional unsteady flow theory and the hydrokinetical software LS-DYNA. The result derived from the pole-curve method shows that, when the one dimensional planar shock wave moves from the wedge-shaped explosive to the LiF window, the moving direction of the particles changes with the angle of about 3.3°, whereas the change angle actually obtained by the LS-DYNA simulation is 2.77°~3.03°. This disparity may be due to the fact that the rarefaction wave is ignored in the pole-curve method.
2018, 38(3): 555-562.
doi: 10.11883/bzycj-2016-0149
Abstract:
In this paper we studied the scattering of the transmission SH-wave form by elliptic elastic inclusions in a bi-material half-space using mainly the Green function, the complex function, the conformal mapping and the polar coordinate. Firstly, by introducing the complex variables and using the conformal mapping, we mapped the elliptic boundary into a unit circle. Next, by dividing the double-phase medium along the vertical boundaries into two quarter-spaces and adding a force system on the subdivision surface to make the SH wave satisfy the displacement and stress continuity conditions in the vertical boundary, we constructed a quarter-space Green function of displacement field. Then, we established an additional force system of integro differential equations in the vertical interface using the "conjunction" technique, and solved the unknown additional force system using the effective truncation. Finally, we presented and discussed the distribution of the dynamic stress concentration factor around the elliptic inclusion under the conditions of different parameters. The results show that the SH-wave incident angle and frequency, and the media both have influences on the distribution of the dynamic stress concentration factor.
In this paper we studied the scattering of the transmission SH-wave form by elliptic elastic inclusions in a bi-material half-space using mainly the Green function, the complex function, the conformal mapping and the polar coordinate. Firstly, by introducing the complex variables and using the conformal mapping, we mapped the elliptic boundary into a unit circle. Next, by dividing the double-phase medium along the vertical boundaries into two quarter-spaces and adding a force system on the subdivision surface to make the SH wave satisfy the displacement and stress continuity conditions in the vertical boundary, we constructed a quarter-space Green function of displacement field. Then, we established an additional force system of integro differential equations in the vertical interface using the "conjunction" technique, and solved the unknown additional force system using the effective truncation. Finally, we presented and discussed the distribution of the dynamic stress concentration factor around the elliptic inclusion under the conditions of different parameters. The results show that the SH-wave incident angle and frequency, and the media both have influences on the distribution of the dynamic stress concentration factor.
2018, 38(3): 563-571.
doi: 10.11883/bzycj-2016-0275
Abstract:
A numerical model of bedding bench slope was established by using the FLAC3D software. First, the dynamic response law of the slope under the blasting waves was analyzed. Then the stability of the slope was analyzed based on the displacement and the shear strain increment. Finally, according to the judgment of the slope stability, the appropriate velocity safety threshold of blasting vibration was developed. Studies reveal that the attenuation rate of the vibration amplitude decreases with higher blaster center distance value. There is elevation amplification effect on the slope and the main vibration velocity reaches the maximum at the foot of the free surface due to the existence of the weak interlayer. The deformation and failure of the slope is controlled by the weak interlayer. The upper rock mass becomes the potential sliding mass when there is tensile and shear failure in the weak interlayer, and the fracture surface can be determined according to the horizontal displacement map and the plastic zone distribution map. The slope failure is a progressive accumulation, displacement and shear strain accumulation will lead to the mechanical parameters of the rock weakening. There is a permanent displacement after blasting vibration if the slope has a great safety margin, while the slope near the state of limit equilibrium will be instability. The safety threshold of the slope is 21 cm/s if the strata inclination is from 15° to 23°. When the cut out height of the slope with the weak interlayer is 14 m and the strata inclinations are 24°, 29°, 31° and 34°, the corresponding safety threshold will be 10, 8, 6 and 5 cm/s.
