Method for forecasting dynamic buckling characteristics of reinforced cylindrical shell with initial defects under underwater explosion load based on artificial intelligence

Abstract

The invention provides a prediction method for dynamic buckling characteristics of a reinforced cylindrical shell with initial defects under an underwater explosion load based on artificial intelligence. The method specifically comprises the following steps: establishing a dynamic response characteristic three-dimensional numerical calculation model of initial defects of the reinforced cylindrical shell structure under an underwater explosion load; obtaining a rib and skin rigidity matching factor; calculating dynamic response characteristics of the reinforced cylindrical shell under different explosive equivalents, different explosion distances and different laying water depths; determining a dynamic buckling critical load and a buckling mode of the reinforced cylindrical shell under the underwater explosion load by combining a dynamic buckling criterion and annular and axial typical response characteristic space distribution characteristics; and establishing an artificial intelligence proxy model of the dynamic buckling characteristics of the reinforced cylindrical shell under the underwater explosion load. According to the method, different defect amplitudes and rib and skin rigidities under the underwater explosion shock wave and bubble pulsation coupling load can be rapidly given based on the proxy model, the dynamic buckling critical load and modality of the reinforced cylindrical shell under the water depth are laid, and the method has high practicability and engineering significance.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of dynamic buckling of cylindrical shells under underwater explosion load and artificial intelligence technology, in particular to a method for predicting the dynamic buckling characteristics of a cylindrical shell with initial defects under underwater explosion load based on artificial intelligence. BACKGROUND

[0002] Typical stiffened cylindrical shells such as submarines, underwater vehicles and submarine pipelines may fail due to dynamic buckling under underwater explosion load, resulting in instantaneous loss of load-carrying capacity. At the same time, initial geometric and thickness defects, which have a significant impact on the critical buckling load and buckling mode of the shell, are inevitable during manufacturing and transportation. Therefore, it is of great significance to predict the dynamic buckling characteristics of stiffened cylindrical shells under underwater explosion load while taking into account the initial defects of the shell for the optimization design and safety evaluation of underwater cylindrical shell structures. However, the actual initial defect distribution is complex, making it difficult to consider, underwater explosion tests are difficult to carry out on a large scale due to high cost and site requirements, and the buckling critical load and mode are obtained with low efficiency. In view of the above problems, the initial defects of the stiffened cylindrical shell are introduced by using reverse modeling technology and modal superposition method, and the artificial intelligence algorithm is combined to realize the rapid prediction of the dynamic buckling characteristics of the stiffened cylindrical shell with initial defects under underwater explosion load, which has extremely important engineering significance. SUMMARY

[0003] The present application aims to solve the problems of not considering the initial defects of the stiffened cylindrical shell, low efficiency of obtaining the critical buckling load and buckling mode in the prior art, and proposes a method for predicting the dynamic buckling characteristics of a stiffened cylindrical shell with initial defects under underwater explosion load based on artificial intelligence.

