A high-precision calculation method for underwater explosion shock waves without numerical dissipation based on a dynamic stretching coordinate system

By combining partitioning and improved solver in dynamic stretching coordinate system, the numerical dissipation and non-physical oscillation problems of underwater explosion caused by traditional fixed coordinate system are solved, and the high-precision simulation of underwater explosion shock wave is realized, and the calculation is improved. accuracy and stability.

CN117763898BActive Publication Date: 2024-11-26HARBIN ENG UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202311669593.6
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-12-07
Publication Date
2024-11-26
Estimated Expiration
2043-12-07

AI Technical Summary

Technical Problem

In numerical simulation, traditional fixed coordinate systems often lead to numerical dissipation and non-physical oscillation of underwater explosion shock waves, and the assumption of uniform internal pressure of bubbles in bubble dynamics leads to calculation errors, making it difficult to achieve accurate solution to the entire flow field of underwater explosion.

Method used

The high-precision calculation method for underwater explosion shock wave myriad dissipation based on dynamic stretching coordinate system is adopted. By dividing the flow field into inner region, outer region and undisturbed region, and the improved HLLC Li is adopted in the dynamic stretching coordinate system. Mann solver and Hugoniot conditions realize coupling and flux calculation of inner and outer regions, and discrete solutions are performed by combining high-precision intermittent Galerkin method.

Benefits of technology

Effectively eliminate numerical dissipation and non-physical oscillation, improving the accuracy of numerical calculations, especially in the simulation of near-field underwater explosion shock waves, high-precision solution of shock wave generation and propagation in the early stage of underwater explosion is realized, reducing the influence of bubbles. , improves the accuracy and stability of the simulation.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN117763898B_ABST
    Figure CN117763898B_ABST
Patent Text Reader

Abstract

The present invention discloses a non-oscillating high-precision calculation method for underwater explosion shock waves based on a dynamic stretching coordinate system, and belongs to the field of underwater explosion numerical calculation. The flow field is divided into an inner zone, an outer zone, and an undisturbed zone according to the bubble surface and the strong discontinuity position of the shock wave; the control equations of the inner zone and the outer zone are respectively transformed into a dynamic stretching coordinate system; in the dynamic stretching coordinate system, a two-phase improved HLLC Riemann solver is used to calculate the bubble surface flux to achieve the coupling of the inner zone and the outer zone; in the dynamic stretching coordinate system, the Hugoniout condition is used to calculate the flux at the shock wave discontinuity to achieve the coupling of the outer zone and the undisturbed zone; the initial conditions are obtained based on the improved HLLC Riemann solver; and the high-precision discontinuous Galerkin method is used to perform discrete solutions to the inner zone and the outer zone. The present invention avoids the non-physical oscillation and dissipation of the traditional method for simulating underwater explosion shock waves and bubble processes, improves the calculation accuracy, and makes the calculation results closer to the actual situation.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to a high-precision calculation method for underwater explosion shock waves without numerical dissipation based on a dynamic stretching coordinate system, and belongs to the field of underwater explosion numerical calculation. Background Art

[0002] The shock waves generated by underwater explosions are extremely powerful and can cause serious damage to marine structures. Accurate calculation of underwater explosion loads has important application value. Underwater explosions generate bubbles and shock waves. The shock waves generated by underwater explosions have the characteristics of strong discontinuity, nonlinearity, large energy transfer, and fast propagation speed. In the experiment, only the pressure changes of the far-field measurement points can be measured. It is difficult to capture the shock waves at the near-field measurement points and the experimental equipment is easily damaged by the shock waves. Therefore, numerical simulation has become an effective means to study the shock waves of near-field underwater explosions. Li et al. used an improved characteristic method to simulate underwater shock waves in one-dimensional model underwater explosions. Nguyen et al. proposed an accurate shock and interface accuracy method that uses curve body modeling to simulate compressible multiphase flow carrying shock waves. Liu et al. developed an Euler finite element method (EFEM) for underwater explosion simulation. The method was verified by shock wave and bubble oscillation simulation. However, due to the use of a fixed coordinate system and Euler grid, it is necessary to express the bubble surface contact discontinuity and the strong shock wave discontinuity between the grids. It is inevitable to balance numerical dissipation and non-physical oscillation, and the two cannot be eliminated at the same time. The discontinuous Galerkin method (DG method) is highly accurate and researchers have found it to be very suitable for dealing with the discontinuity of shock waves from underwater explosions. The basic idea of ​​the discontinuous Galerkin method can be traced back to the 1970s, but it was not systematically proposed at the time. At this stage, researchers began to explore the application of piecewise polynomial approximation to solving partial differential equations. However, it was not until the early 1990s that these ideas were systematically applied to numerical methods. The DG method is widely used in solving Navier-Stokes equations and other fluid dynamics problems. It can effectively deal with complex flow phenomena such as shock waves, interfaces and turbulence, as well as multi-field coupling problems such as multi-media flow and multiphase flow. The DG method is very suitable for dealing with the complex phenomenon of discontinuity of shock waves from underwater explosions. The fourth-order Runge-Kutta method (RK method) is very suitable for time discretization, and combined with the DG method to form the RKDG algorithm.

[0003] In recent years, the DG method has been gradually applied to the study of underwater explosion shock waves. Wu et al. used the local discrete Galerkin (LDG) method to establish a local discontinuous axis estimation model to calculate the dynamic pressure in the fluid field and capture high-resolution shock waves. Zhang et al. used the (RKDG) method and the connection scheme of the finite element method to study the pressure characteristics of near-field underwater explosions. Liang et al. used the combined RKDG-FEM method to study underwater explosions near composite structures. However, since the traditional fixed coordinate system often brings numerical dissipation and non-physical oscillations to the numerical simulation, and the calculation errors caused by the traditional assumption of uniform pressure inside the bubble in bubble dynamics, the numerical simulation of the shock wave of the near-field underwater explosion is greatly affected by the bubble, and the calculation results of the far-field underwater explosion are also somewhat different from the experimental results, and it is ultimately difficult to achieve an accurate solution of the entire flow field of the underwater explosion. The dynamic stretching coordinate system RKDG algorithm of the present invention breaks away from the constraints of the traditional fixed coordinate system and eliminates the numerical dissipation and non-physical oscillation of the fixed coordinate system algorithm at strong discontinuities. It not only improves the accuracy of numerical calculations, but also can observe the physical process of the generation and movement of secondary shock waves (rarefaction waves) with high precision, thus achieving high-precision solution of shock wave generation and propagation in the early stage of underwater explosions. Summary of the invention

[0004] The present invention proposes a high-precision calculation method for underwater explosion shock waves without numerical dissipation based on a dynamic stretching coordinate system, aiming to solve the technical problems that the traditional fixed coordinate system often brings numerical dissipation and non-physical oscillation to numerical simulation, and the calculation error caused by the traditional assumption of uniform internal pressure of bubbles in bubble dynamics, resulting in the numerical simulation of shock waves of near-field underwater explosions being greatly affected by bubbles, and the calculation results of far-field underwater explosions having a certain gap with experimental results.

[0005] A high-precision calculation method for underwater explosion shock waves without numerical dissipation based on a dynamic stretching coordinate system, the high-precision calculation method for underwater explosion shock waves without numerical dissipation based on a dynamic stretching coordinate system comprising the following steps:

[0006] S100, dividing the flow field into an inner zone, an outer zone, and an undisturbed zone according to the bubble surface contact discontinuity and the shock wave intensity discontinuity positions;

[0007] S200, respectively transforming the control equations of the inner zone and the outer zone into a dynamic stretching coordinate system;

[0008] S300, the improved HLLC Riemann solver is the underwater explosion Riemann solver, and in the dynamic stretching coordinate system, the two-phase improved HLLC Riemann solver is used to calculate the bubble surface flux to achieve the coupling between the inner and outer zones;

[0009] S400, in the dynamic stretching coordinate system, the flux at the shock wave discontinuity is calculated using the Hugoniot condition to achieve coupling between the outer zone and the undisturbed zone;

[0010] S500, initial conditions are obtained based on the improved HLLC Riemann solver;

[0011] S600, uses high-precision discontinuous Galerkin method to discretely solve the inner and outer regions.

[0012] Further, in S100, the inner zone is the space inside the bubble of the underwater explosion, the outer zone is the space outside the bubble swept by the shock wave, and the undisturbed zone is the space outside the bubble that is not swept by the shock wave, that is, the still water area.

[0013] Further, in S200, the fixed coordinate system is converted into a dynamic stretching coordinate system, wherein both the fixed coordinate system and the dynamic stretching coordinate system are spherical coordinate systems. For the underwater explosion of spherical TNT explosives, it is assumed that all items are isotropic, and only the radial motion of the spherical shock wave and bubbles generated by the underwater explosion of the spherical TNT explosives is concerned.