A numerical model of bedding bench slope was established by using the FLAC3D software. First, the dynamic response law of the slope under the blasting waves was analyzed. Then the stability of the slope was analyzed based on the displacement and the shear strain increment. Finally, according to the judgment of the slope stability, the appropriate velocity safety threshold of blasting vibration was developed. Studies reveal that the attenuation rate of the vibration amplitude decreases with higher blaster center distance value. There is elevation amplification effect on the slope and the main vibration velocity reaches the maximum at the foot of the free surface due to the existence of the weak interlayer. The deformation and failure of the slope is controlled by the weak interlayer. The upper rock mass becomes the potential sliding mass when there is tensile and shear failure in the weak interlayer, and the fracture surface can be determined according to the horizontal displacement map and the plastic zone distribution map. The slope failure is a progressive accumulation, displacement and shear strain accumulation will lead to the mechanical parameters of the rock weakening. There is a permanent displacement after blasting vibration if the slope has a great safety margin, while the slope near the state of limit equilibrium will be instability. The safety threshold of the slope is 21 cm/s if the strata inclination is from 15° to 23°. When the cut out height of the slope with the weak interlayer is 14 m and the strata inclinations are 24°, 29°, 31° and 34°, the corresponding safety threshold will be 10, 8, 6 and 5 cm/s.
2018, 38(3): 572-578.
doi: 10.11883/bzycj-2016-0280
Abstract:
In order to achieve traceable dynamic pressure measurement in the liquid pulse pressure calibration, a dynamic pressure measurement method by laser interferometry based on Newton's second law is used. The pulse pressure is generated by a drop mass' impact on the piston in the liquid cavity to calibrate the amplitude sensitivity of the piezoelectric sensor. The amplitude of the pulse pressure is calculated from the mass and the mass' acceleration measured by the laser interferometer. Theoretical and experimental analysis of the pressure distribution in the pressure cavity, the acceleration distribution in the mass and the frictions are done, and then the measurement uncertainty of the piezoelectric pressure sensor's amplitude sensitivity calibration is analyzed. The measurement range of the calibration device is (10-500) MPa and the extend uncertainty is less than 1.8%.
In order to achieve traceable dynamic pressure measurement in the liquid pulse pressure calibration, a dynamic pressure measurement method by laser interferometry based on Newton's second law is used. The pulse pressure is generated by a drop mass' impact on the piston in the liquid cavity to calibrate the amplitude sensitivity of the piezoelectric sensor. The amplitude of the pulse pressure is calculated from the mass and the mass' acceleration measured by the laser interferometer. Theoretical and experimental analysis of the pressure distribution in the pressure cavity, the acceleration distribution in the mass and the frictions are done, and then the measurement uncertainty of the piezoelectric pressure sensor's amplitude sensitivity calibration is analyzed. The measurement range of the calibration device is (10-500) MPa and the extend uncertainty is less than 1.8%.
2018, 38(3): 579-585.
doi: 10.11883/bzycj-2016-0245
Abstract:
In this paper, we measured the blast overpressures of TNT, SEFAE and DEFAE using a pressure measuring system and pressure transducers installed in the free field and on the ground. The data obtained were analyzed and fitted with the blast overpressure development formulas for three explosives. The results showed that the average TNT equivalence of DEFAE in the free field and on the ground were 60% and 69% bigger than that of SEFAE respectively, suggesting that the power of DEFAE was bigger than that of SEFAE. For each of the three explosives the overpressure measured was 10% lower than that on the ground due to the reflection wave generated by the ground. In the actual test, a fixed correction factor needs to be adopted. These results may add to analysis of small quantity of FAE and traditional explosives and serve as reference for the investigations of the blasting power of FAE.
In this paper, we measured the blast overpressures of TNT, SEFAE and DEFAE using a pressure measuring system and pressure transducers installed in the free field and on the ground. The data obtained were analyzed and fitted with the blast overpressure development formulas for three explosives. The results showed that the average TNT equivalence of DEFAE in the free field and on the ground were 60% and 69% bigger than that of SEFAE respectively, suggesting that the power of DEFAE was bigger than that of SEFAE. For each of the three explosives the overpressure measured was 10% lower than that on the ground due to the reflection wave generated by the ground. In the actual test, a fixed correction factor needs to be adopted. These results may add to analysis of small quantity of FAE and traditional explosives and serve as reference for the investigations of the blasting power of FAE.