[0004] The present application is realized by the following technical solutions, the present application proposes a method for predicting the dynamic buckling characteristics of a stiffened cylindrical shell with initial defects under underwater explosion load based on artificial intelligence, the method comprises: Step 1, establishing a three-dimensional numerical calculation model of the dynamic response characteristics of the stiffened cylindrical shell with initial defects under underwater explosion load; Step 2, calculating the nonlinear static buckling critical load of the un-stiffened and stiffened cylindrical shell with initial defects under uniform external pressure load P sc,u 、 P sc,s , and obtaining the stiffness matching factor of the rib and the cylindrical shell skin K = P sc,u / P sc,s ; according to P sc,s , obtaining the water depth value range of the stiffened cylindrical shell H ​P sc,s / pg , where p represents the density of water, g represents the acceleration of gravity; Step 3, change the water depth of the stiffened cylindrical shell H , the equivalent of the charge W , the blast distance d , calculate the underwater explosion impact factor ; the dynamic response characteristics of the stiffened cylindrical shell with initial defects under underwater explosion load are calculated by using three-dimensional numerical method, and the maximum value of the response characteristics of each typical monitoring position of the stiffened cylindrical shell under underwater explosion load in the final deformation state is recorded X max ; Step 4, combine the dynamic buckling criterion and the impact factor at the typical monitoring position C and X max curve to determine whether dynamic buckling occurs, and the dynamic buckling critical load of the stiffened cylindrical shell is obtained C s ; at the same time, according to the spatial distribution characteristics of the hoop and axial typical response characteristics, the hoop and axial dynamic buckling modes m , n ) are obtained; Step 5, for the case that the spatial distribution of the initial defects is basically the same but the defect amplitude L is different, or the spatial distribution of the initial defects and the defect amplitude are the same but the number of rib strips is different, repeat steps 2-4 above for numerical calculation, and use artificial intelligence method to establish the proxy model of the dynamic buckling critical load C s , the hoop and axial buckling modes m , n ) and the initial defect amplitude L / stiffness matching factor K , the water depth H C s = f ( L, H ) / or C s = f ( K, H )、( m , n )= Y ( L, H ) / ( m , n )= Y ( K, H ), and according to the proxy model, the different L or different​K Reinforced cylindrical shells at different water depths H Dynamic buckling critical load C s 、 Circumferential and axial dynamic buckling modes ( m , n ).

[0005] Furthermore, the initial defects in step 1 include shape defects and thickness defects. For the processed stiffened cylindrical shell structure, a 3D scanner and a thickness gauge are used to scan the shape defects and thickness defects, and the measurement results are used to perform 3D imaging to obtain the spatial distribution characteristics of the defects. For the unprocessed stiffened cylindrical shell structure, the modal superposition method and the existing thickness defect distribution law are used to obtain the shape and thickness defects.

[0006] Furthermore, for both fabricated and unfabricated reinforced cylindrical shell structures, based on reverse modeling technology and modal superposition method to couple shape and thickness defects, a three-dimensional numerical calculation model of the dynamic response of the reinforced cylindrical shell under underwater explosive load, taking into account the initial defects of the structure, is established using the finite element method or meshless method.

[0007] Furthermore, in step 2, the arc-length method, which takes into account material nonlinearity, is used to calculate the critical nonlinear static buckling load of unreinforced and reinforced cylindrical shells under uniform external pressure load, taking into account initial defects. P sc,u , P sc,s This leads to the stiffness matching factor between the ribs and the cylindrical shell skin. K .

[0008] Furthermore, in step 3, when performing numerical calculations of the dynamic response characteristics of the stiffened cylindrical shell under underwater explosive loads, taking into account the initial defects, a single variable principle is adopted, i.e., the variable is fixed. W , d , H Two of these values ​​are used to calculate the critical load magnitude by determining the corresponding value of another physical quantity when dynamic buckling occurs.

[0009] Furthermore, in step 4, different typical measuring point locations are considered. C - X max The curves are used to determine whether dynamic buckling has occurred using the Budiansky-Roth dynamic buckling criterion or the Southwell method, thereby obtaining the critical load for dynamic buckling. Based on the circumferential and axial spatial distribution characteristics of typical response features at different measuring points of the stiffened cylindrical shell, the circumferential and axial dynamic buckling modes are obtained. m , n ).

[0010] Further, in step 5, the laying water depth is first obtained according to step 2 above H , and then the laying water depth of the stiffened cylindrical shell and the charge is fixed H , the charge equivalent W , the burst distance d or the laying water depth of the stiffened cylindrical shell and the charge H , the burst distance d , the charge equivalent W， , the above steps 3-4 are repeated to obtain different L / or K , and on this basis, the C s is changed and further different H / or L and K is obtained. H C s