[0014]

[0015] Among them, ζ is the radial coordinate of the dynamic stretching coordinate system, r is the radial coordinate of the fixed spherical coordinate system, and r 1 is the coordinate of the fixed coordinate system corresponding to the left endpoint of the dynamic stretching coordinate system, r 2 is the coordinate of the fixed coordinate system corresponding to the right endpoint of the dynamic stretching coordinate system,

[0016] For the flow field inside the bubble, r 1 =0, r 2 =R, r 1 represents the explosion center, i.e., the bubble center, r 2 represents the bubble surface, R is the bubble radius, and for the flow field outside the bubble, r 1 =R, r 2 =R S , r 1 represents the bubble surface, r 2 Indicates the interruption of the shock wave, R s is the distance from the shock wave to the explosion center, thus forming two dynamic stretching coordinate systems, one is the dynamic stretching coordinate system of the flow field inside the bubble, and the other is the dynamic stretching coordinate system of the flow field outside the bubble. Their r 1 and r 2 Different sizes represent different meanings;

[0017] Substituting equation (1) into the one-dimensional spherical coordinate system Euler equation in the fixed coordinate system, after a series of processing, the conservation equation of the one-dimensional spherical coordinate system Euler equation in the dynamic stretching coordinate system is obtained:

[0018]

[0019] Where U = (ρ, ρu, E) T is the conserved quantity of the conservation formula, which is the same as the conserved quantity of the fixed coordinate system.

[0020] is the numerical flux in the dynamic stretching coordinate system,

[0021] is the source intensity in the dynamic stretching coordinate system.

[0022] Further, the series of processing is:

[0023] The equation (1) is derived with respect to time t, and the velocity relationship between the dynamic stretching coordinate system and the fixed coordinate system is obtained:

[0024]

[0025] u is the velocity in the fixed coordinate system. It is known that the velocity in the dynamic stretching coordinate system is So the above formula (5) becomes:

[0026]

[0027] Next, the velocity in the dynamic stretching coordinate system is solved according to the above formula:

[0028]

[0029] Then solve the material derivatives of the conserved quantity U in the fixed coordinate system and the dynamic stretching coordinate system respectively:

[0030]

[0031] Subtract equation (9) from equation (8), and then combine equation (1) and equation (6) to obtain:

[0032]

[0033] Finally, equation (2) is obtained through equation (10), which is the conservation equation of the spherical coordinate one-dimensional Euler equation in the dynamic stretching coordinate system.

[0034] Furthermore, in S300, the following steps are included:

[0035] S310, defining an initial moment, assuming that at the initial moment the explosion has occurred and a bubble has been formed, and a rarefaction wave and a shock wave have been formed on the bubble surface and are about to propagate;

[0036] S320. Clarify the description of the Riemann problem, including the state description of two fluids and the separation condition of the two fluids, wherein the two fluids include explosion gas and water;

[0037] S330, determining the initial conditions inside and outside the bubble, and calculating and defining the pressure, velocity and density inside and outside the bubble;

[0038] S340, using Hugoniot conditions to determine the relationship between the shock wave's output pressure and wave velocity;

[0039] S350, using the velocity relationship of the basic HLLC Riemann solver, improving the velocity relationship, and obtaining the wave velocities of the rarefaction wave and the shock wave based on the velocity relationship;

[0040] S360. Obtain the final improved Riemann solver.

[0041] Furthermore, in S300, specifically,

[0042] The underwater explosion Riemann solver is an improved solver based on the HLLC Riemann solver. The inside of the bubble is gas, and the outside of the bubble is water. An inward rarefaction wave is formed inside the bubble surface, and an outward shock wave is formed outside the bubble. It is a classic two-phase flow problem. At the initial moment t = ε, it is assumed that at this moment, the underwater explosion has occurred and a bubble has formed, and the rarefaction wave and shock wave are formed on the bubble surface and are about to propagate. At this time, t is close to 0.

[0043] The Riemann problem for underwater explosions is described as follows: the two fluids are separated by the bubble surface. Inside the bubble, the high-pressure explosion gas products are initially at rest, and outside the bubble, the undisturbed water is also at rest. The initial conditions inside and outside the bubble, i.e., the conserved quantities, are respectively given by U L and U R Given that, in addition, U L and U R The JWL state equation and the Tammann state equation are used to describe the rarefaction wave, i.e. the second shock wave, respectively. The wave velocity is denoted as S L , the velocity of the bubble surface is S * , the shock wave speed is S R , so there is

[0044]

[0045] Among them, u * is the initial average velocity of the bubble surface in a short time, is the initial average velocity of the first element on the left side of the bubble surface, is the initial average velocity of all cells to the right of the bubble surface, p * is the average pressure on the bubble surface over a short period of time. is the initial average pressure of the first cell to the left of the bubble surface, is the initial average pressure of all cells on the right side of the bubble surface. Then, using the Hugoniot condition, we get the relationship between the shock wave pressure and wave velocity:

[0046]

[0047] Among them, P 0 is the hydrostatic pressure, C R =C 0 is the speed of sound in still water. Using the velocity relationship of the HLLC Riemann solver, we have:

[0048]

[0049] Combining (18) and (19), we get

[0050]

[0051] Among them, ρ L is the initial density of area A, i.e., the density of the explosion gas, ρ R is the initial density of the undisturbed zone, i.e., the still water density. The negative value of the average value of the sound speed in the inner zone and zone B is calculated as the rarefaction wave speed S L ,Right now

[0052]

[0053] The sound speed in the inner zone and zone B is calculated by the following formula:

[0054]

[0055] The density of area B is calculated by the conservation of mass before and after the explosion:

[0056]

[0057] where h L is the actual space length of zone B when the explosion occurs at time t = ε, h L -εu * This is the actual spatial length of area B before the explosion.

[0058] The actual space length h of zone B when the explosion occurs at time t = ε L is the length of space swept by the rarefaction wave and bubble surface, so:

[0059]

[0060] Among them, R c is the initial radius of the spherical TNT explosive, N = 100 is the number of grids, because h Lis the actual spatial length of area B, which is the 100th unit from the left to the right starting from the center of the bubble. Therefore, at the initial time t = ε, h L is the actual length of a unit.

[0061] At this point, the improved Riemann solver is set up.

[0062] Furthermore, in S400, the flux at the shock wave discontinuity is calculated using the Hugoniot condition, and the static water condition of the undisturbed zone is used as the boundary condition at the shock wave discontinuity to achieve mutual coupling between the outer zone and the undisturbed zone.

[0063] Furthermore, in S500, the following steps are included:

[0064] S510, performing area division, dividing the entire initial flow field into four areas, namely area A, area B, area C and area D, wherein:

[0065] Area A represents the interior of the bubble of the underwater explosion. This area is not swept by the second shock wave, that is, not swept by the inward rarefaction wave. It is part of the inner zone and belongs to the inner zone.

[0066] Area B represents the interior of the bubble of the underwater explosion. This area is swept by the second shock wave and also belongs to the inner area.

[0067] Zone C represents the outside of the bubble of the underwater explosion, which is swept by the shock wave and belongs to the outer zone;

[0068] Area D represents the outside of the bubble of the underwater explosion, which is not swept by the shock wave and belongs to the calm water area;

[0069] S520, obtaining initial conditions of four regions, where:

[0070] The initial conditions of regions A and B are provided by the underwater explosion Riemann solver, which considers the properties and initial state of the internal bubbles;

[0071] The initial conditions in region C were calculated based on the Tamman equation of state;

[0072] The initial conditions in region D are those in the still water region, usually the pressure value of still water;

[0073] The initial conditions include velocity, pressure and density.

[0074] S530. At the bubble surface, that is, at the junction of area B and the outer area, the conditions of pressure continuity and velocity equality are used as boundary conditions.

[0075] Furthermore, in S600, for equation (2), the discontinuous Galerkin method, namely the DG method, is used for spatial discretization, namely:

[0076]

[0077] In terms of time, the fourth-order Runge-Kutta method, namely the RK method, is used for discretization, thus forming the dynamic stretching coordinate system RKDG algorithm.

[0078] A storage medium stores a computer program, which, when executed by a processor, implements the above-mentioned high-precision calculation method for underwater explosion shock waves without numerical dissipation based on a dynamic stretching coordinate system.