2018, 38(3): 586-595.
doi: 10.11883/bzycj-2016-0266
Abstract:
In order to understand the deformation behavior and microstructure evolution of a solution treated AM80 magnesium alloy under quasi-static and impact loadings, quasi-static and high-speed impact compression tests at room temperature were performed by an Instron universal compression machine and a slip Hopkinson pressure bar apparatus, respectively. Under quasi-static loadings, the flow stress of the AM80 magnesium alloy decreases gradually with the increase of the strain rate (3×10-5 s-1 ≤\begin{document}$\dot \varepsilon $\end{document} \begin{document}$\dot \varepsilon $\end{document}
In order to understand the deformation behavior and microstructure evolution of a solution treated AM80 magnesium alloy under quasi-static and impact loadings, quasi-static and high-speed impact compression tests at room temperature were performed by an Instron universal compression machine and a slip Hopkinson pressure bar apparatus, respectively. Under quasi-static loadings, the flow stress of the AM80 magnesium alloy decreases gradually with the increase of the strain rate (3×10-5 s-1 ≤
2018, 38(3): 596-602.
doi: 10.11883/bzycj-2016-0286
Abstract:
In this work we conducted a quasi-static tensile test, a high temperature tensile test and a dynamic tensile test on Q235B steel, the most widely used in steel structures in China, using a multi-functional material testing machine and a split Hopkinson tension bar (SHTB) and, based on the test data obtained, fitted three frequently used material models, i.e. the Cowper-Symonds model, the Johnson-Cook model and the Zerilli-Armstrong model, in LS-DYNA. We then verified their validity by conducting Taylor impact tests. The results showed that Q235B steel was temperature and strain-rate sensitive, that the Cowper-Symonds model was applicable in low velocity impact simulations, that the Johnson-Cook model was suitable for simulations with a wider range of strain-rates, and that the Zerilli-Armstrong model was not recommendable for low velocity impact simulation.
In this work we conducted a quasi-static tensile test, a high temperature tensile test and a dynamic tensile test on Q235B steel, the most widely used in steel structures in China, using a multi-functional material testing machine and a split Hopkinson tension bar (SHTB) and, based on the test data obtained, fitted three frequently used material models, i.e. the Cowper-Symonds model, the Johnson-Cook model and the Zerilli-Armstrong model, in LS-DYNA. We then verified their validity by conducting Taylor impact tests. The results showed that Q235B steel was temperature and strain-rate sensitive, that the Cowper-Symonds model was applicable in low velocity impact simulations, that the Johnson-Cook model was suitable for simulations with a wider range of strain-rates, and that the Zerilli-Armstrong model was not recommendable for low velocity impact simulation.
2018, 38(3): 603-615.
doi: 10.11883/bzycj-2016-0265
Abstract:
Based on the nonlinear finite element software LS-DYNA and smoothed particle hydrodynamics (SPH) method, a computational model of high-speed railway axles subjected to ballast impact was built. Influences of impact velocity, ballast shape and size, and impact angle on the dynamic response of axles were examined. The response characteristics of the impact force and the deformation at the impact point were given, and the relationship between the peak impact force and the maximum residual deformation of the axle was also analyzed. Finally, the impact damage regularity of the axle under different conditions was also explored. Numerical results show that both the peak impact force and the deformation of the axle, including both transient and residual deformation, increase with the increase of initial impact velocity, ballast size and impact angle, respectively, and an approximately linear increase between the maximum residual deformation and peak impact force is found. Besides, the normalized indentation depth (residual deformation) of axles increases linearly with the square root of impact energy.
Based on the nonlinear finite element software LS-DYNA and smoothed particle hydrodynamics (SPH) method, a computational model of high-speed railway axles subjected to ballast impact was built. Influences of impact velocity, ballast shape and size, and impact angle on the dynamic response of axles were examined. The response characteristics of the impact force and the deformation at the impact point were given, and the relationship between the peak impact force and the maximum residual deformation of the axle was also analyzed. Finally, the impact damage regularity of the axle under different conditions was also explored. Numerical results show that both the peak impact force and the deformation of the axle, including both transient and residual deformation, increase with the increase of initial impact velocity, ballast size and impact angle, respectively, and an approximately linear increase between the maximum residual deformation and peak impact force is found. Besides, the normalized indentation depth (residual deformation) of axles increases linearly with the square root of impact energy.
