Experiment and Numerical Simulation on Damage Behavior of Honeycomb Sandwich Composites under Low-Energy Impact

: It is well-established that the honeycomb sandwich composite structures are easily prone to damage under low-energy impact. Consequently, it would lead to a dramatic decrease in structural load-bearing capacity and a threat to overall safety. Both experimental and numerical simulations are carried out to investigate the impact damage behavior of honeycomb sandwich composite specimens. The damage mode, damage parameters, and contact force-time curves of three types of panel materials with T300, T700, and T800 are obtained under different impact energies of 10 J, 20 J, and 40 J by the drop-weight impact experiment. Moreover, digital image correlation (DIC) tests are used to measure the deformation and strain of the lower panel. The experimental results reveal that the degree of damage increases with increasing impact energy. Particularly, the T300 panel specimen exhibits visible ﬁber fracture when subjected to an impact energy of 40 J. The impact process involves matrix cracking, ﬁber fracture, and delamination of the upper panel occurring ﬁrst, followed by immediate crush damage to the honeycomb core and, ﬁnally, slight ﬁber damage to the lower panel. Due to its higher strength, the T800 panel specimen exhibits the highest damage resistance compared to the T700 and T300 panel specimens. To consider the microscopic failure criteria and various types of contact during the impact process, a ﬁnite element model of honeycomb sandwich composites is established, and numerical simulation analysis of low-energy impact is performed to determine the damage mode, damage size, and contact-force curves. Comparative analysis demonstrates good agreement between the simulation and experimental results. The ﬁndings of this study provide valuable technical support for the widespread application of honeycomb sandwich composites in the aviation ﬁeld.