[0079] Beneficial effects of the present invention: The present invention proposes a high-precision calculation method for underwater explosion shock waves without numerical dissipation based on a dynamic stretching coordinate system, which aims to solve the problems of numerical dissipation and non-physical oscillation in the traditional numerical simulation of underwater explosion shock waves. By breaking away from the limitations of the traditional fixed coordinate system, the accuracy of numerical calculations is improved, especially in the simulation of near-field underwater explosion shock waves. The present invention improves the bubble dynamics modeling, making the numerical simulation closer to the actual situation, and has potential applications in multiple fields. It is expected to improve the reliability of numerical simulation of underwater explosion loads, which is beneficial to the progress of related research and application fields. BRIEF DESCRIPTION OF THE DRAWINGS

[0080] Figure 1 The flowchart of the dynamic stretching coordinate system RKDG algorithm proposed by the present invention to achieve accurate solution of the entire flow field in the early stage of underwater explosion of spherical explosives;

[0081] Figure 2 The spatial division and coordinate system transformation diagram of the whole flow field of underwater explosion of spherical explosive proposed by the present invention;

[0082] Figure 3 It is a space partition diagram of the Riemann problem at the initial moment of the whole flow field of the underwater explosion of the spherical explosive proposed by the present invention;

[0083] Figure 4 A comparison diagram of the reference solution of the dynamic stretching coordinate system RKDG algorithm and the fixed coordinate system RKDG and one-dimensional RCM method for solving the one-dimensional spherically symmetric air explosion space pressure distribution;

[0084] Figure 5 The pressure distribution diagram of a 7.0kg spherical TNT explosive at a water depth of 6m calculated by the dynamic stretching coordinate system RKDG algorithm proposed by the present invention at different positions and times;

[0085] Figure 6 This is a comparison chart of the pressure results of a 7.0kg spherical TNT explosive at a water depth of 6m calculated using the dynamic tensile coordinate system RKDG algorithm and LS-dyna software proposed in the present invention at a measuring point at 4m and the experimental results. DETAILED DESCRIPTION

[0086] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0087] A high-precision calculation method for underwater explosion shock waves without numerical dissipation based on a dynamic stretching coordinate system, the high-precision calculation method for underwater explosion shock waves without numerical dissipation based on a dynamic stretching coordinate system comprising the following steps:

[0088] S100, dividing the flow field into an inner zone, an outer zone, and an undisturbed zone according to the bubble surface contact discontinuity and the shock wave intensity discontinuity positions;

[0089] S200, respectively transforming the control equations of the inner zone and the outer zone into a dynamic stretching coordinate system;

[0090] S300, the improved HLLC Riemann solver is the underwater explosion Riemann solver, and in the dynamic stretching coordinate system, the two-phase improved HLLC Riemann solver is used to calculate the bubble surface flux to achieve the coupling between the inner and outer zones;

[0091] S400, in the dynamic stretching coordinate system, the flux at the shock wave discontinuity is calculated using the Hugoniot condition to achieve coupling between the outer zone and the undisturbed zone;

[0092] S500, initial conditions are obtained based on the improved HLLC Riemann solver;

[0093] S600, uses high-precision discontinuous Galerkin method to discretely solve the inner and outer regions.

[0094] Specifically, by dynamically stretching the coordinate system and improving the numerical method, the common numerical dissipation and non-physical oscillation problems in the traditional fixed coordinate system can be effectively reduced, and the accuracy and stability of the simulation results can be improved. The error caused by the assumption of uniform pressure inside the bubble in the traditional method is solved. The method of the present invention adopts the improved HLLC Riemann solver and Hugoniot condition to more accurately calculate the bubble surface flux and the flux at the shock wave discontinuity, thereby more accurately describing the bubble dynamics and reducing the degree to which the simulation is affected by the bubble. By dividing the flow field into an inner zone, an outer zone and an undisturbed zone, and transforming and coupling them in a dynamically stretched coordinate system, the coupling between the inner and outer zones and the connection between the outer zone and the undisturbed zone are achieved, and the overall consistency and accuracy of the simulation are improved. The initial conditions obtained based on the improved HLLC Riemann solver are more accurate, which helps to improve the convergence speed and accuracy of the simulation. The high-precision discontinuous Galerkin method is used to discretely solve the inner and outer zones, which improves the accuracy of the entire calculation method and makes the simulation results closer to the actual situation.

[0095] Further, in S100, the inner zone is the space inside the bubble of the underwater explosion, the outer zone is the space outside the bubble swept by the shock wave, and the undisturbed zone is the space outside the bubble that is not swept by the shock wave, that is, the still water area.

[0096] Specifically, distinguishing between internal, external, and undisturbed areas allows for more accurate simulation of the different impact ranges of underwater explosions. This classification allows the simulation to more accurately describe the different dynamic behaviors of the bubble interior, the external area affected by the shock wave, and the undisturbed still water area, which helps the simulation to more accurately reflect the impact and behavior of underwater explosion shock waves in different spatial states, improving the authenticity and credibility of the simulation.

[0097] Further, in S200, the fixed coordinate system is converted into a dynamic stretching coordinate system, wherein both the fixed coordinate system and the dynamic stretching coordinate system are spherical coordinate systems. For the underwater explosion of spherical TNT explosives, it is assumed that all items are isotropic, and only the radial motion of the spherical shock wave and bubbles generated by the underwater explosion of the spherical TNT explosives is concerned.

[0098]

[0099] Among them, ζ is the radial coordinate of the dynamic stretching coordinate system, r is the radial coordinate of the fixed spherical coordinate system, and r 1 is the coordinate of the fixed coordinate system corresponding to the left endpoint of the dynamic stretching coordinate system, r 2 is the coordinate of the fixed coordinate system corresponding to the right endpoint of the dynamic stretching coordinate system,

[0100] For the flow field inside the bubble, r 1 =0, r 2 =R, r 1represents the explosion center, i.e., the bubble center, r 2 represents the bubble surface, R is the bubble radius, and for the flow field outside the bubble, r 1 =R, r 2 =R S , r 1 represents the bubble surface, r 2 Indicates the interruption of the shock wave, R s is the distance from the shock wave to the explosion center, thus forming two dynamic stretching coordinate systems, one is the dynamic stretching coordinate system of the flow field inside the bubble, and the other is the dynamic stretching coordinate system of the flow field outside the bubble. Their r 1 and r 2 Different sizes represent different meanings;

[0101] Substituting equation (1) into the one-dimensional spherical coordinate system Euler equation in the fixed coordinate system, after a series of processing, the conservation equation of the one-dimensional spherical coordinate system Euler equation in the dynamic stretching coordinate system is obtained:

[0102]

[0103] Where U = (ρ, ρu, E) T is the conserved quantity of the conservation formula, which is the same as the conserved quantity of the fixed coordinate system.

[0104] is the numerical flux in the dynamic stretching coordinate system,

[0105] is the source intensity in the dynamic stretching coordinate system.

[0106] Further, the series of processing is:

[0107] The equation (1) is derived with respect to time t, and the velocity relationship between the dynamic stretching coordinate system and the fixed coordinate system is obtained:

[0108]

[0109] u is the velocity in the fixed coordinate system. It is known that the velocity in the dynamic stretching coordinate system is So the above formula (5) becomes:

[0110]

[0111] Next, the velocity in the dynamic stretching coordinate system is solved according to the above formula:

[0112]

[0113] Then solve the material derivatives of the conserved quantity U in the fixed coordinate system and the dynamic stretching coordinate system respectively:

[0114]

[0115] Subtract equation (9) from equation (8), and then combine equation (1) and equation (6) to obtain:

[0116]

[0117] Finally, equation (2) is obtained through equation (10), which is the conservation equation of the spherical coordinate one-dimensional Euler equation in the dynamic stretching coordinate system.

[0118] Specifically, in S200, the fixed spherical coordinate system is transformed into a dynamic stretching coordinate system, and the conservation formula of the one-dimensional spherical coordinate system Euler equation in the dynamic stretching coordinate system is derived through a series of processing for the radial shock wave and bubble motion of the underwater explosion of spherical TNT explosives. This transformation makes the model more suitable for the characteristics of underwater explosions, simplifies the description of the problem, focuses on the radial motion caused by the explosion, and provides the relationship between conservation quantities and velocities, which helps to more accurately understand and simulate the evolution of the flow field after the explosion, especially the dynamic behavior of different regions inside and outside the bubble.

[0119] Furthermore, in S300, the following steps are included:

[0120] S310, defining an initial moment, assuming that at the initial moment the explosion has occurred and a bubble has been formed, and a rarefaction wave and a shock wave have been formed on the bubble surface and are about to propagate;

[0121] S320. Clarify the description of the Riemann problem, including the state description of two fluids and the separation condition of the two fluids, wherein the two fluids include explosion gas and water;

[0122] S330, determining the initial conditions inside and outside the bubble, and calculating and defining the pressure, velocity and density inside and outside the bubble;

[0123] S340, using Hugoniot conditions to determine the relationship between the shock wave's output pressure and wave velocity;

[0124] S350, using the velocity relationship of the basic HLLC Riemann solver, improving the velocity relationship, and obtaining the wave velocities of the rarefaction wave and the shock wave based on the velocity relationship;

[0125] S360. Obtain the final improved Riemann solver.

[0126] Specifically, in S300, a series of beneficial improvements are made to the underwater explosion shock wave. First, the initial moment is defined in S310, and the state of the explosion forming bubbles, rarefaction waves and shock waves on the bubble surface is clarified. Then, the Riemann problem is clarified in S320, including the description and separation conditions of the two fluid states of explosion gas and water. In S330, the initial conditions inside and outside the bubble are calculated and defined, providing the necessary parameters for the simulation. By determining the characteristics of the shock wave using the Hugoniot condition (S340) and improving the velocity relationship of the basic HLLC Riemann solver (S350), the wave speeds of the rarefaction wave and the shock wave are obtained. Finally, in S360, an improved Riemann solver is obtained, which can more accurately simulate the propagation of rarefaction waves and shock waves during underwater explosions, and improve the accuracy and reliability of the simulation. These improvements ensure a more accurate description of the underwater explosion shock wave, thereby providing a more reliable tool for understanding and analyzing the underwater explosion process.