2018, 38(3): 616-621.
doi: 10.11883/bzycj-2016-0263
Abstract:
In order to further improve the damage efficiency of a multiple explosively-formed penetrator warhead, the multiple explosively-formed rod-like penetrator warheads were designed by composite charge and were researched by numerical simulation. Based on the detonation loading control method by composite charge, the liners can be formed to be dense rod-like penetrators. By adjusting the oblique angle of a half-prefabricated liner, the rotation speed of the penetrator is controlled and the stability of its flight attitude in the air is controlled. And then, the damage power of the penetrator was improved. The principle prototypes of the warheads with different oblique notching angles were designed and the static blasting experiments were performed. The comparison between the experimental results and the numerical simulation ones displays that the impact attitude of the explosively-formed rod-like penetrator is the best and its penetration into the steel-45 plate is the deepest, when the oblique angle is 1.5°. At the base of ensuring the formed quality, the penetration power of the rod-like penetrator can be improved significantly by obliquely placing the liner.
In order to further improve the damage efficiency of a multiple explosively-formed penetrator warhead, the multiple explosively-formed rod-like penetrator warheads were designed by composite charge and were researched by numerical simulation. Based on the detonation loading control method by composite charge, the liners can be formed to be dense rod-like penetrators. By adjusting the oblique angle of a half-prefabricated liner, the rotation speed of the penetrator is controlled and the stability of its flight attitude in the air is controlled. And then, the damage power of the penetrator was improved. The principle prototypes of the warheads with different oblique notching angles were designed and the static blasting experiments were performed. The comparison between the experimental results and the numerical simulation ones displays that the impact attitude of the explosively-formed rod-like penetrator is the best and its penetration into the steel-45 plate is the deepest, when the oblique angle is 1.5°. At the base of ensuring the formed quality, the penetration power of the rod-like penetrator can be improved significantly by obliquely placing the liner.
2018, 38(3): 622-631.
doi: 10.11883/bzycj-2016-0309
Abstract:
Numerical simulation was carried out by applying the fluid dynamics software Fluent to explore the influences of different ignition conditions, such as ignition locations (the distances from the left wall of the closed pipe are 100, 200, and 500 mm, respectively), ignition temperatures (1 000, 1 500 and 2 000 K) and ignition area (ignition radius:50, 35 and 20 mm) on the deflagration characteristics of the premixed H2/air mixture in a closed pipe with 1 000 mm in length. The results show that, when the ignition positions are far away from the left wall of the closed pipe, the temperature of the intermediate node in the flow field is higher and the temperature rising is faster in the closed pipe. The rising rates of the maximum temperatures are basically synchronous on the conditions of the three different ignition temperatures (1 000, 1 500 and 2 000 K). Meanwhile, the combustion reaction of H2/air is more intense with the increasing of the ignition temperatures. The temperature rising rate in the closed pipe is accelerated. However, the peak pressure in the closed pipe is reduced. Moreover, the smaller the ignition area, the faster the temperature rising of H2/air in the early stage. When the radius of the ignition area is 35 mm and the ignition position away from the left side wall of the closed pipe is 100 mm, the deflagration parameters of H2/air are relatively higher. The influence of different ignition conditions on the kinetic energy and internal energy is similar to the influence of different ignition conditions on the velocity and temperature of the premixed gas, but the ignition conditions hardly influence the vorticity.
Numerical simulation was carried out by applying the fluid dynamics software Fluent to explore the influences of different ignition conditions, such as ignition locations (the distances from the left wall of the closed pipe are 100, 200, and 500 mm, respectively), ignition temperatures (1 000, 1 500 and 2 000 K) and ignition area (ignition radius:50, 35 and 20 mm) on the deflagration characteristics of the premixed H2/air mixture in a closed pipe with 1 000 mm in length. The results show that, when the ignition positions are far away from the left wall of the closed pipe, the temperature of the intermediate node in the flow field is higher and the temperature rising is faster in the closed pipe. The rising rates of the maximum temperatures are basically synchronous on the conditions of the three different ignition temperatures (1 000, 1 500 and 2 000 K). Meanwhile, the combustion reaction of H2/air is more intense with the increasing of the ignition temperatures. The temperature rising rate in the closed pipe is accelerated. However, the peak pressure in the closed pipe is reduced. Moreover, the smaller the ignition area, the faster the temperature rising of H2/air in the early stage. When the radius of the ignition area is 35 mm and the ignition position away from the left side wall of the closed pipe is 100 mm, the deflagration parameters of H2/air are relatively higher. The influence of different ignition conditions on the kinetic energy and internal energy is similar to the influence of different ignition conditions on the velocity and temperature of the premixed gas, but the ignition conditions hardly influence the vorticity.