Introduction
As a lightweight and high-strength material with exceptional specific rigidity and strength, honeycomb sandwich composites find extensive applications in the aerospace industry [1][2][3][4][5][6][7]. These applications include aircraft components such as rudders, flaps, elevators, wings/tail trailing edges, landing gear compartment doors, APU compartment doors, and radar covers, among others. Typically, a honeycomb sandwich composite consists of an upper panel, a lower panel, and a honeycomb core sandwiched between them. The adhesive film is used to bond these components together; the honeycomb sandwich composites are shown in Figure 1. The upper and lower panels primarily contribute to the overall structure's strength and stiffness. On the other hand, the honeycomb core bears It is important to note that the panels of honeycomb sandwich composite are thin and susceptible to impacts during usage, including runway debris, hail, and tool drops. These impacts can lead to local fiber breakage, thereby reducing the overall strength and service life of the structure [17]. When subjected to higher impact energy, the panel may experience penetration, exposing the core. The damping properties of the core can influence the adhesive strength, potentially resulting in large-scale failure between the panel and the core, leading to significant damage. Numerous failure accidents of sandwich structures have been attributed to impact damage.
Researchers have conducted a series of studies on the damage behavior of honeycomb sandwich composites under impact energy. The research on damage to sandwich composite materials primarily relies on experiment and simulation analysis. Baran and Weijermars [9] investigated the residual bending behavior of sandwich panels after impact, using bending tests to obtain shear cracks with progressive failure. Kolopp [18] conducted medium-velocity impact experiments on sandwich structures to enhance armor solutions for aeronautical applications. The research revealed that the combination of aluminum honeycomb core and aluminum skins yielded mitigated results. Lv [12] explored the low-velocity impact response of composite sandwich structures with grid-honeycomb hybrid cores through experimental and numerical simulation. The study demonstrated that the sandwich panels exhibited varying degrees of impact resistance at different locations. Zhang et al. [19] examined the damage response of sandwich panels with Nomex honeycomb cores under low-velocity impact conditions. They employed experiment and numerical simulation to establish a failure model based on the Hashin failure criterion and obtained insights into the intralaminar damage behavior of the laminate. The simulation results, including contact force histories, damage shapes, and indentation depths, were found to agree well with the experimental findings. Chen et al. [20] developed a Sandwich structures typically consist of panels made of either metal materials or composites [10][11][12]. Composite panels are commonly employed in the aerospace industry to achieve weight reduction. The core material used in sandwich structures usually includes options such as aluminum honeycomb, Nomex honeycomb, aluminum foam, and non-metallic foam [13][14][15][16]. Aramid paper-impregnated resin is often utilized for paper honeycomb cores due to its lightweight nature, non-flammability, and high specific strength. Consequently, it is the most widely employed core material in sandwich structures.
It is important to note that the panels of honeycomb sandwich composite are thin and susceptible to impacts during usage, including runway debris, hail, and tool drops. These impacts can lead to local fiber breakage, thereby reducing the overall strength and service life of the structure [17]. When subjected to higher impact energy, the panel may experience penetration, exposing the core. The damping properties of the core can influence the adhesive strength, potentially resulting in large-scale failure between the panel and the core, leading to significant damage. Numerous failure accidents of sandwich structures have been attributed to impact damage.
Researchers have conducted a series of studies on the damage behavior of honeycomb sandwich composites under impact energy. The research on damage to sandwich composite materials primarily relies on experiment and simulation analysis. Baran and Weijermars [9] investigated the residual bending behavior of sandwich panels after impact, using bending tests to obtain shear cracks with progressive failure. Kolopp [18] conducted medium-velocity impact experiments on sandwich structures to enhance armor solutions for aeronautical applications. The research revealed that the combination of aluminum honeycomb core and aluminum skins yielded mitigated results. Lv [12] explored the low-velocity impact response of composite sandwich structures with grid-honeycomb hybrid cores through experimental and numerical simulation. The study demonstrated that the sandwich panels exhibited varying degrees of impact resistance at different locations. Zhang et al. [19] examined the damage response of sandwich panels with Nomex honeycomb cores under low-velocity impact conditions. They employed experiment and numerical simulation to establish a failure model based on the Hashin failure criterion and obtained insights into the intralaminar damage behavior of the laminate. The simulation results, including contact force histories, damage shapes, and indentation depths, were found to agree well with the experimental findings. Chen et al. [20] developed a numerical model to investigate intra-laminar damage, inter-laminar adhesive delamination, and the strain rate effect of sandwich structures subjected to low-velocity impact. The numerical model successfully described the essential perforation mechanisms associated with damage patterns and exhibited accurate predictions of total energy absorption, among others. Zhang et al. [21] and Sun et al. [22] separately formulated simulation and theoretical models that could better predict the damage dent response. Feng et al. [23] proposed a progressive damage model based on continuum damage mechanics. They introduced cohesive interface elements to predict interlaminar damage and failure mechanisms in composites.
The low-energy impact damage experiment and finite element simulation analysis are conducted on honeycomb sandwich composites in this paper. The study aims to determine the damage mode and assess the damage resistance of these structures. The research findings will serve as valuable technical support for the widespread application of honeycomb sandwich structures in the aviation industry. Figure 2 illustrates the specimen of the honeycomb sandwich composite structure. The specimen consists of an upper panel, a lower panel, and a paper honeycomb core positioned in between. The panels and honeycomb core are bonded together using a high-strength adhesive film. The dimensions of the specimen are depicted in Figure 3. The length of the specimen measures 150 mm, the width is 100 mm, and the height of the honeycomb core is 19.5 mm.