[0127] Furthermore, in S300, specifically,

[0128] The underwater explosion Riemann solver is an improved solver based on the HLLC Riemann solver. The inside of the bubble is gas, and the outside of the bubble is water. An inward rarefaction wave is formed inside the bubble surface, and an outward shock wave is formed outside the bubble. It is a classic two-phase flow problem. At the initial moment t = ε, it is assumed that at this moment, the underwater explosion has occurred and a bubble has formed, and the rarefaction wave and shock wave are formed on the bubble surface and are about to propagate. At this time, t is close to 0.

[0129] The Riemann problem for underwater explosions is described as follows: the two fluids are separated by the bubble surface. Inside the bubble, the high-pressure explosion gas products are initially at rest, and outside the bubble, the undisturbed water is also at rest. The initial conditions inside and outside the bubble, i.e., the conserved quantities, are respectively given by U L and U R Given that, in addition, U L and U R The JWL state equation and the Tammann state equation are used to describe the rarefaction wave, i.e. the second shock wave, respectively. The wave velocity is denoted as S L , the velocity of the bubble surface is S * , the shock wave speed is S R , so there is

[0130]

[0131]

[0132] Among them, u * is the initial average velocity of the bubble surface in a short time, is the initial average velocity of the first element on the left side of the bubble surface, is the initial average velocity of all cells to the right of the bubble surface, p * is the average pressure on the bubble surface over a short period of time. is the initial average pressure of the first cell to the left of the bubble surface, is the initial average pressure of all cells on the right side of the bubble surface. Then, using the Hugoniot condition, we get the relationship between the shock wave pressure and wave velocity:

[0133]

[0134]

[0135] Among them, P 0 is the hydrostatic pressure, C R =C 0 is the speed of sound in still water. Using the velocity relationship of the HLLC Riemann solver, we have:

[0136]

[0137] Combining (18) and (19), we get

[0138]

[0139] Among them, ρ L is the initial density of area A, i.e., the density of the explosion gas, ρ R is the initial density of the undisturbed zone, i.e., the still water density. The negative value of the average value of the sound speed in the inner zone and zone B is calculated as the rarefaction wave speed S L ,Right now

[0140]

[0141] The sound speed in the inner zone and zone B is calculated by the following formula:

[0142]

[0143] The density of area B is calculated by the conservation of mass before and after the explosion:

[0144]

[0145] where h L is the actual space length of zone B when the explosion occurs at time t = ε, h L -εu * That is the actual space length of area B before the explosion.

[0146] The actual space length h of zone B when the explosion occurs at time t = ε L is the length of space swept by the rarefaction wave and bubble surface, so:

[0147]

[0148]

[0149] Among them, R c is the initial radius of the spherical TNT explosive, N = 100 is the number of grids, because h L is the actual spatial length of area B, which is the 100th unit from the left to the right starting from the center of the bubble. Therefore, at the initial time t = ε, h L is the actual length of a unit.

[0150] At this point, the improved Riemann solver is set up.

[0151] Specifically, in S300, the underwater explosion Riemann solver is improved in detail, and the simulation accuracy of the underwater explosion problem is improved based on the modification of the HLLC Riemann solver. The solver takes into account the situation that the inside of the bubble is gas and the outside of the bubble is water, and describes the inward rarefaction wave formed inside the bubble surface and the outward shock wave formed outside the bubble. At the initial moment, when the underwater explosion occurs and the bubble is formed, the rarefaction wave and shock wave that are about to propagate on the bubble surface are taken into account.

[0152] Specifically, the Riemann problem describes the situation where two fluids are separated by the surface of a bubble, with the high-pressure explosion gas inside the bubble initially at rest and the undisturbed water outside the bubble also at rest. The initial conditions are given by conservation quantities, the JWL equation of state, and the Tammann equation of state, while the velocities of the rarefaction and shock waves are considered. The relationship between the exit pressure and the velocity of the shock wave is established through the Hugoniot condition and the velocity relation of the HLLC Riemann solver. In addition, the velocity of the rarefaction wave is calculated, which is obtained by calculating the speed of sound.

[0153] Throughout the process, through the detailed description of the Riemann problem and the derivation of mathematical formulas, the improved solver more accurately simulates the fluid behavior inside and outside the bubble during underwater explosions, providing a more reliable tool for understanding and analyzing underwater explosion processes and enhancing the accuracy and credibility of rarefaction wave and shock wave propagation.

[0154] Furthermore, in S400, the flux at the shock wave discontinuity is calculated using the Hugoniot condition, and the static water condition of the undisturbed zone is used as the boundary condition at the shock wave discontinuity to achieve mutual coupling between the outer zone and the undisturbed zone.

[0155] Specifically, in S400, by using the Hugoniot condition to calculate the flux at the shock wave discontinuity and taking the still water condition of the undisturbed zone as the boundary condition at the shock wave discontinuity, the mutual coupling between the outer zone and the undisturbed zone is successfully achieved. This means that the impact of the shock wave propagation to the undisturbed zone is taken into account in the simulation, and the interaction between the outer zone and the undisturbed zone is achieved by calculating the flux at the shock wave discontinuity and taking the still water condition of the undisturbed zone as the boundary condition.

[0156] The beneficial effect of this step is that it more comprehensively considers the propagation and impact of shock waves, making the simulation closer to reality. By coupling the outer zone and the undisturbed zone, the model can more accurately capture the propagation behavior of shock waves, and considers the static water conditions in the undisturbed zone, further improving the authenticity and reliability of the simulation. This mutually coupled treatment method helps to more fully understand the propagation mechanism of underwater explosion shock waves.

[0157] Furthermore, in S500, the following steps are included:

[0158] S510, performing area division, dividing the entire initial flow field into four areas, namely area A, area B, area C and area D, wherein:

[0159] Area A represents the interior of the bubble of the underwater explosion. This area is not swept by the second shock wave, that is, not swept by the inward rarefaction wave. It is part of the inner zone and belongs to the inner zone.

[0160] Area B represents the interior of the bubble of the underwater explosion. This area is swept by the second shock wave and also belongs to the inner area.

[0161] Zone C represents the outside of the bubble of the underwater explosion, which is swept by the shock wave and belongs to the outer zone;

[0162] Area D represents the outside of the bubble of the underwater explosion, which is not swept by the shock wave and belongs to the calm water area;

[0163] S520, obtaining initial conditions of four regions, where:

[0164] The initial conditions of regions A and B are provided by the underwater explosion Riemann solver, which considers the properties and initial state of the internal bubbles;

[0165] The initial conditions in region C were calculated based on the Tamman equation of state;

[0166] The initial conditions in region D are those in the still water region, usually the pressure value of still water;

[0167] The initial conditions include velocity, pressure and density.

[0168] S530. At the bubble surface, that is, at the junction of area B and the outer area, the conditions of pressure continuity and velocity equality are used as boundary conditions.

[0169] Specifically, 1. Region division: In S510, the entire initial flow field is divided into four regions, namely, region A, region B, region C, and region D. Such division takes into account the inside and outside of the bubble of the underwater explosion and the influence of different waves, making the simulation more detailed.

[0170] 2. Obtaining initial conditions: In S520, the initial conditions of the four regions are obtained. The initial conditions of regions A and B are provided by the underwater explosion Riemann solver, which takes into account the properties and initial state of the internal bubbles. The initial conditions of region C are calculated based on the Tamman state equation, while the initial conditions of region D are the initial conditions of the still water region, which are usually the pressure values ​​of still water. Such detailed initial condition settings help to more accurately reflect the characteristics of different regions.

[0171] 3. Setting boundary conditions: In S530, at the bubble surface, that is, at the junction of area B and the outer area, the conditions of pressure continuity and velocity equality are used as boundary conditions. Such boundary condition setting helps to simulate the physical behavior of the bubble surface, ensures a smooth transition between adjacent areas, and promotes the stability and accuracy of the simulation.

[0172] Furthermore, in S600, for equation (2), the discontinuous Galerkin method, namely the DG method, is used for spatial discretization, namely:

[0173]

[0174] In terms of time, the fourth-order Runge-Kutta method, namely the RK method, is used for discretization, thus forming the dynamic stretching coordinate system RKDG algorithm.

[0175] Specifically, in S600, the discontinuous Galerkin method (DG method) is used to discretize the space of equation (2), and the fourth-order Runge-Kutta method (RK method) is used to discretize the time, and the RKDG algorithm for the dynamic stretching coordinate system is successfully established. The specific numerical solution method is as follows:

[0176] 1. Spatial discretization (DG method): For equation (2), the discontinuous Galerkin method is used for spatial discretization. The DG method is a high-order finite element method that divides the calculation area into non-overlapping units and introduces numerical fluxes between units, so that different polynomial interpolations can be used on different units. This helps to capture the gradient and discontinuity of the solution more accurately and improves the accuracy of spatial discretization.