2018, 38(3): 632-638.
doi: 10.11883/bzycj-2016-0293
Abstract:
In the present study we carried out sympathetic detonation experiments, using RDX-8701 as donor and booster pipe and using JO-9C charge as acceptor explosive, to study the response of the fuse explosion device under shock wave. By observing the remainder of the JO-9C explosive, the deformation of the witness and the steel shell, and judging the explosive reaction state, we analyzed the histories in the sympathetic detonation reaction of the acceptor, and established the calculation model of the sympathetic detonation using AUTODYN. The calculated results of the model included the effects of the shock waves produced by the donor that acted on the booster pipe. Based on the fluid-solid coupling method, we also obtained the critical and safety distance of the sympathetic detonation through numerical simulation by adjusting experimental distance. The results show that the detonation wave propagated down firstly along the slope direction of the booster charge, thereby leading to the detonation of the booster charge, and then detonated the detonator explosive. According to the numerical simulation results, the critical and the safety distances of the sympathetic detonation were 5.7 mm and 8.8 mm respectively.
In the present study we carried out sympathetic detonation experiments, using RDX-8701 as donor and booster pipe and using JO-9C charge as acceptor explosive, to study the response of the fuse explosion device under shock wave. By observing the remainder of the JO-9C explosive, the deformation of the witness and the steel shell, and judging the explosive reaction state, we analyzed the histories in the sympathetic detonation reaction of the acceptor, and established the calculation model of the sympathetic detonation using AUTODYN. The calculated results of the model included the effects of the shock waves produced by the donor that acted on the booster pipe. Based on the fluid-solid coupling method, we also obtained the critical and safety distance of the sympathetic detonation through numerical simulation by adjusting experimental distance. The results show that the detonation wave propagated down firstly along the slope direction of the booster charge, thereby leading to the detonation of the booster charge, and then detonated the detonator explosive. According to the numerical simulation results, the critical and the safety distances of the sympathetic detonation were 5.7 mm and 8.8 mm respectively.
2018, 38(3): 639-646.
doi: 10.11883/bzycj-2016-0295
Abstract:
Firstly, the distribution of load and displacement of the free surface of the arch were obtained by using the geometrical relation and the empirical formula, and the related parameters were discussed. Secondly, the modified soil-structure interaction model was used to solve the vibration equation of the arch orthogonally, and get the analytic solution of the elastic dynamic response of the structure under arbitrary angle loading. Then the time histories curve of displacement, velocity and acceleration were predicted. The analysis results indicate that the responses of the arch are greatly influenced by the acoustic impedance. The larger the acoustic impedance, the greater the displacement, velocity and acceleration of the structure. So the protective structures are better to be constructed in a site with low acoustic impedance.
Firstly, the distribution of load and displacement of the free surface of the arch were obtained by using the geometrical relation and the empirical formula, and the related parameters were discussed. Secondly, the modified soil-structure interaction model was used to solve the vibration equation of the arch orthogonally, and get the analytic solution of the elastic dynamic response of the structure under arbitrary angle loading. Then the time histories curve of displacement, velocity and acceleration were predicted. The analysis results indicate that the responses of the arch are greatly influenced by the acoustic impedance. The larger the acoustic impedance, the greater the displacement, velocity and acceleration of the structure. So the protective structures are better to be constructed in a site with low acoustic impedance.
2018, 38(3): 647-653.
doi: 10.11883/bzycj-2016-0289
Abstract:
In this study we performed a series of penetration tests on specimens of carbon fiber reinforced plastics (CFRPs) using a one-stage gas gun at impact angles of 0°, 30° and 45° at speeds ranging from 70 to 280 m/s to investigate their ballistic resistance behavior in oblique penetration. High speed photography was employed to measure the projectile velocity and ballistic trajectory, and analyzed the influence of the impact angle on their energy absorption, ballistic limit and projectile obliquity variations. The results show that the energy absorption coefficient of the normal impact outperformed the oblique impact at lower impact-energies whereas at the higher impact-energies the ballistic performance was observed to be approximately just the opposite. In addition, due to the increase of the penetration length through the laminate with the impact angle, the ballistic limit increases with the impact angle; the influence of the impact angle on the projectile obliquity varies generally with the impact velocity.