Specimen
Aerospace 2023, 10, x FOR PEER REVIEW 3 of 21 numerical model to investigate intra-laminar damage, inter-laminar adhesive delamination, and the strain rate effect of sandwich structures subjected to low-velocity impact. The numerical model successfully described the essential perforation mechanisms associated with damage patterns and exhibited accurate predictions of total energy absorption, among others. Zhang et al. [21] and Sun et al. [22] separately formulated simulation and theoretical models that could better predict the damage dent response. Feng et al. [23] proposed a progressive damage model based on continuum damage mechanics. They introduced cohesive interface elements to predict interlaminar damage and failure mechanisms in composites. The low-energy impact damage experiment and finite element simulation analysis are conducted on honeycomb sandwich composites in this paper. The study aims to determine the damage mode and assess the damage resistance of these structures. The research findings will serve as valuable technical support for the widespread application of honeycomb sandwich structures in the aviation industry. Figure 2 illustrates the specimen of the honeycomb sandwich composite structure. The specimen consists of an upper panel, a lower panel, and a paper honeycomb core positioned in between. The panels and honeycomb core are bonded together using a highstrength adhesive film. The dimensions of the specimen are depicted in Figure 3. The length of the specimen measures 150 mm, the width is 100 mm, and the height of the honeycomb core is 19.5 mm.  numerical model to investigate intra-laminar damage, inter-laminar adhesive delamination, and the strain rate effect of sandwich structures subjected to low-velocity impact. The numerical model successfully described the essential perforation mechanisms associated with damage patterns and exhibited accurate predictions of total energy absorption, among others. Zhang et al. [21] and Sun et al. [22] separately formulated simulation and theoretical models that could better predict the damage dent response. Feng et al. [23] proposed a progressive damage model based on continuum damage mechanics. They introduced cohesive interface elements to predict interlaminar damage and failure mechanisms in composites. The low-energy impact damage experiment and finite element simulation analysis are conducted on honeycomb sandwich composites in this paper. The study aims to determine the damage mode and assess the damage resistance of these structures. The research findings will serve as valuable technical support for the widespread application of honeycomb sandwich structures in the aviation industry. Figure 2 illustrates the specimen of the honeycomb sandwich composite structure. The specimen consists of an upper panel, a lower panel, and a paper honeycomb core positioned in between. The panels and honeycomb core are bonded together using a highstrength adhesive film. The dimensions of the specimen are depicted in Figure 3. The length of the specimen measures 150 mm, the width is 100 mm, and the height of the honeycomb core is 19.5 mm.   The upper and lower panels of the honeycomb sandwich composite structure are made of fiber-reinforced unidirectional strips. The strips are arranged in a [45/0/90/-45]s stacking sequence. In this study, three types of materials are chosen for the panels: T300, T700, and T800. Table 1 provides information about the specimen, while Table 2 presents the data on the unidirectional strip material. As for the honeycomb core, it is composed of a regular hexagonal paper honeycomb. The typical sizes of the honeycomb cells are W = 1.8 mm and D = 3.2 mm, with a single layer wall thickness of 0.05 mm. Figure 4 illustrates the honeycomb core structure.   The upper and lower panels of the honeycomb sandwich composite structure are made of fiber-reinforced unidirectional strips. The strips are arranged in a [45/0/90/-45]s stacking sequence. In this study, three types of materials are chosen for the panels: T300, T700, and T800. Table 1 provides information about the specimen, while Table 2 presents the data on the unidirectional strip material. As for the honeycomb core, it is composed of a regular hexagonal paper honeycomb. The typical sizes of the honeycomb cells are W = 1.8 mm and D = 3.2 mm, with a single layer wall thickness of 0.05 mm. Figure 4 illustrates the honeycomb core structure.

Experimental Device Description
In accordance with the ASTM D3763 standard for impact experiments, the impact tests on the honeycomb sandwich composite specimens are conducted using the Instron CEAST 9350 drop hammer impact testing machine, as depicted in Figure 5. The drop hammer used in the experiment has a spherical shape, a mass of 5.52 kg, and a diameter of 20 mm. During the experiment, the impact energies are varied by adjusting the height of the drop hammer punch. The honeycomb sandwich composite specimens are securely fixed in a fixture, which is installed on the bedstand of the testing machine. The installation method can be seen in Figure 6.