[0177] 2. Time discretization (RK method): In terms of time, the fourth-order Runge-Kutta method is used for discretization. The RK method is a commonly used numerical integration method that can more accurately approximate changes in the time direction through multiple stages of calculation. Using the fourth-order RK method means that four calculations are required for each time step, which improves the accuracy of time changes.

[0178] Combining spatial discretization with time discretization forms the dynamic stretching coordinate system RKDG algorithm. This numerical solution method combines high-order spatial interpolation and accurate time approximation, making the simulation of underwater explosion problems more accurate and stable. The introduction of this algorithm helps to improve the numerical stability and convergence of the simulation, making the numerical solution of underwater explosion problems more reliable.

[0179] A storage medium stores a computer program, which, when executed by a processor, implements the above-mentioned high-precision calculation method for underwater explosion shock waves without numerical dissipation based on a dynamic stretching coordinate system.

[0180] Specifically, 1. High-precision calculation method: This computer program implements a numerical calculation method for underwater explosion shock waves based on a dynamic stretching coordinate system. The discontinuous Galerkin method (DG method) is used for spatial discretization, combined with the fourth-order Runge-Kutta method (RK method) for time discretization. This combination provides a high-precision calculation method. The DG method can handle gradients and discontinuities, while the RK method can provide accurate time approximation, which together improves the accuracy of the simulation.

[0181] 2. Dynamic stretching coordinate system: By introducing the dynamic stretching coordinate system, the method can more flexibly adapt to the propagation of underwater explosion shock waves under different conditions. This helps to simulate more complex and realistic scenarios and improves the applicability and accuracy of the method.

[0182] 3. Reduced numerical dissipation: By using high-precision numerical methods, especially the combination of the DG method and the RK method, it is expected to reduce numerical dissipation and improve the stability of the simulation. This is crucial for simulating complex physical phenomena such as underwater explosion shock waves.

[0183] 4. Storage medium for practical application: The computer program is stored on a practical storage medium, which shows that this high-precision calculation method can be effectively applied on practical computing devices. This helps to more accurately simulate the behavior of underwater explosion shock waves in engineering and scientific fields.

[0184] The following is a specific embodiment of the invention:

[0185] Reference Figure 1-Figure 6As shown, the present invention includes: an RKDG algorithm in a dynamic stretching coordinate system for the outer flow field of the bubble, an RKDG algorithm in a dynamic stretching coordinate system for the inner flow field of the bubble, and an underwater explosion Riemann solver connecting the two flow fields. The inner flow field of the bubble generated by the underwater explosion is gas, and the outer flow field of the bubble is liquid water. The two flow fields have different media, different physical quantities, and different dynamic stretching coordinate systems, so an underwater explosion Riemann solver is needed to solve the flux between the two coordinate systems. The present invention aims to solve the technical problems that the traditional fixed coordinate system often brings numerical dissipation and non-physical oscillations to the numerical simulation, and the calculation errors caused by the traditional assumption of uniform internal pressure of bubbles in bubble dynamics, resulting in the numerical simulation of shock waves of near-field underwater explosions being greatly affected by bubbles, and the calculation results of far-field underwater explosions are somewhat different from the experimental results. In order to solve the above technical problems, the present invention will adopt the following technical solutions:

[0186] 1. The flow field is divided into inner zone, outer zone and undisturbed zone according to the position of bubble surface contact discontinuity and shock wave strong discontinuity;

[0187] 2. Transform the control equations of the inner and outer zones into the dynamic stretching coordinate system;

[0188] 3. Use the two-phase HLLC Riemann solver to calculate the bubble surface flux and realize the coupling between the inner and outer zones;

[0189] 4. The Hugoniot condition is used to calculate the flux at the discontinuity of the shock wave to achieve coupling between the outer zone and the undisturbed zone;

[0190] 5. Obtain initial conditions based on the improved HLLC Riemann solver;

[0191] 6. Use high-precision discontinuous Galerkin method to discretely solve the inner and outer regions.

[0192] In step 1, we first divide the entire flow field into three major areas, namely the inner area, the outer area, and the undisturbed area. This division is for the entire calculation process, which is convenient for distinguishing the two dynamic stretching coordinate systems inside and outside the bubble. Among them, the inner area refers to the space inside the bubble of the underwater explosion. The outer area refers to the space outside the bubble swept by the shock wave. The undisturbed area refers to the space outside the bubble that has not been swept by the shock wave, that is, the still water area. See the attached figure for specific area division.

[0193] In step 2, we need to transform the fixed coordinate system into the dynamic stretching coordinate system. Both the fixed coordinate system and the dynamic stretching coordinate system are spherical coordinate systems. For the underwater explosion of spherical TNT explosives, we assume that all the components are isotropic. So we only focus on the radial motion of the spherical shock wave and bubbles generated by the underwater explosion of spherical TNT explosives. Therefore, we let

[0194]

[0195] Among them, ζ is the radial coordinate of the dynamic stretching coordinate system, r is the radial coordinate of the fixed spherical coordinate system, and r 1 is the coordinate of the fixed coordinate system corresponding to the left endpoint of the dynamic stretching coordinate system, r 2 It is the coordinate of the fixed coordinate system corresponding to the right endpoint of the dynamic stretching coordinate system.

[0196] For the flow field inside the bubble, r 1 =0, r 2 =R. r 1 represents the explosion center, i.e., the bubble center, r 2 represents the bubble surface, R is the bubble radius. For the flow field outside the bubble, r 1 =R, r 2 =R S . r 1 represents the bubble surface, r 2 Indicates the interruption of the shock wave, R s is the distance from the shock wave to the explosion center. Thus, two dynamic stretching coordinate systems are formed, one is the dynamic stretching coordinate system of the flow field inside the bubble, and the other is the dynamic stretching coordinate system of the flow field outside the bubble. Their r 1 and r 2 Different sizes represent different meanings.

[0197] Then we substitute equation (1) into the one-dimensional spherical coordinate system Euler equation in the fixed coordinate system. After a series of processing (the specific process is described in the specific implementation method below), we obtain the conservation equation of the one-dimensional spherical coordinate system Euler equation in the dynamic stretching coordinate system:

[0198]

[0199] Where U = (ρ, ρu, E) T is the conserved quantity of the conservation formula, which is the same as the conserved quantity of the fixed coordinate system. is the numerical flux in the dynamic stretching coordinate system,

[0200] is the source intensity in the dynamic stretching coordinate system.

[0201] In step 3, we improved the HLLC Riemann solver into an underwater explosion Riemann solver. The HLLC Riemann solver is a solver used to solve two-phase flow problems. The two-phase flow here refers to conventional gas and water, which can be described using the Tamman state equation. However, for underwater explosion problems, the gas inside the bubbles generated is an explosion gas, which cannot be described using the Tamman state equation and needs to be described using the JWL underwater explosion state equation. Therefore, we improved the HLLC Riemann solver into an underwater explosion Riemann solver. The specific process is described in the specific implementation method below.

[0202] In step 4, we use the Hugoniot condition to calculate the flux at the shock wave discontinuity, and use the static water condition of the undisturbed zone as the boundary condition at the shock wave discontinuity to achieve the mutual coupling between the external zone and the undisturbed zone.

[0203] In step 5, in order to describe the initial conditions of the entire flow field, we divide the entire initial flow field into four regions, namely, region A, region B, region C, and region D. Region A and region B both belong to the inner region. This division is for the initial process, which is convenient for determining the initial conditions of different regions. Region A refers to the space inside the bubble of the underwater explosion that is not swept by the second shock wave. The second shock wave is also called the inward rarefaction wave. The underwater explosion will instantly produce an outward shock wave, namely the shock wave, and an inward rarefaction wave, namely the second shock wave. Region B refers to the space inside the bubble of the underwater explosion that is swept by the second shock wave. Region C (outer region) refers to the space outside the bubble that is swept by the shock wave. Region D (undisturbed region) refers to the space outside the bubble that is not swept by the shock wave, namely the still water region. See the attached figure for specific region division. Next, we use the underwater explosion Riemann solver to solve the initial conditions after the region is divided, as well as the boundary conditions at the bubble surface (i.e., the junction of region B and the outer region). The boundary conditions on the bubble surface are continuous pressure and equal velocity. The pressure in the inner zone is calculated using the JWL underwater explosion state equation, the pressure in the outer zone is calculated using the Tamman state equation, and the pressure in the undisturbed zone is the hydrostatic pressure.

[0204] In step 6, for equation (2), we use the discontinuous Galerkin method (DG method) for spatial discretization, that is:

[0205]

[0206] In terms of time, we use the fourth-order Runge-Kutta method (RK method) for discretization, thus forming our dynamic stretching coordinate system RKDG algorithm.