In this study we performed a series of penetration tests on specimens of carbon fiber reinforced plastics (CFRPs) using a one-stage gas gun at impact angles of 0°, 30° and 45° at speeds ranging from 70 to 280 m/s to investigate their ballistic resistance behavior in oblique penetration. High speed photography was employed to measure the projectile velocity and ballistic trajectory, and analyzed the influence of the impact angle on their energy absorption, ballistic limit and projectile obliquity variations. The results show that the energy absorption coefficient of the normal impact outperformed the oblique impact at lower impact-energies whereas at the higher impact-energies the ballistic performance was observed to be approximately just the opposite. In addition, due to the increase of the penetration length through the laminate with the impact angle, the ballistic limit increases with the impact angle; the influence of the impact angle on the projectile obliquity varies generally with the impact velocity.
2018, 38(3): 654-658.
doi: 10.11883/bzycj-2016-0308
Abstract:
In this paper we proposed a method for modifying classical similarity theories to meet the guidance need in the impact experiment of strain rate sensitive materials. This method, based on recent dynamic similarity theories, took into account the effect of the initial load's similarity coefficients on the strain-rate effect during the impact experiment due to the differences in materials between the prototype and the model. We rebuilt the calculation formulas of the initial load's similarity coefficients using the Cowper-Symonds equation, and discussed the results obtained in the impact experiment on two scaled models, made from two different materials, as to whether they were strain-rate sensitive or not, showing that, compared with the classical similarity theories, our newly proposed method could provide the initial load's accurate similarity coefficients and thus improve the model prediction precision.
In this paper we proposed a method for modifying classical similarity theories to meet the guidance need in the impact experiment of strain rate sensitive materials. This method, based on recent dynamic similarity theories, took into account the effect of the initial load's similarity coefficients on the strain-rate effect during the impact experiment due to the differences in materials between the prototype and the model. We rebuilt the calculation formulas of the initial load's similarity coefficients using the Cowper-Symonds equation, and discussed the results obtained in the impact experiment on two scaled models, made from two different materials, as to whether they were strain-rate sensitive or not, showing that, compared with the classical similarity theories, our newly proposed method could provide the initial load's accurate similarity coefficients and thus improve the model prediction precision.
2018, 38(3): 659-664.
doi: 10.11883/bzycj-2016-0279
Abstract:
This paper presents a study on the multi-spall of ductile metal induced by triangular impulse, using the simple cut-off pressure model and Tuler-Batcher's accumulative damage model to describe the spall. Numerical results show that the spall thickness increases with increasing ratio of pulse length to strength of shock wave and the destroy depth of the target approximates to half of pulse width under extreme loading.
This paper presents a study on the multi-spall of ductile metal induced by triangular impulse, using the simple cut-off pressure model and Tuler-Batcher's accumulative damage model to describe the spall. Numerical results show that the spall thickness increases with increasing ratio of pulse length to strength of shock wave and the destroy depth of the target approximates to half of pulse width under extreme loading.
2018, 38(3): 665-670.
doi: 10.11883/bzycj-2016-0310
Abstract:
The intact and pre-existing flaw granite specimens were tested by using the split Hopkinson pressure bar device with 50 mm rod diameter, and the whole process of spalling fracture was recorded by a high-speed video camera. Based on the theoretical analysis, the possible initial spalling positions of the granite specimens under half sine wave loading were calculated. The results of high-speed photography show that the pre-existing flaws have certain influence on the initial spalling positions, and spalling fractures generally occur at pre-existing flaws. According to the PFC2D (particle flow code in two dimensions) model, it is found that the initial spalling positions are influenced by the pre-existing flaws and the rising edge of the reflected tensile wave. The longer the rising edge of the reflection wave, the more likely the initial spalling occurs at the pre-existing flaw. The shorter the rising edge of the reflected tensile wave, the less likely the initial spalling occurs at the pre-existing flaw.