Experimental Device Description
In accordance with the ASTM D3763 standard for impact experiments, the impact tests on the honeycomb sandwich composite specimens are conducted using the Instron CEAST 9350 drop hammer impact testing machine, as depicted in Figure 5. The drop hammer used in the experiment has a spherical shape, a mass of 5.52 kg, and a diameter of 20 mm. During the experiment, the impact energies are varied by adjusting the height of the drop hammer punch. The honeycomb sandwich composite specimens are securely fixed in a fixture, which is installed on the bedstand of the testing machine. The installation method can be seen in Figure 6.

Impact Experiment
The low-energy impact experiment on the honeycomb sandwich composite specimen is performed with impact energies of 10 J, 20 J, and 40 J, as indicated in Table 3. Various parameters and data are recorded during the experiment, including the damage mode, damage sizes, and the contact force-time curves for the punch. Additionally, the displacement and strain measurements are obtained using a digital image correlation (DIC) test system specifically for lower panels; the type of camera for the DIC is ix_speed, with a sampling frequency of 10~50 wHz. This system allows for accurate recording and analysis of the panel's deformation and strain behavior during the impact process.

Impact Experiment
The low-energy impact experiment on the honeycomb sandwich composite specimen is performed with impact energies of 10 J, 20 J, and 40 J, as indicated in Table 3. Various parameters and data are recorded during the experiment, including the damage mode, damage sizes, and the contact force-time curves for the punch. Additionally, the displacement and strain measurements are obtained using a digital image correlation (DIC) test system specifically for lower panels; the type of camera for the DIC is ix_speed, with a sampling frequency of 10~50 wHz. This system allows for accurate recording and analysis of the panel's deformation and strain behavior during the impact process.

Impact Experiment
The low-energy impact experiment on the honeycomb sandwich composite specimen is performed with impact energies of 10 J, 20 J, and 40 J, as indicated in Table 3. Various parameters and data are recorded during the experiment, including the damage mode, damage sizes, and the contact force-time curves for the punch. Additionally, the displacement and strain measurements are obtained using a digital image correlation (DIC) test system specifically for lower panels; the type of camera for the DIC is ix_speed, with a sampling frequency of 10~50 wHz. This system allows for accurate recording and analysis of the panel's deformation and strain behavior during the impact process.

Damage Modes and Discussion
Based on the low-energy impact experiment, the impact damage of the honeycomb sandwich composite specimens at different energies is obtained and is presented in  Figure 10 displays the damage profile of the T700 panel specimen after being cut from the middle. This provides a detailed view of the internal damage caused by the impact. To analyze the displacement behavior of the T700 lower panel during an impact with an energy of 40 J, Figure 11 shows the displacement contour obtained using the DIC test system. This contour map demonstrates the magnitude and distribution of displacements on the panel's surface. For a comparison of impact damage on the upper panels, Figure 12 shows the damage profiles, while Table 4 presents a summarized evaluation of the damage data for each panel specimen; the damage data is measured by a micrometer.         After analyzing the experimental results, damage modes and data of the specimens are obtained. By comparing Figure 7 to Figure 12 and Table 4, the following conclusions can be drawn: (1) Under low-energy impact, the upper panel exhibits the most severe damage, including fiber fracture, matrix cracking, and delamination. The honeycomb core in the middle experiences compressive feedback damage, while the damage to the lower panel is relatively slight.
(2) In the case of the T300 honeycomb sandwich composite specimen, subjected to an impact energy of 10 J, the main damages in the impact area of the upper panel  After analyzing the experimental results, damage modes and data of the specimens are obtained. By comparing Figure 7 to Figure 12 and Table 4, the following conclusions can be drawn:  Table 4 and Figure 12, it can be observed that as the impact energy increases, the damage width, dent depth, and other damage data increase for the same type of specimens. Under the same impact energy, different panel materials exhibit different resistance to impact damage. The T300 panel specimen suffers the largest damage area, damage width, and dent depth, indicating the most severe impact damage. The T700 panel ranks second in terms of damage severity, while the T800 panel specimen experiences the smallest amount of damage thanks to its superior strength performance and ability to resist impact damage (as shown in Table 2).