[0207] The following will combine the content of the invention and the accompanying drawings to introduce the RKDG algorithm of the dynamic stretching coordinate system in detail, specifically and comprehensively. In the space of the entire flow field of the underwater explosion, we use the discontinuous Galerkin method to separate the inside of the bubble (inner zone) and the outside of the bubble (outer zone) into 100 grids. In the initial stage, the inner zone is divided into zone A and zone B. Zone A is a total of 99 units from left to right starting from the center of the bubble. Zone B is the 100th unit from left to right starting from the center of the bubble, and the right boundary of this unit is the bubble surface. The outer zone is 100 units from the bubble surface to the shock wave, and the right boundary of the last unit is the shock wave discontinuity.

[0208] Next, we derive the basic equations of the dynamic stretching coordinate system. We know that the conservation equation of the one-dimensional Euler equation in the spherical coordinate system under the fixed coordinate system is:

[0209]

[0210] Among them, U=(u,ρu,E) T is the conservation quantity of the one-dimensional Euler equation in the spherical coordinate system under the fixed coordinate system, F(U)=(ρu,ρu 2 +p,u(E+p)) T is the numerical flux, which is a function of U. S=-2(ρu,ρu 2 ,u(E+p)) T / r is the source intensity. ρ in their elements is the density of the fluid particle, u is the radial velocity of the fluid particle in the spherical coordinate system, and p is the pressure at the fluid particle. E = 0.5ρu 2 +ρe is the total energy per unit volume of the fluid.

[0211] We take the derivative of equation (1) with respect to time t and obtain the velocity relationship between the dynamic stretching coordinate system and the fixed coordinate system:

[0212]

[0213] u is the velocity in the fixed coordinate system. We can easily know that the velocity in the dynamic stretching coordinate system is So formula (5) becomes:

[0214]

[0215] Next, we can solve the velocity in the dynamic stretching coordinate system based on the above formula:

[0216]

[0217] Then we solve the material derivatives of the conserved quantity U in the fixed coordinate system and the dynamically stretched coordinate system respectively:

[0218]

[0219] We subtract equation (9) from equation (8), and then combine equation (1) and equation (6) to obtain:

[0220]

[0221] Finally, by substituting equation (10) into equation (4), we can obtain equation (2) mentioned above, which is the conservation equation of the spherical coordinate one-dimensional Euler equation in the dynamic stretching coordinate system. The dynamic stretching coordinate systems of the inner and outer regions are because their

[0222] Next, we introduce the state equation of the dynamic stretching coordinate system RKDG algorithm of the present invention. For the bubble surface and the bubble interior, we use the JWL state equation to calculate the pressure on the bubble surface and inside the bubble. The JWL equation describing the pressure inside and on the surface of the bubble at the initial stage of underwater explosion is:

[0223]

[0224] in:

[0225]

[0226]

[0227] For the flow field outside the bubble, we choose the Tamman state equation to describe the pressure of the flow field:

[0228] p=ρe(γ-1)-γP w (14)

[0229] Where p is the pressure of the flow field outside the bubble, ρ is the density of the flow field outside the bubble, γ is the insulation coefficient (taken as 7.5), P w is a constant (taken as 3.309GPa).

[0230] Different explosives correspond to different explosive parameters. The spherical explosive parameters of the numerical model of the underwater explosion bubble external flow field in this paper are selected from the underwater explosion parameters of TNT explosives, and the water parameters are selected from the Tamman state equation parameters. The specific parameters of the Tamman state equation and the JWL state equation of underwater explosion TNT explosives are shown in Table 1.

[0231] Table 1 Main parameters of Tamman state equation and JWL state equation for underwater TNT explosive

[0232]

[0233] Next, we introduce the initial conditions and boundary conditions of the algorithm. In order to calculate reasonable initial conditions and boundary conditions, we need to use the underwater explosion Riemann solver. The underwater explosion Riemann solver is an improved solver based on the HLLC Riemann solver. There is gas on the left side (inside) of the bubble surface, and water on the right side (outside) of the bubble surface. An inward rarefaction wave is formed inside the bubble surface, and an outward shock wave is formed outside the bubble. This is a classic two-phase flow problem. The difficulty lies in the fact that the gas inside the bubble is an explosion gas, and it is not appropriate to use the state equation under adiabatic conditions. In order to better study this two-phase flow problem, we study the initial moment t = ε, that is, at this moment, the underwater explosion has occurred and bubbles have formed, and rarefaction waves and shock waves are formed on the bubble surface and are about to propagate. At this time, t is very small, almost close to 0.

[0234] The Riemann problem for underwater explosions is described as follows: Two fluids are separated by a bubble surface. Inside the bubble, the high-pressure explosion gas products are initially at rest. Outside the bubble, the undisturbed water is also at rest. The initial conditions inside and outside the bubble, i.e., the conserved quantities, are respectively given by U L and U R In addition, they are described using the JWL equation of state and the Tammann equation of state, respectively. The rarefaction wave (second shock wave) velocity is denoted by S L , the velocity of the bubble surface is S * , the shock wave speed is S R . Then, we have

[0235]

[0236] where u * is the initial average velocity of the bubble surface in a short time, is the initial average velocity of the first element on the left side of the bubble surface, is the initial average velocity of all cells to the right of the bubble surface. p * is the average pressure on the bubble surface over a short period of time. is the initial average pressure of the first cell to the left of the bubble surface, is the initial average pressure of all cells on the right side of the bubble surface. Then, using the Amemiya condition, we get the relationship between the shock wave pressure and wave velocity:

[0237]

[0238] Among them, P 0 is the hydrostatic pressure, C R =C 0 is the speed of sound in still water. Using the velocity relation of the HLLC Riemann solver, we have:

[0239]

[0240] Combining (18) and (19), we get

[0241]

[0242] Among them, ρ L is the initial density of area A, i.e., the density of the explosion gas, ρ R is the initial density of the undisturbed zone, i.e., the still water density, as shown in Table 1. We calculate the negative value of the average value of the sound speed in the inner zone and zone B as the rarefaction wave speed S L ,Right now

[0243]

[0244] The sound speed in the inner zone and zone B is calculated by the following formula:

[0245]

[0246] The specific parameters are shown in Table 1. For the density of area B, we calculate it by mass conservation before and after the explosion:

[0247]

[0248] where h L is the actual space length of zone B when the explosion occurs at time t = ε, h L -εu * This is the actual spatial length of zone B before the explosion.

[0249] The actual space length h of zone B when the explosion occurs at time t = ε L is the length of space swept by the rarefaction wave and bubble surface, so we have:

[0250]

[0251] Among them, R c is the initial radius of the spherical TNT explosive. N = 100 is the number of grids. L is the actual spatial length of area B, which is the 100th unit from the left to the right starting from the center of the bubble. Therefore, at the initial time t = ε, h L It is the actual length of a unit.

[0252] At this point, the improved Riemann solver is set up. By combining equations (21), (22), (23), (24) and (25), we can obtain S L The value of S is then calculated by formula (20) R However, S is directly solved by solving the system of equations L It is not simple, so we can use the following two methods to solve S L To solve. The first method is to estimate C L By directly solving equation (22), we can directly calculate the first S L , then we put the first S L Substituting into equation (24), and combining equation (23) and equation (25) in parallel, we get the first Put the first Substituting into equation (22), we get the second Then we can calculate a more accurate S through formula (21) L In the second method, we iterate through the first method multiple times until S LWhen S is almost no longer changing, L is very close to the correct value. Solve for S L Then, we calculate S through formula (20) R The value of . Formula (20) is a complex cubic equation with one variable. The expression of its root is quite complicated, so it needs to be solved by software.

[0253] In calculating S R After we find the value of , we can use equations (15), (16), (17) and (18) to solve for the initial pressure of area B: and the initial pressure in the outer zone and the initial velocity of region B and the initial velocity of the outer zone The initial bubble radius and shock wave radius (the distance from the shock wave to the explosion point, i.e. the center of the bubble) can also be obtained:

[0254] R=R C +εu * (26)

[0255] R S =R C +εS R (27)

[0256] Therefore, we can obtain the initial conditions of the entire flow field at the explosion time t = ε:

[0257]

[0258] U L is the initial conserved quantity of region A, that is, the initial condition. is the initial condition of region B, is the initial condition of region C (outer region), U R is the initial condition of the undisturbed region.