[0011] Further, in step 5, the data characteristics are normalized and an artificial intelligence prediction model is established, the input layer parameters include L, H ) or H, K , and the output layer parameters include the critical load of dynamic buckling C s and the circumferential and axial dynamic buckling modes m , n , the artificial intelligence prediction model hyperparameters are obtained by an automatic parameter optimization method, and thus the critical load of dynamic buckling C s , the circumferential and axial dynamic buckling modes m , n of the stiffened cylindrical shell under underwater explosion load and the initial defect amplitude L / or the stiffness matching factor K , the laying water depth H are matched to establish a proxy model C s = f ( L, H ) / or C s = f ( K, H ), ( m , n )= Y ( L, H ) / or( m , n )= Y ( K, H ), and different L or different K ​​Reinforced cylindrical shells at different water depths H Dynamic buckling critical load C s Circumferential and axial dynamic buckling modes ( m , n ).

[0012] The beneficial effects of this invention are: This invention proposes an artificial intelligence-based method for predicting the dynamic buckling characteristics of stiffened cylindrical shells with initial defects under underwater explosive loads. Based on inverse modeling techniques and modal superposition methods, it incorporates the initial geometric and thickness defects of the stiffened cylindrical shell. An artificial intelligence surrogate model is used to rapidly predict the dynamic buckling critical load and modes of the stiffened cylindrical shell under coupled loads of underwater explosive shock waves and bubble pulsations, considering different defect amplitudes, stiffener and shell skin stiffness, and the dynamic buckling critical load at the specified water depth. In step 1, inverse modeling techniques and modal superposition methods are used to incorporate the initial geometric and thickness defects of the stiffened cylindrical shell. In step 2, the arc-length method, which incorporates material nonlinearity, is used to calculate the nonlinear static buckling critical load of unstiffened and stiffened cylindrical shells with initial defects under uniform external pressure loads, and this is used to obtain the matching factor between different stiffener stiffnesses and shell skin stiffness. In step 5, different defect amplitudes, stiffness matching factors between stiffeners and cylindrical shell skin, and the set water depth are used as input layer parameters, and dynamic buckling critical load, circumferential and axial buckling modes are used as output layer parameters. A proxy model of the stiffened cylindrical shell under underwater explosive load is established with the dynamic buckling critical load, circumferential and axial buckling modes, defect amplitude, stiffener and cylindrical shell skin stiffness matching factors, and set water depth, so as to realize the rapid prediction of dynamic buckling characteristics of stiffened cylindrical shells with initial defects under underwater explosive load. Attached Figure Description

[0013] Figure 1 This is a flowchart of a method for predicting the dynamic buckling characteristics of a reinforced cylindrical shell with initial defects under underwater explosive load based on artificial intelligence, according to the present invention.

[0014] Figure 2 is a measured initial geometric defect distribution diagram of the machined cylindrical shell obtained by reverse modeling technology according to the present invention.

[0015] Figure 3 is a diagram showing the measured initial thickness defect distribution of the processed cylindrical shell obtained by reverse modeling technology according to the present invention.

[0016] Figure 4 This is a schematic diagram of the numerical calculation results of the critical load for dynamic buckling of a reinforced cylindrical shell with initial defects under underwater explosive load according to the present invention.

[0017] Figure 5 This is a schematic diagram of the prediction results of the artificial intelligence surrogate model for the critical load of dynamic buckling of a reinforced cylindrical shell with initial defects under underwater explosive load according to the present invention.

[0018] Figure 6 is a schematic diagram of the numerical calculation result of the dynamic buckling mode of the initial defect-containing stiffened cylindrical shell under underwater explosion load of the application and the prediction result of the artificial intelligence agent model m = 6, n = 3. DETAILED DESCRIPTION

[0019] The application proposes a method for predicting the dynamic buckling characteristics of an initial defect-containing stiffened cylindrical shell under underwater explosion load based on artificial intelligence. The method introduces the spatial distribution characteristics of the initial defect by using reverse modeling technology and modal superposition method, obtains the stiffener and skin matching factor based on the nonlinear static buckling critical load of the initial defect-containing unstiffened and stiffened cylindrical shell under uniform external pressure load, obtains the dynamic buckling critical load and mode of the stiffened cylindrical shell with different defect amplitudes and stiffener numbers under different layout water depths based on the dynamic buckling criterion and the spatial distribution characteristics of the circumferential and axial typical response characteristics, and establishes the agent model of the dynamic buckling critical load and buckling mode of the stiffened cylindrical shell and the initial defect amplitude, stiffness matching factor and layout water depth by using artificial intelligence method, so as to realize the rapid prediction of the dynamic buckling characteristics of the initial defect-containing stiffened cylindrical shell under underwater explosion load.