Impact Response
The contact force-time curves of the punch during the experiment are depicted in Figure 13. These curves can be divided into three stages: the rising stage, the fluctuating stage, and the falling stage. When the punch contacts the upper panel initially, the contact force increases rapidly in a short period of time, reaching the first peak. Subsequently, the contact force decreases and fluctuates within a small range before reaching the second peak, which represents the maximum value of the entire contact force-time curve. Finally, the contact force gradually decreases over time until it reaches zero.

Finite Element Model
A finite element model of a honeycomb sandwich composite specimen is created by using the finite element software ABAQUS. The modal consists of four components: the upper panel, lower panel, honeycomb core, and drop hammer. The simulation focuses on a T700 panel specimen, where the dimensions, panel material, and laying methods in the finite element model are consistent with the experimental specimen of the T700 panel.
For the paper honeycomb core, specific features are taken into consideration. The modeling incorporates W = 1.8 mm and D = 3.2 mm dimensions, as depicted in Figure 4; Figure 13. The curves of contact force-time: (a) T300 panels; (b) T700 panels; (c) T800 panels. Table 5 presents the peak impact forces for the honeycomb sandwich composite specimen under impact energies of 10 J, 20 J, and 40 J. Through comparative analysis; the following conclusions can be drawn: (1) Under the same impact energy, the panel's material has little influence on the peak contact force, with a difference of about 4%. (2) Under the same impact energy, the duration time during the impact process is basically the same for the different panel materials, and the arrival time of the peak contact force is also the same. (3) With the increase in impact energy, the peak contact force increases, but the rate of increase in the peak impact force is smaller than the rate of energy's increase. The increase in impact energy causes the contact force peak to arrive earlier.
(4) Variations in rebound are observed among panels composed of different materials when subjected to diverse levels of impact energy, leading to discrepancies in the ultimate contact duration.