The intact and pre-existing flaw granite specimens were tested by using the split Hopkinson pressure bar device with 50 mm rod diameter, and the whole process of spalling fracture was recorded by a high-speed video camera. Based on the theoretical analysis, the possible initial spalling positions of the granite specimens under half sine wave loading were calculated. The results of high-speed photography show that the pre-existing flaws have certain influence on the initial spalling positions, and spalling fractures generally occur at pre-existing flaws. According to the PFC2D (particle flow code in two dimensions) model, it is found that the initial spalling positions are influenced by the pre-existing flaws and the rising edge of the reflected tensile wave. The longer the rising edge of the reflection wave, the more likely the initial spalling occurs at the pre-existing flaw. The shorter the rising edge of the reflected tensile wave, the less likely the initial spalling occurs at the pre-existing flaw.
2018, 38(3): 671-676.
doi: 10.11883/bzycj-2016-0300
Abstract:
In this study, based on ANSYS/LS-DYNA, we described the tempered glass using the Lagrange method, reasonably simplified the blast wave load by simulating the failure of the tempered glass using the erosion algorithm, with the maximum elongation line strain theory used as the failure criterion for the monolithic tempered glass. We then simulated the monolithic tempered glass under the blast wave using model thus established. Through a large number of debugging and calculations, we identified the strain threshold of the failure of the monolithic tempered glass as 0.002 2. The numerical results were verified by the experimental results and proved as reasonable.
In this study, based on ANSYS/LS-DYNA, we described the tempered glass using the Lagrange method, reasonably simplified the blast wave load by simulating the failure of the tempered glass using the erosion algorithm, with the maximum elongation line strain theory used as the failure criterion for the monolithic tempered glass. We then simulated the monolithic tempered glass under the blast wave using model thus established. Through a large number of debugging and calculations, we identified the strain threshold of the failure of the monolithic tempered glass as 0.002 2. The numerical results were verified by the experimental results and proved as reasonable.
2018, 38(3): 677-682.
doi: 10.11883/bzycj-2016-0282
Abstract:
In the measurement of a projectile's acceleration history using an acceleration transducer, the results achieved contain the acceleration resulting from the hindrance of a target that decelerates the projectile and the acceleration resulting from the structural vibration of the missile. In this study, a simple model was established simplifying the projectile as a long straight round bar and analyses the frequency characteristics of the axial vibration of a projectile. A finite element model of the projectile was built using ANSYS to solve the natural frequency and mode of the projectile via modal analysis, then the harmonic response analysis about the projectile was conducted. The theoretical results and simulated results were quite consistent, both about 1 200 Hz. The frequency characteristics of the actual acceleration signal obtained from the target experiment were analyzed via the Fourier transform and wavelet analysis. The peak value of the signal power density spectrum was at 1 114 Hz, consistent with both the theoretical value and the simulated result.
In the measurement of a projectile's acceleration history using an acceleration transducer, the results achieved contain the acceleration resulting from the hindrance of a target that decelerates the projectile and the acceleration resulting from the structural vibration of the missile. In this study, a simple model was established simplifying the projectile as a long straight round bar and analyses the frequency characteristics of the axial vibration of a projectile. A finite element model of the projectile was built using ANSYS to solve the natural frequency and mode of the projectile via modal analysis, then the harmonic response analysis about the projectile was conducted. The theoretical results and simulated results were quite consistent, both about 1 200 Hz. The frequency characteristics of the actual acceleration signal obtained from the target experiment were analyzed via the Fourier transform and wavelet analysis. The peak value of the signal power density spectrum was at 1 114 Hz, consistent with both the theoretical value and the simulated result.
2018, 38(3): 683-687.
doi: 10.11883/bzycj-2016-0302
Abstract:
In the present study, based on the physical process of combined pulse fracturing, we proposed a model for the coupling of the whole process of multi-pulse conflagration fracturing using different combinations of the multi-level pulse propellant conflagration loading model, the pressurized liquid column movement model, the perforation discharge model, the fracture initiation and the fracture dynamic extension model. Furthermore, we analyzed the wellbore pressure changes and the fractures propagation under single and combined propellants at different ratios. The calculated results demonstrated that fast deflagration fracturing could be initiated using combinations of the first-phase propellants with different burning rates and the multi-directional wellbore pressure was reached at once; a longer burning time of the propellant might be held to maintain the high pressure for the full extension of the fracture, using the second-and the third-phase propellants; when the amount of the first-phase propellants and the total amount of the propellants were determined, the deflagration loading rate and the multi-directional fracture initiation pressure tended to decrease slightly as the lowest burning rate of the third-phrase propellant's proportion rose up, though not obviously; the eventual extension lengths of the fracture in different directions were, however, very sensitive to the proportion of the second-and third-phase propellants. These results improved our understanding of the coupling mechanism of the combined pulse fracturing, and might be used to optimize the mass ratio of propellants.