[0020] Specifically, in combination with Figures 1-6 , the application proposes a method for predicting the dynamic buckling characteristics of an initial defect-containing stiffened cylindrical shell under underwater explosion load based on artificial intelligence. The method comprises the following steps: Step 1: based on reverse modeling technology and modal superposition method, a high-precision three-dimensional numerical calculation model of the dynamic response characteristics of the initial defect-containing stiffened cylindrical shell under underwater explosion load is established. Step 2: the nonlinear static buckling critical load of the initial defect-containing unstiffened and stiffened cylindrical shell under uniform external pressure load is calculated P sc,u 、 P sc,s , and the stiffener and cylindrical shell skin stiffness matching factor K = P sc,u / P sc,s is obtained; according to P sc,s , the layout water depth value range of the stiffened cylindrical shell H <( P sc,s / pg ) is obtained, wherein p represents the density of water, g represents the acceleration of gravity; Step 3: the layout water depth H , the charge equivalent W , and the blast distance d of the stiffened cylindrical shell are changed, and the underwater explosion impact factor The dynamic response characteristics of a stiffened cylindrical shell, taking into account initial defects, under underwater explosive loading were calculated using a three-dimensional numerical method. The maximum values ​​of the response characteristics at typical measuring points under the final deformation state of the stiffened cylindrical shell under underwater explosive loading were recorded. X max ; Step 4: Combine dynamic buckling criteria and impact factors at typical monitoring locations. C and X max The curve determines whether dynamic buckling has occurred, and from this, the critical load for dynamic buckling of the stiffened cylindrical shell is obtained. C s Simultaneously, based on the spatial distribution characteristics of typical circumferential and axial response features, the circumferential and axial dynamic buckling modes were obtained. m , n ); Step 5: For initial defects with basically consistent spatial distribution patterns but different defect amplitudes... L For cases where the initial defect spatial distribution pattern and defect amplitude are the same but the number of ribs differs, repeat steps 2-4 above for numerical calculation, and use artificial intelligence methods to establish the critical dynamic buckling load of the stiffened cylindrical shell. C s Circumferential and axial buckling modes ( m , n ) and initial defect amplitude L Stiffness Matching Factor K Deployment water depth H proxy model C s = f ( L, H ) / or C s = f ( K, H ), ( m , n )= Y ( L, H ) / ( m , n )= Y ( K, H And quickly determine different based on the proxy model. L Or different K Reinforced cylindrical shells at different water depths H Dynamic buckling critical load C s 、 Circumferential and axial dynamic buckling modes ( m , n ).

[0021] Furthermore, the initial defects in step 1 include shape defects and thickness defects. For the processed stiffened cylindrical shell structure, a 3D scanner and a thickness gauge are used to scan the shape defects and thickness defects, and the measurement results are used to perform 3D imaging to obtain the spatial distribution characteristics of the defects. For the unprocessed stiffened cylindrical shell structure, the modal superposition method and the existing thickness defect distribution law are used to obtain the shape and thickness defects.

[0022] Furthermore, for both fabricated and unfabricated reinforced cylindrical shell structures, based on reverse modeling technology and modal superposition method to couple shape and thickness defects, a high-precision three-dimensional numerical calculation model of the dynamic response of reinforced cylindrical shells under underwater explosive loads, taking into account the initial defects of the structure, is established using finite element methods such as ABAQUS and LS-DYNA or meshless methods such as SPH-RKPM.