Finite Element Model
A finite element model of a honeycomb sandwich composite specimen is created by using the finite element software ABAQUS. The modal consists of four components: the upper panel, lower panel, honeycomb core, and drop hammer. The simulation focuses on a T700 panel specimen, where the dimensions, panel material, and laying methods in the finite element model are consistent with the experimental specimen of the T700 panel.
For the paper honeycomb core, specific features are taken into consideration. The modeling incorporates W = 1.8 mm and D = 3.2 mm dimensions, as depicted in Figure 4; the material parameters of honeycomb core used in the calculation is taken from references [24], E 1 = 2.5 GPa, E 2 = E 3 = 1.3 GPa, ν 12 = ν 13 = 0.2, ν 23 = 0.3, ρ = 0.72 g/cm 3 . The adhesive relationship between the double-layer aramid paper is neglected, and the honeycomb structure is simplified to a single layer with a wall thickness of 0.05 mm. The drop hammer is represented as an analytical rigid body with a diameter of 20 mm. An inertial mass point measuring 5.52 kg is added at the center of the drop hammer punch. The impact energy is adjusted by manipulating the impact velocity. The finite element model, illustrated in Figure 14 (Different colors in the model represent different materials), employs 8-node linear reduced integral elements (C3D8R) for the upper and lower panels. To prevent significant mesh distortion and calculation termination, hourglass control is implemented. The laying method for the upper and lower panels follows the sequence [45/0/90/-45]s, as shown in Figure 15 (Different colors represent different fiber laying methods). The material parameters of the panel are taken from Table 2.
Considering the relatively small wall thickness of the honeycomb core compared to the core cell size, the S4R element is adopted for the Nomex honeycomb core. During the impact process, the deformation caused by the drop hammer itself is not considered, and the drop hammer is defined as an undeformed analytical, rigid body.  Table 2. Considering the relatively small wall thickness of the honeycomb core compared to the core cell size, the S4R element is adopted for the Nomex honeycomb core. During the impact process, the deformation caused by the drop hammer itself is not considered, and the drop hammer is defined as an undeformed analytical, rigid body.    In the finite element model, the damage failure criterion is incorporated by defining the material properties. For the in-plane damage model of the panel, a progressive damage evolution approach is adopted. The Hashin damage failure criterion is utilized to determine the initial damage of the panel elements. This criterion encompasses four types of failure modes: fiber tensile damage, fiber compressive damage, matrix tensile damage, and matrix compressive damage. The initial criterion serves as the starting point for material damage, and the simulation also considers damage evolution.
All contact conditions during the entire impact process are taken into account in the finite element simulation analysis. This includes the contact between the outer surface of the drop hammer and the upper panel, the contact between the outer surface of the drop hammer and the paper honeycomb core, the self-contact between the paper honeycomb core resulting from compression during the impact process, and the contact between the paper honeycomb core and the upper and lower panels. A penalty function is employed for tangential contact, while "hard" contact is used for normal contact. The friction coefficient is set to 0.1. In the finite element model, the damage failure criterion is incorporated by defining the material properties. For the in-plane damage model of the panel, a progressive damage evolution approach is adopted. The Hashin damage failure criterion is utilized to determine the initial damage of the panel elements. This criterion encompasses four types of failure modes: fiber tensile damage, fiber compressive damage, matrix tensile damage, and matrix compressive damage. The initial criterion serves as the starting point for material damage, and the simulation also considers damage evolution.
All contact conditions during the entire impact process are taken into account in the finite element simulation analysis. This includes the contact between the outer surface of the drop hammer and the upper panel, the contact between the outer surface of the drop hammer and the paper honeycomb core, the self-contact between the paper honeycomb core resulting from compression during the impact process, and the contact between the paper honeycomb core and the upper and lower panels. A penalty function is employed for tangential contact, while "hard" contact is used for normal contact. The friction coefficient is set to 0.1.
To replicate the experimental conditions, the motion of the six degrees of freedom for the four edge regions of the upper and lower panels is constrained in the model. The constraint mode is illustrated in Figure 16. The drop hammer is restricted to only moving downward in the vertical panel direction and is prevented from moving in any other direction. To replicate the experimental conditions, the motion of the six degrees of freedom for the four edge regions of the upper and lower panels is constrained in the model. The constraint mode is illustrated in Figure 16. The drop hammer is restricted to only moving downward in the vertical panel direction and is prevented from moving in any other direction.
The impact load is applied to the panel through a drop hammer punch with a mass of 5.52 kg. The impact energy is adjusted by varying the velocity of the punch. In the simulation analysis, three different impact energies are considered: 10 J, 20 J, and 40 J.

Simulation Results and Discussion
The impact simulation analysis of honeycomb sandwich composite structures is performed using the finite element software ABAQUS, and the damage modes, damage sizes, and curves of contact force-time are obtained.  The impact load is applied to the panel through a drop hammer punch with a mass of 5.52 kg. The impact energy is adjusted by varying the velocity of the punch. In the simulation analysis, three different impact energies are considered: 10 J, 20 J, and 40 J.

Simulation Results and Discussion
The impact simulation analysis of honeycomb sandwich composite structures is performed using the finite element software ABAQUS, and the damage modes, damage sizes, and curves of contact force-time are obtained.