In the present study, based on the physical process of combined pulse fracturing, we proposed a model for the coupling of the whole process of multi-pulse conflagration fracturing using different combinations of the multi-level pulse propellant conflagration loading model, the pressurized liquid column movement model, the perforation discharge model, the fracture initiation and the fracture dynamic extension model. Furthermore, we analyzed the wellbore pressure changes and the fractures propagation under single and combined propellants at different ratios. The calculated results demonstrated that fast deflagration fracturing could be initiated using combinations of the first-phase propellants with different burning rates and the multi-directional wellbore pressure was reached at once; a longer burning time of the propellant might be held to maintain the high pressure for the full extension of the fracture, using the second-and the third-phase propellants; when the amount of the first-phase propellants and the total amount of the propellants were determined, the deflagration loading rate and the multi-directional fracture initiation pressure tended to decrease slightly as the lowest burning rate of the third-phrase propellant's proportion rose up, though not obviously; the eventual extension lengths of the fracture in different directions were, however, very sensitive to the proportion of the second-and third-phase propellants. These results improved our understanding of the coupling mechanism of the combined pulse fracturing, and might be used to optimize the mass ratio of propellants.
2018, 38(3): 688-695.
doi: 10.11883/bzycj-2016-0133
Abstract:
In this paper, we studied the program implosion process and the influence of the magnetic field under a high-energy density state on charged particles and the mechanism of the compression process, using the MHD equations and finite difference method. The results show that the implosion process parameters meet the following conditions:the electron temperature plasma (50 keV), the pressure (1 TPa), the particle number density (1024 cm-3). The liner material had a major influence on the confined time and ignition conditions; at the same time, the electronic thermal conductivity was reduced by more than 100 times in the absence of a magnetic field when the magnetic field was greater than 5 T, and the ion thermal conductivity also experienced a significant decrease in the compression peak. The alpha particle energy deposition density in the 5 T embedded magnetic field was relatively raised by more than about 200 times than in the 1 T embedded magnetic field. Also, it was found that magnetization hindered implosion compression process to a certain extent.
In this paper, we studied the program implosion process and the influence of the magnetic field under a high-energy density state on charged particles and the mechanism of the compression process, using the MHD equations and finite difference method. The results show that the implosion process parameters meet the following conditions:the electron temperature plasma (50 keV), the pressure (1 TPa), the particle number density (1024 cm-3). The liner material had a major influence on the confined time and ignition conditions; at the same time, the electronic thermal conductivity was reduced by more than 100 times in the absence of a magnetic field when the magnetic field was greater than 5 T, and the ion thermal conductivity also experienced a significant decrease in the compression peak. The alpha particle energy deposition density in the 5 T embedded magnetic field was relatively raised by more than about 200 times than in the 1 T embedded magnetic field. Also, it was found that magnetization hindered implosion compression process to a certain extent.
2018, 38(3): 696-706.
doi: 10.11883/bzycj-2016-0303
Abstract:
In this paper, various collision protection methods were presented to ameliorate naval ships' collision protective capability. Conventional collision protection methods may result in the increase of the ship's weight and reduction of its functional performance. Composite sandwich plates serve as a new approach to the design of the collision protection structure, and have become one of the hottest research topics worldwide. With naval ships' collision resistance as its academic background, this paper reviewed the recent advances in the areas of experimental methods, deformation damage mechanisms, energy absorption factors, analysis methods and offered suggestions of future research directions in composite sandwich plates.
In this paper, various collision protection methods were presented to ameliorate naval ships' collision protective capability. Conventional collision protection methods may result in the increase of the ship's weight and reduction of its functional performance. Composite sandwich plates serve as a new approach to the design of the collision protection structure, and have become one of the hottest research topics worldwide. With naval ships' collision resistance as its academic background, this paper reviewed the recent advances in the areas of experimental methods, deformation damage mechanisms, energy absorption factors, analysis methods and offered suggestions of future research directions in composite sandwich plates.