[0023] Furthermore, in step 2, the arc-length method, which takes into account material nonlinearity, is used to calculate the critical nonlinear static buckling load of unreinforced and reinforced cylindrical shells under uniform external pressure load, taking into account initial defects. P sc,u , P sc,s This leads to the stiffness matching factor between the ribs and the cylindrical shell skin. K .

[0024] Furthermore, in step 3, when performing numerical calculations of the dynamic response characteristics of the stiffened cylindrical shell under underwater explosive loads, taking into account the initial defects, a single variable principle is adopted, i.e., the variable is fixed. W , d , H Two of these values ​​are used to calculate the critical load magnitude by taking the corresponding value of another physical quantity when dynamic buckling occurs. Typical measurement points for stiffened cylindrical shells include at least three circumferential lines at the head, middle, and tail, and at least four axial lines including the blast-facing surface, the back blast surface, and the side blast surface. Typical response characteristics of stiffened cylindrical shells include plastic deformation and plastic strain.

[0025] Furthermore, in step 4, different typical measuring point locations are considered. C - X max The curves are used to determine whether dynamic buckling has occurred using the Budiansky-Roth dynamic buckling criterion or the Southwell method, thereby obtaining the critical load for dynamic buckling. Based on the circumferential and axial spatial distribution characteristics of typical response features at different measuring points of the stiffened cylindrical shell, the circumferential and axial dynamic buckling modes are obtained. m , n ).

[0026] Furthermore, in step 5, for cases where the initial defect spatial distribution pattern is basically consistent but the defect amplitude is different... LDifferent, or the initial defect spatial distribution pattern and defect amplitude are the same but the number of ribs are different (i.e. K In different cases, first obtain the deployment depth according to step 2 above. H The range was then determined, and the reinforced cylindrical shell and the medicine pack were laid out at the specified water depth. H Medicine pack equivalent W By adjusting the detonation distance d Alternatively, a reinforced cylindrical shell and a medicine pack can be fixed at different water depths. H Explosion distance d By adjusting the equivalent amount of the medicine pack W， Repeat steps 3-4 above to obtain different results. L / or K Down C s Value, and change it based on this. H And further obtained different L / or K and H Down C s value.

[0027] Furthermore, in step 5, the data features are normalized and an artificial intelligence forecasting model is established. The input layer parameters include ( H, L )or( H, K The output layer parameters include the dynamic buckling critical load. C s and circumferential and axial dynamic buckling modes ( m , n The hyperparameters of the artificial intelligence prediction model were obtained through automated parameter optimization methods, and the critical dynamic buckling load of the stiffened cylindrical shell under underwater explosive load was established accordingly. C s Circumferential and axial dynamic buckling modes ( m , n ) and initial defect amplitude L / or stiffness matching factor K Deployment water depth H Down-the-proxy model C s = f ( L, H ) / or C s = f ( K, H ), ( m , n )= Y ( L, H ) / or( m , n )= Y ( K, H And quickly determine different based on the proxy model.L Or different K Reinforced cylindrical shells at different water depths H Dynamic buckling critical load C s Circumferential and axial dynamic buckling modes ( m , n ).

[0028] This invention proposes an artificial intelligence-based method for predicting the dynamic buckling characteristics of stiffened cylindrical shells with initial defects under underwater explosive loading. The method introduces initial defects into the stiffened cylindrical shell using inverse modeling and modal superposition techniques. It calculates the nonlinear static buckling critical load of unstiffened and stiffened cylindrical shells with initial defects under uniform external pressure using the arc-length method that incorporates material nonlinearity, and obtains matching factors between different stiffener stiffnesses and the shell skin stiffness. Based on artificial intelligence, the method establishes the dynamic buckling critical load, circumferential and axial buckling modes and defect amplitudes of the stiffener under underwater explosive loading, the matching factor between stiffener and shell skin stiffness, and deploys a surrogate model at different water depths, enabling rapid prediction of the dynamic buckling characteristics of stiffened cylindrical shells with initial defects under underwater explosive loading.