Damage Simulation Results of the Upper Panel
The finite element simulation analysis reveals that the upper panel experiences the most severe damage, including fiber fracture, matrix cracking, etc. Specifically, for the impact energy of 20 J, the damage to the top layer of the upper panel is depicted in Figure 17, the color in the figures represent the degree of damage, while the red color represents the most severe damage. The shape of fiber damage in each layer of the upper panel appears as peanut shells, perpendicular to the direction of unidirectional fiber placement. The overall damage is determined by accumulating the damage from each layer; the damage accumulation under the impact energy of 10 J is shown in Figure 18. A comparison of the damage width between the numerical simulation results and experimental results is presented in Table 6. The analysis demonstrated that the error between the simulation results and experimental results regarding the damage width is less than 6.1%. This indicates that the simulation results are in good agreement with the experimental findings, further validating their accuracy and reliability. The overall damage is determined by accumulating the damage from each layer; the damage accumulation under the impact energy of 10 J is shown in Figure 18. A comparison of the damage width between the numerical simulation results and experimental results is presented in Table 6. The analysis demonstrated that the error between the simulation results and experimental results regarding the damage width is less than 6.1%. This indicates that the simulation results are in good agreement with the experimental findings, further validating their accuracy and reliability.  To examine the internal damage behavior of the honeycomb sandwich composites, the structures are cut along the centerline of the short side. The comparison of cross-sectional damage conditions between the simulation results and experimental results is shown in Figure 19, the color represents the degree of damage, while red color represents the most severe damage. Under an impact energy of 20 J, the fibers in the upper panel near the impact area are fractured, while the adhesive film between the upper panel and the honeycomb core remains intact. As the drop hammer punch presses down on the honeycomb core, significant compressive deformation occurs. When the impact energy increases to 40 J, the upper panel experiences overall penetration, displaying a petal-shaped failure pattern. Additionally, the honeycomb core exhibits more severe deformation, characterized by large-scale crushing failure and adhesive failure in the impact area.  To examine the internal damage behavior of the honeycomb sandwich composites, the structures are cut along the centerline of the short side. The comparison of cross-sectional damage conditions between the simulation results and experimental results is shown in Figure 19, the color represents the degree of damage, while red color represents the most severe damage. Under an impact energy of 20 J, the fibers in the upper panel near the impact area are fractured, while the adhesive film between the upper panel and the honeycomb core remains intact. As the drop hammer punch presses down on the honeycomb core, significant compressive deformation occurs. When the impact energy increases to 40 J, the upper panel experiences overall penetration, displaying a petal-shaped failure pattern. Additionally, the honeycomb core exhibits more severe deformation, characterized by large-scale crushing failure and adhesive failure in the impact area.

Damage Simulation Results of Honeycomb Core
The main damage observed in the honeycomb core includes collapse in the impact center area, as well as fracture, shear failure, and buckling. The damage area extends from the impact center position towards both sides, with the largest depth of honeycomb damage occurring in the center area, forming an arc-shaped pattern extending to both sides; the displacement contour is shown in Figure 20. The comparison of the damage between the simulation results and experimental results is illustrated in Figure 19. It can be observed that the simulation results are in good agreement with the experimental findings, confirming the effectiveness and rationality of the finite element simulation model. Utilizing this model, the size and mode of damage in the honeycomb core can be accurately predicted. The damage evolution for the specimen is depicted in Figure 21.

Damage Simulation Results of Honeycomb Core
The main damage observed in the honeycomb core includes collapse in the impact center area, as well as fracture, shear failure, and buckling. The damage area extends from the impact center position towards both sides, with the largest depth of honeycomb damage occurring in the center area, forming an arc-shaped pattern extending to both sides; the displacement contour is shown in Figure 20. The comparison of the damage between the simulation results and experimental results is illustrated in Figure 19. It can be observed that the simulation results are in good agreement with the experimental findings, confirming the effectiveness and rationality of the finite element simulation model. Utilizing this model, the size and mode of damage in the honeycomb core can be accurately predicted. The damage evolution for the specimen is depicted in Figure 21.