[0259] The calculation of the numerical flux of the present invention uses two methods, one is the numerical flux calculation method using the LF form, and the other is the numerical flux calculation method using the Riemann problem solver. The numerical flux in the LF form is a numerical flux calculation format commonly used for solving hyperbolic conservation law equations. The full name of the numerical flux in the LF form is the Lax-Friedrich flux. The numerical flux calculation in the LF form is relatively simple. First, the value of the physical quantity of the right boundary of the left unit and the value of the physical quantity of the left boundary of the right unit must be calculated, and the left flux is calculated according to the value of the physical quantity of the right boundary of the left unit, and the right flux is calculated according to the value of the physical quantity of the left boundary of the right unit. After obtaining the left flux and the right flux, the average flux is calculated. After calculating the average flux, a correction term is added to obtain the LF numerical flux between the two units. The correction term is the wave velocity at the LF numerical flux multiplied by the difference in the transport of the two units. The famous LF numerical flux calculation formula is:

[0260]

[0261] Among them, f(u - ) is the left flux, f(u + ) is the right flux, u + is the right transport, u - is the left transport. α is the wave speed, which is calculated as:

[0262]

[0263] Finally, we set the boundary conditions. We know that the general form of the dynamic stretching coordinate coefficient flux is For the flow field inside the bubble, we know from equation (7) that the velocity at the explosion center in the dynamic stretching coordinate system is u=0. We believe that there is no gas particle at the center of the underwater explosion of a spherical explosive, which is an absolute boundary point. The velocity at this point is 0, and the pressure is equal to the left boundary pressure of the unit closest to the explosion center. Therefore, we can get the numerical flux of the underwater explosion center as:

[0264]

[0265] After considering the influence of the internal flow field of the bubble on the external flow field of the bubble, the bubble surface boundary will become a two-phase flow Riemann discontinuity problem with underwater explosion gas on one side and water on the other. u=u * , we can get the numerical flux of the bubble surface boundary in the dynamic stretching coordinate system as:

[0266]

[0267] The internal and external flow fields of underwater explosion bubbles1 and r 2 Because the two dynamic stretching coordinate systems on the left and right sides of the underwater explosion bubble surface boundary are different, it is necessary to calculate the numerical flux of the two units in the dynamic stretching coordinate system separately. For the flow field inside the bubble, r 1 =0, r 2 = R. So we get the numerical flux at the bubble surface in the dynamic stretching coordinate system of the bubble internal flow field:

[0268]

[0269] Similar to the dynamic stretching coordinate coefficient value flux of the bubble internal flow field, for the bubble external flow field, r 1 =R, r 2 =R S So we get the numerical flux at the bubble surface in the dynamic stretching coordinate system of the bubble external flow field:

[0270]

[0271] From equation (7), we know that the velocity of the outward shock wave discontinuity is u=0 (because the outside of the shock wave discontinuity is still water). We get the numerical flux at the shock wave discontinuity in the dynamic stretching coordinate system:

[0272]

[0273] At this point, the initial conditions, boundary conditions and state equations of the RKDG algorithm of the dynamic stretching coordinate system of the entire flow field of the underwater explosion have been set up, forming the internal numerical model of the underwater explosion bubble and the external numerical model of the underwater explosion bubble, which are connected through the improved Riemann solver.

[0274] After connecting two underwater explosion RKDG numerical models using a one-dimensional improved Riemann solver, negative physical quantities appeared in the numerical model inside the bubble, especially negative density and negative square of the speed of sound. This led to calculation errors after the Fortran code of the entire RKDG numerical model ran for thousands of steps. Therefore, in order to solve this problem, we need to use a positive limiter to ensure that the physical quantities at the numerical flux and the integration point are positive. The main idea of ​​the positive limiter is to limit the weight function, that is, the coefficient of the basis function, of the unit where negative physical quantities appear, so that the weight function of the unit is equal to the average value of the physical quantity of the unit. In the actual code writing, we set a conditional statement so that the weight function of the unit that meets the condition is equal to the average value of the physical quantity of the unit, that is, replace the original erroneous weight function. The average value of the physical quantity of the unit is calculated as:

[0275]

[0276] Among them, U=(u 1 ,u 2 ,u 3 )=(ρ,ρu,E) T , r is the distance from the unit to the explosion center. When using Gauss-Legendre integration, r is the distance from the integration point to the explosion center. l is the distance from the left edge of the unit to the explosion center, R r is the distance from the right edge of the cell to the center of the explosion.

[0277] Next, we use the discontinuous Galerkin method to perform spatial separation on the conservation equations of the spherical coordinate one-dimensional Euler equations in the dynamically stretched coordinate system. We first convert the conservation quantity U = (u 1 ,u 2 ,u 3 )=(u,ρu,E) T Separation is performed as in equation (3). Where n = 1, 2, 3, corresponding to the independent variables (u, ρu, E) of the three differential equations in the dynamic stretching coordinate Euler equation conservation equation. M is the calculation accuracy order of the discontinuous Galerkin method, and M = 3 is taken in this paper. k is the weight function, which is only related to time t. Determined by initial conditions. is a basis function, which is only related to the spatial position. The basis function selected in this paper is the Legendre polynomial with a computational domain of [-0.5, 0.5]. The Legendre polynomial with a computational domain of [-0.5, 0.5] of third-order accuracy is:

[0278]

[0279] in [-0.5,0.5] is the computational domain of Legendre polynomials, not the spatial domain of discontinuous Galerkin. The spatial domain of discontinuous Galerkin is [0,1], so coordinate transformation is required.

[0280] Therefore, we multiply both sides of the dynamic stretching coordinate Euler equation, that is, equation (2), by the basis function and integrate it over [-0.5, 0.5], and then perform distributed integration to obtain:

[0281]

[0282] Where n = 1, 2, 3. dζ = 1 / N , N is the number of spatial grids, that is, the solution domain is divided into N segments. dη=0.5-(-0.5)=1. (u n ) t is u n The partial derivative with respect to time t is yes Take the spatial partial derivative of equation (2-24) with respect to x on the computational domain [-0.5,0.5]. It is the difference between two adjacent numerical fluxes multiplied by the basis function value on the unit boundary. For continuous places in the spatial solution domain, directly use formula (3) to calculate the independent physical quantity and then calculate the numerical flux. For discontinuous places in the spatial solution domain, such as the discontinuity of the outward shock wave of the bubble external flow field and the bubble boundary, it is necessary to solve the Riemann problem, that is, to obtain the physical quantity by solving the boundary conditions and then calculate the numerical flux at the discontinuity. The integral term in formula (38) is calculated using the Gauss-Legendre integral method. The Gauss-Legendre integral method is a method of discretizing the integral into nodes multiplied by function values ​​for calculation. The nodes are the zeros of the Legendre polynomials in the computational domain [-0.5, 0.5]. Using the orthogonality of the basis function (Legendre polynomials are orthogonal polynomials), formula (3) is further separated to obtain the derivative of the weight function with respect to time t, and the following is obtained:

[0283]

[0284] The first M on the right side of the equation 0 is the mass matrix, It is only related to the selected basis function and integral calculation domain. This means that as long as the Legendre polynomial is selected as the basis function and the calculation domain is fixed to [-0.5, 0.5], no matter what problem is solved by using the discontinuous Galerkin method, the mass matrix M 0 The mass matrix M 0 The latter part is called rhs. The left side of equation (39) is the column vector of the derivative of the weight function of the nth independent variable with respect to time t. The more elements in the column vector represent the higher the degree of discreteness, that is, the greater the order of accuracy. After calculating the derivative of the weight function of the independent variable with respect to time t through equation (39), the value of the independent variable at the next time step can be obtained by performing the fourth-order Runge-Kutta method.

[0285] The present invention uses the fourth-order Runge Kutta method for the time discretization of the Euler equations in the dynamic stretching coordinate system. The Runge Kutta method is a numerical method for time discretizing the conservation formula of differential squares or differential equations in the solution domain. The main idea of ​​the Runge Kutta method is to iteratively calculate the physical quantities of all subsequent time steps through the time derivatives of the physical quantities in the differential equations within a very small time step. The main principle of the fourth-order Runge Kutta method is to perform weighted calculations in four parts within a time step, and the final weighted average is used as the physical quantity of the next time step. The fourth-order Runge Kutta method in this article is used to calculate the weight function related to time. The weighted average in each time step is used as the weight function of the next time step. Finally, the physical quantity at any time of the spatial discrete point can be obtained by multiplying the weight function at each time step by the basis function and then superimposing it.

[0286] Finally, we used the dynamic stretching coordinate system RKDG algorithm of the present invention and LS-dyna software to calculate the change of the measuring point at 4m of a 7.0kg spherical TNT explosive at a water depth of 6m over time, and compared it with the experimental results, obtaining significant results.

[0287] Although the present invention has been disclosed as above in the preferred embodiment, it is not intended to limit the present invention. Anyone familiar with this technology can make various changes and modifications without departing from the spirit and scope of the present invention. Therefore, the scope of protection of the present invention should be based on the definition of the claims.