[0029] Specifically, in combination Figures 1-6 The implementation steps of this invention are described as follows: 1. Based on inverse modeling techniques and modal superposition methods, and considering the initial geometry and thickness defects of coupled stiffened cylindrical shells, a high-precision three-dimensional numerical model of the dynamic response characteristics of stiffened cylindrical shells with initial defects under underwater explosive loads is established using meshed or meshless methods. The matching factor between different stiffeners and the stiffness of the cylindrical shell skin is obtained using the arc-length method that incorporates material nonlinearity. K Change the water depth of the reinforced cylindrical shell. H Medicine pack equivalent W Explosion distance d Calculate the underwater explosion impact factor C Record the maximum values ​​of the response characteristics at typical measuring points of a stiffened cylindrical shell under underwater explosive loading in its final deformed state. X max .

[0030] 2. Based on C - X max The curve was combined with the dynamic buckling criterion to obtain the critical load for dynamic buckling of the stiffened cylindrical shell. C s Based on the spatial distribution characteristics of typical circumferential and axial response features, the circumferential and axial dynamic buckling modes are obtained. m , n Based on this, for cases where the initial spatial distribution of defects is basically consistent but the defect amplitude is large,L Different, or the initial defect spatial distribution pattern and defect amplitude are the same but the number of ribs are different (i.e. K For cases where the conditions differ, repeat the above steps and use artificial intelligence methods to establish the critical dynamic buckling load of a stiffened cylindrical shell under underwater explosive loading. C s Circumferential and axial dynamic buckling modes ( m , n ) and initial defect amplitude L (or stiffness matching factor) K ), Deployment water depth H The proxy model enables rapid prediction of the dynamic buckling characteristics of stiffened cylindrical shells with initial defects under underwater explosive loads.

Claims

1. A method for predicting the dynamic buckling characteristics of a stiffened cylindrical shell with initial defects under underwater explosive loading based on artificial intelligence, characterized in that, The method includes: Step 1: Establish a three-dimensional numerical model of the dynamic response characteristics of a stiffened cylindrical shell with initial defects under underwater explosive loading; Step 2: Calculate the critical nonlinear static buckling load for unstiffened and stiffened cylindrical shells under uniform external pressure load, taking into account initial defects. P sc,u , P sc,s And obtain the stiffness matching factor between the ribs and the cylindrical shell skin. κ = P sc,u / P sc,s ;according to P sc,s The range of water depth values ​​for the reinforced cylindrical shell layout was obtained. H <( P sc,s / ρg ),in ρ Represents the density of water. g Represents gravitational acceleration; Step 3: Change the water depth of the reinforced cylindrical shell. H Medicine pack equivalent W Explosion distance d Calculate the underwater explosion impact factor The dynamic response characteristics of a stiffened cylindrical shell, taking into account initial defects, under underwater explosive loading were calculated using a three-dimensional numerical method. The maximum values ​​of the response characteristics at typical measuring points under the final deformation state of the stiffened cylindrical shell under underwater explosive loading were recorded. ξ max ; Step 4: Combine dynamic buckling criteria and impact factors at typical monitoring locations. C and ξ max The curve determines whether dynamic buckling has occurred, and from this, the critical load for dynamic buckling of the stiffened cylindrical shell is obtained. C s Simultaneously, based on the spatial distribution characteristics of typical circumferential and axial response features, the circumferential and axial dynamic buckling modes were obtained. m , n ); Step 5: For initial defects with basically consistent spatial distribution patterns but different defect amplitudes... λ For cases where the initial defect spatial distribution pattern and defect amplitude are the same but the number of ribs differs, repeat steps 2-4 above for numerical calculation, and use artificial intelligence methods to establish the critical dynamic buckling load of the stiffened cylindrical shell. C s Circumferential and axial buckling modes ( m , n ) and initial defect amplitude λ Stiffness Matching Factor κ Deployment water depth H proxy model C s = f ( λ,H ) / or C s = f ( κ,H ), ( m , n )= Ψ ( λ,H ) / ( m , n )= Ψ ( κ,H And quickly determine different based on the proxy model. λ Or different κ Reinforced cylindrical shells at different water depths H Dynamic buckling critical load C s 、 Circumferential and axial dynamic buckling modes ( m , n ).