Damage Simulation Results of Honeycomb Core
The main damage observed in the honeycomb core includes collapse in the impact center area, as well as fracture, shear failure, and buckling. The damage area extends from the impact center position towards both sides, with the largest depth of honeycomb damage occurring in the center area, forming an arc-shaped pattern extending to both sides; the displacement contour is shown in Figure 20. The comparison of the damage between the simulation results and experimental results is illustrated in Figure 19. It can be observed that the simulation results are in good agreement with the experimental findings, confirming the effectiveness and rationality of the finite element simulation model. Utilizing this model, the size and mode of damage in the honeycomb core can be accurately predicted. The damage evolution for the specimen is depicted in Figure 21.

Curves of contact force-time
The curves of contact force-time of drop hammer punch between simulation and experiment are shown in Figure 24, which shows that the simulation results have good agreement with the experimental results.

Curves of Contact Force-Time
The curves of contact force-time of drop hammer punch between simulation and experiment are shown in Figure 24, which shows that the simulation results have good agreement with the experimental results.

Curves of contact force-time
The curves of contact force-time of drop hammer punch between simulation and experiment are shown in Figure 24, which shows that the simulation results have good agreement with the experimental results.

Summary
(1) In order to study the damage behavior of the honeycomb sandwich composite structure with a T700 panel, a comprehensive finite element model is established. This model takes into account factors such as the microscopic failure criteria of the panel, various contact conditions during the impact process, and the simplification of the drop hammer punch. By considering these aspects, the model aims to accurately represent the behavior of the composite structure during impact. (2) Through the finite element simulation analysis, the damage modes, damage parameters, and contact force-time curves of the honeycomb sandwich composites under different impact energies are obtained. These results provide valuable insights into the response and failure mechanisms of the composite structure under impact loading. (3) A comparison between the simulation results and experimental results demonstrates good consistency between the two. This agreement between the simulation and experimental findings confirms the correctness and validity of the finite element model employed in the study.

Conclusions
This study focuses on the impact damage behavior of paper honeycomb sandwich composite specimens through experiments and finite element simulations. Various factors

Summary
(1) In order to study the damage behavior of the honeycomb sandwich composite structure with a T700 panel, a comprehensive finite element model is established. This model takes into account factors such as the microscopic failure criteria of the panel, various contact conditions during the impact process, and the simplification of the drop hammer punch. By considering these aspects, the model aims to accurately represent the behavior of the composite structure during impact. (2) Through the finite element simulation analysis, the damage modes, damage parameters, and contact force-time curves of the honeycomb sandwich composites under different impact energies are obtained. These results provide valuable insights into the response and failure mechanisms of the composite structure under impact loading. (3) A comparison between the simulation results and experimental results demonstrates good consistency between the two. This agreement between the simulation and experimental findings confirms the correctness and validity of the finite element model employed in the study.

Conclusions
This study focuses on the impact damage behavior of paper honeycomb sandwich composite specimens through experiments and finite element simulations. Various factors influencing the damage, such as panel material and impact energy, are considered. The following conclusions are drawn: (1) Low-energy impact experiments are conducted on honeycomb sandwich composite specimens to determine the damage modes and damage parameters. During the impact process, the upper panel exhibits severe damage, including matrix cracking, fiber breakage, and delamination. The honeycomb core experiences compressive feedback damage, while the lower panel shows relatively minor damage, indicating the energy absorption capabilities of the honeycomb core. (2) The impact resistance of panels made from T300, T700, or T800 laminates increases with the strength of the laminate. Among them, the T800 panel demonstrates the highest resistance to impact. (3) Increasing the impact energy results in an increase in peak contact force. However, the rate of increase in peak impact force is smaller than the rate of energy increase. The panel material has minimal influence on the peak contact force, while the impact energy affects the impact time history.
(4) Through finite element simulation analysis of honeycomb sandwich composites, the damage modes and damage sizes for the panels and honeycomb core are obtained, and the contact force-time curves of the punch are obtained. Comparative analysis demonstrates good consistency between the simulation and experimental results, confirming the rationality and accuracy of the finite element model.