Claims

1. A high-precision calculation method for underwater explosion shock waves without numerical dissipation based on a dynamic stretching coordinate system. It is characterized in that The underwater explosion shock wave non-numerical dissipation high-precision calculation method based on a dynamic stretching coordinate system comprises the following steps: S100, dividing the flow field into an inner zone, an outer zone, and an undisturbed zone according to the bubble surface contact discontinuity and the shock wave intensity discontinuity positions; S200, respectively transforming the control equations of the inner zone and the outer zone into a dynamic stretching coordinate system; S300, the improved HLLC Riemann solver is the underwater explosion Riemann solver, and in the dynamic stretching coordinate system, the two-phase improved HLLC Riemann solver is used to calculate the bubble surface flux to achieve the coupling between the inner and outer zones; S400, in the dynamic stretching coordinate system, the flux at the shock wave discontinuity is calculated using the Hugoniot condition to achieve coupling between the outer zone and the undisturbed zone; S500, initial conditions are obtained based on the improved HLLC Riemann solver; S600, using high-precision discontinuous Galerkin method to discretely solve the inner and outer regions; In S200, the fixed coordinate system is converted into a dynamic stretching coordinate system. Both the fixed coordinate system and the dynamic stretching coordinate system are spherical coordinate systems. For the underwater explosion of spherical TNT explosives, it is assumed that all items are isotropic and only the radial motion of the spherical shock wave and bubbles generated by the underwater explosion of the spherical TNT explosives is concerned. Among them, ζ is the radial coordinate of the dynamic stretching coordinate system, r is the radial coordinate of the fixed spherical coordinate system, and r 1 is the coordinate of the fixed coordinate system corresponding to the left endpoint of the dynamic stretching coordinate system, r 2 is the coordinate of the fixed coordinate system corresponding to the right endpoint of the dynamic stretching coordinate system, For the flow field inside the bubble, r 1 =0, r 2 =R, r 1 represents the explosion center, i.e., the bubble center, r 2 represents the bubble surface, R is the bubble radius, and for the flow field outside the bubble, r 1 =R, r 2 =R S , r 1 represents the bubble surface, r 2 Indicates the interruption of the shock wave, R s is the distance from the shock wave to the explosion center, thus forming two dynamic stretching coordinate systems, one is the dynamic stretching coordinate system of the flow field inside the bubble, and the other is the dynamic stretching coordinate system of the flow field outside the bubble. Their r 1 and r 2 Different sizes represent different meanings; Substituting equation (1) into the one-dimensional spherical coordinate system Euler equation in the fixed coordinate system, after a series of processing, the conservation equation of the one-dimensional spherical coordinate system Euler equation in the dynamic stretching coordinate system is obtained: Where U = (ρ, ρu, E) T is the conserved quantity of the conservation formula, which is the same as the conserved quantity of the fixed coordinate system. is the numerical flux in the dynamic stretching coordinate system, is the source intensity in the dynamic stretching coordinate system; In S500, the following steps are included: S510, performing area division, dividing the entire initial flow field into four areas, namely area A, area B, area C and area D, wherein: Area A represents the interior of the bubble of the underwater explosion. This area is not swept by the second shock wave, that is, not swept by the inward rarefaction wave. It is part of the inner zone and belongs to the inner zone. Area B represents the interior of the bubble of the underwater explosion. This area is swept by the second shock wave and also belongs to the inner area. Zone C represents the outside of the bubble of the underwater explosion, which is swept by the shock wave and belongs to the outer zone; Area D represents the outside of the bubble of the underwater explosion, which is not swept by the shock wave and belongs to the calm water area; S520, obtaining initial conditions of four regions, where: The initial conditions of regions A and B are provided by the underwater explosion Riemann solver, which considers the properties and initial state of the internal bubbles; The initial conditions in region C were calculated based on the Tamman equation of state; The initial conditions in region D are those in the still water region, usually the pressure value of still water; The initial conditions include velocity, pressure and density; S530. At the bubble surface, that is, at the junction of area B and the outer area, the conditions of pressure continuity and velocity equality are used as boundary conditions.

2. According to claim 1, a high-precision calculation method for underwater explosion shock wave without numerical dissipation based on a dynamic stretching coordinate system, It is characterized in that In S100, the inner zone is the space inside the bubble of the underwater explosion, the outer zone is the space outside the bubble swept by the shock wave, and the undisturbed zone is the space outside the bubble that is not swept by the shock wave, that is, the still water area.

3. According to claim 1, a high-precision calculation method for underwater explosion shock wave without numerical dissipation based on a dynamic stretching coordinate system, It is characterized in that The series of processing is: The equation (1) is derived with respect to time t, and the velocity relationship between the dynamic stretching coordinate system and the fixed coordinate system is obtained: u is the velocity in the fixed coordinate system. It is known that the velocity in the dynamic stretching coordinate system is So the above formula (5) becomes: Next, the velocity in the dynamic stretching coordinate system is solved according to the above formula: Then solve the material derivatives of the conserved quantity U in the fixed coordinate system and the dynamic stretching coordinate system respectively: Subtract equation (9) from equation (8), and then combine equation (1) and equation (6) to obtain: Finally, equation (2) is obtained through equation (10), which is the conservation equation of the spherical coordinate one-dimensional Euler equation in the dynamic stretching coordinate system.

4. According to claim 3, a high-precision calculation method for underwater explosion shock wave without numerical dissipation based on a dynamic stretching coordinate system, It is characterized in that In S300, the following steps are included: S310, defining an initial moment, assuming that at the initial moment the explosion has occurred and a bubble has been formed, and a rarefaction wave and a shock wave have been formed on the bubble surface and are about to propagate; S320. Clarify the description of the Riemann problem, including the state description of two fluids and the separation condition of the two fluids, wherein the two fluids include explosion gas and water; S330, determining the initial conditions inside and outside the bubble, and calculating and defining the pressure, velocity and density inside and outside the bubble; S340, using Hugoniot conditions to determine the relationship between the shock wave's output pressure and wave velocity; S350, using the velocity relationship of the basic HLLC Riemann solver, improving the velocity relationship, and obtaining the wave velocities of the rarefaction wave and the shock wave based on the velocity relationship; S360. Obtain the final improved Riemann solver.

5. According to claim 4, a high-precision calculation method for underwater explosion shock wave without numerical dissipation based on a dynamic stretching coordinate system, It is characterized in that In S300, specifically, The underwater explosion Riemann solver is an improved solver based on the HLLC Riemann solver. The inside of the bubble is gas, and the outside of the bubble is water. An inward rarefaction wave is formed inside the bubble surface, and an outward shock wave is formed outside the bubble. It is a classic two-phase flow problem. At the initial moment t = ε, it is assumed that at this moment, the underwater explosion has occurred and a bubble has formed, and the rarefaction wave and shock wave are formed on the bubble surface and are about to propagate. At this time, t is close to 0. The Riemann problem for underwater explosions is described as follows: the two fluids are separated by the bubble surface. Inside the bubble, the high-pressure explosion gas products are initially at rest, and outside the bubble, the undisturbed water is also at rest. The initial conditions inside and outside the bubble, i.e., the conserved quantities, are respectively given by U L and U R Given that, in addition, U L and U R The JWL state equation and the Tammann state equation are used to describe the rarefaction wave, i.e. the second shock wave, respectively. The wave velocity is denoted as S L , the velocity of the bubble surface is S * , the shock wave speed is S R , so there is Among them, u * is the initial average velocity of the bubble surface in a short time, is the initial average velocity of the first element on the left side of the bubble surface, is the initial average velocity of all cells to the right of the bubble surface, p * is the average pressure on the bubble surface over a short period of time. is the initial average pressure of the first cell to the left of the bubble surface, is the initial average pressure of all cells on the right side of the bubble surface. Then, using the Hugoniot condition, we get the relationship between the shock wave pressure and wave velocity: Among them, P 0 is the hydrostatic pressure, C R =C 0 is the speed of sound in still water. Using the velocity relationship of the HLLC Riemann solver, we have: Combining (18) and (19), we get Among them, ρ L is the initial density of area A, i.e., the density of the explosion gas, ρ R is the initial density of the undisturbed zone, i.e., the still water density. The negative value of the average value of the sound speed in the inner zone and zone B is calculated as the rarefaction wave speed S L ,Right now The sound speed in the inner zone and zone B is calculated by the following formula: The density of area B is calculated by the conservation of mass before and after the explosion: where h L is the actual space length of zone B when the explosion occurs at time t = ε, h L -εu * That is the actual space length of area B before the explosion. The actual space length h of zone B when the explosion occurs at time t = ε L is the length of space swept by the rarefaction wave and bubble surface, so: Among them, R c is the initial radius of the spherical TNT explosive, N = 100 is the number of grids, because h L is the actual spatial length of area B, which is the 100th unit from the left to the right starting from the center of the bubble. Therefore, at the initial time t = ε, h L is the actual length of a unit. At this point, the improved Riemann solver is set up.

6. According to claim 5, a high-precision calculation method for underwater explosion shock wave without numerical dissipation based on a dynamic stretching coordinate system, It is characterized in that In S400, the flux at the shock wave discontinuity is calculated using the Hugoniot condition, and the static water condition of the undisturbed zone is used as the boundary condition at the shock wave discontinuity to achieve mutual coupling between the outer zone and the undisturbed zone.

7. The high-precision calculation method for underwater explosion shock wave without numerical dissipation based on a dynamic stretching coordinate system according to claim 6, It is characterized in that In S600, for equation (2), the discontinuous Galerkin method, namely, the DG method, is used for spatial discretization, namely: In terms of time, the fourth-order Runge-Kutta method, namely the RK method, is used for discretization, thus forming the dynamic stretching coordinate system RKDG algorithm.

8. A storage medium having a computer program stored thereon, It is characterized in that When the computer program is executed by a processor, a high-precision calculation method for underwater explosion shock waves without numerical dissipation based on a dynamic stretching coordinate system as described in any one of claims 1 to 7 is implemented.