2. The method according to claim 1, characterized in that, In step 1, the initial defects include shape defects and thickness defects. For the processed stiffened cylindrical shell structure, the shape defects and thickness defects are scanned by a 3D scanner and a thickness gauge, respectively, and the spatial distribution characteristics of the defects are obtained by 3D imaging of the measurement results. For the unprocessed stiffened cylindrical shell structure, the shape and thickness defects are obtained by modal superposition method and existing thickness defect distribution law, respectively.

3. The method according to claim 2, characterized in that, For both fabricated and unfabricated reinforced cylindrical shell structures, a three-dimensional numerical calculation model of the dynamic response of a reinforced cylindrical shell under underwater explosive load, taking into account initial structural defects, is established using reverse modeling technology and modal superposition method to couple shape and thickness defects.

4. The method according to claim 1, characterized in that, In step 2, the arc-length method, which takes into account material nonlinearity, is used to calculate the critical nonlinear static buckling load of unreinforced and reinforced cylindrical shells under uniform external pressure load, taking into account initial defects. P sc,u , P sc,s This leads to the stiffness matching factor between the ribs and the cylindrical shell skin. κ .

5. The method according to claim 1, characterized in that, In step 3, when performing numerical calculations of the dynamic response characteristics of the stiffened cylindrical shell under underwater explosive loads, taking into account the initial defects, a single variable principle is adopted, i.e., a fixed variable is used. W , d , H Two of these values ​​are used to calculate the critical load magnitude by determining the corresponding value of another physical quantity when dynamic buckling occurs.

6. The method according to claim 1, characterized in that, In step 4, different typical measuring point locations are used. C - ξ max The curves are used to determine whether dynamic buckling has occurred using the Budiansky-Roth dynamic buckling criterion or the Southwell method, thereby obtaining the critical load for dynamic buckling. Based on the circumferential and axial spatial distribution characteristics of typical response features at different measuring points of the stiffened cylindrical shell, the circumferential and axial dynamic buckling modes are obtained. m , n ).

7. The method according to claim 1, characterized in that, In step 5, the deployment water depth is first obtained according to step 2 above. H The range was then determined, and the reinforced cylindrical shell and the medicine pack were laid out at the specified water depth. H Medicine pack equivalent W By adjusting the detonation distance d Alternatively, a reinforced cylindrical shell and a medicine pack can be fixed at different water depths. H Explosion distance d By adjusting the equivalent amount of the medicine pack W， Repeat steps 3-4 above to obtain different results. λ / or κ Down C s Value, and change it based on this. H And further obtained different λ / or κ and H Down C s value.

8. The method according to claim 7, characterized in that, In step 5, the data features are normalized and an artificial intelligence forecasting model is established. The input layer parameters include ( H,λ )or( H,κ The output layer parameters include the dynamic buckling critical load. C s and circumferential and axial dynamic buckling modes ( m , n The hyperparameters of the artificial intelligence prediction model were obtained through automated parameter optimization methods, and the critical dynamic buckling load of the stiffened cylindrical shell under underwater explosive load was established accordingly. C s Circumferential and axial dynamic buckling modes ( m , n ) and initial defect amplitude λ / or stiffness matching factor κ Deployment water depth H Down-the-proxy model C s = f ( λ,H ) / or C s = f ( κ,H ), ( m , n )= Ψ ( λ,H ) / or( m , n )= Ψ ( κ,H And quickly determine different based on the proxy model. λ Or different κ Reinforced cylindrical shells at different water depths H Dynamic buckling critical load C s Circumferential and axial dynamic buckling modes ( m , n ).

Images