Bridge and ship impact simulation and evaluation method and system based on mixed reality technology

Abstract

The application discloses a bridge and ship impact simulation evaluation method and system based on a mixed reality technology, and comprises the following steps: a mixed reality environment is created by fusing the physical properties of actual bridges and ships and virtual impact event simulation; an input device in the mixed reality environment receives the impact angle, speed and force of a user about an impact event, and an impact event data set is constructed; a calculation model combined with a deep neural network and a multi-physical field coupling is constructed; based on the impact event data and hydrogeological data in the mixed reality environment, combined with the received weather data set, an impact event is simulated, and the impact process is rendered in real time in the mixed reality environment; the simulation impact event result is analyzed; parameters are adjusted in real time during the mixed reality environment simulation process, and evaluation and optimization are performed. Through the mixed reality technology, the application creates an environment containing real world elements and highly realistic virtual simulation, greatly improving the reality and interactivity of simulation evaluation.

Description

TECHNICAL FIELD

[0001] The present application relates to a bridge and ship impact simulation evaluation method and system, especially to a bridge and ship impact simulation evaluation method and system based on mixed reality technology. BACKGROUND

[0002] In the field of bridge and ship engineering, ensuring the safety and stability of structures in impact events such as ship collisions with bridges has been a long-standing challenge.

[0003] Traditional simulation evaluation methods rely on static calculation models and numerical analysis, which can predict the impact effect to some extent, but often lack realism and interactivity, making it difficult for engineers and decision-makers to intuitively understand the complexity of impact events and their actual impact on structural integrity.

[0004] After searching, the Chinese utility model patent with authorization announcement number CN213740832U discloses a debris flow simulation device for simulating debris flow impact on bridges, which includes a gantry support and an inclined slope. The upper and middle sections of the inclined slope are connected to the gantry support on both sides, and the lower end of the inclined slope is aligned with the pier of the bridge model. A mixing box is connected to the gantry support, and a flap is provided at the bottom of the mixing box. The outlet of the flap is opposite to the inclined slope, and a conveyor belt is provided on the inclined slope to match the slope. This device can simulate bridge impact, but has great limitations in evaluating and optimizing the impact resistance of structures, and cannot provide sufficient detailed data to support decision-making, especially when facing complex environmental conditions and variable impact angles, speeds, and other parameters.

[0005] In addition, the Chinese invention patent with authorization announcement number CN111829738B discloses a bridge bearing capacity lightweight evaluation method based on impact load, which uses a drop hammer to simulate impact load to load the bridge. Displacement sensors and acceleration sensors are used to measure the displacement time history curve and acceleration time history curve of the bridge control section. The extracted dynamic response is compared and analyzed with the theoretical calculation value to establish an evaluation index to evaluate the bearing capacity of the bridge. However, it cannot accurately simulate the dynamic response in the actual impact process, including the distribution of impact force, structural stress, and potential damage.

[0006] Therefore, how to provide a bridge and ship impact simulation evaluation method based on mixed reality technology is a problem that needs to be solved by those skilled in the art. SUMMARY

[0007] The purpose of the present application is to provide a bridge and ship impact simulation evaluation method and system based on mixed reality technology, which greatly improves the realism and interactivity of simulation evaluation, and can accurately simulate the dynamic response in the impact process.

[0008] Technical solution: The steps of the present application are as follows:

[0009] Create a mixed reality environment that combines the physical properties of actual bridges and ships with virtual impact event simulation;

[0010] The input device in the mixed reality environment receives user input on the impact angle, speed and force of the impact event, and builds an impact event dataset;

[0011] Build a computational model that combines deep neural networks with multi-physics coupling;

[0012] Based on the impact event data and hydrogeological data in the mixed reality environment, combined with the received weather data set, use the computational model to simulate the impact event and render the impact process in real time in the mixed reality environment;

[0013] Analyze the results of the simulated impact event to assess the impact on the structural integrity of the bridge and ship;

[0014] Adjust parameters in real time during the simulation process in the mixed reality environment, observe the impact effect under different conditions, and evaluate and optimize the design of the bridge and ship based on the simulation results;

[0015] Record and store each bridge and ship impact simulation process and results.

[0016] The mixed reality environment that combines the physical properties of actual bridges and ships with virtual impact event simulation specifically includes:

[0017] S11, on-site scanning of target bridges and ships to collect geometric structure data, and building three-dimensional models of bridges and ships based on point cloud data:

[0018]

[0019] Where f is the implicit function of the reconstructed surface, V is the normal vector field of the point cloud, And Laplace operator and divergence, respectively;

[0020] S12, physical property setting for the model: for the bridge model, set the material density, elastic modulus, tensile strength and compressive strength parameters; for the ship model, set the navigation parameters;

[0021] S13, build hydrogeological data set and weather data set through integrated environmental sensor network;

[0022] S14, real-time transmission of hydrogeological data set and weather data set data to mixed reality platform, synchronized update with three-dimensional model;

[0023] S15, integrate dynamic response algorithm, simulate the reaction of the ship when subjected to impact force:

[0024]

[0025] Where M is the mass matrix, C is the damping matrix, K is the stiffness matrix, x is the displacement vector, and F(t) is the time-varying external force vector.

[0026] The hydrogeological data set construction includes deploying underwater sonar sensors, flowmeters and wave height sensors to monitor and collect water depth D, flow velocity V w and wave height H w data in real time; the weather data set construction includes deploying wind speed and direction meters and temperature and humidity sensors to monitor and collect wind speed V a , wind direction D a , air temperature T and humidity H data in real time.

[0027] The impact event data set is constructed by receiving the impact angle, speed and force of the user about the impact event in the mixed reality environment; specifically including:

[0028] S21, deploying a user interface in a mixed reality environment to input the impact angle α, impact speed v and impact force F of the impact event;

[0029] S22, the data input by the user is packaged into an impact event data set:

[0030] Data Impact ={ID, T stamp , α, v, F(m, v, v0, Δt)};

[0031] Where F(m, v, v0, Δt) represents that force F is a function of mass m, impact speed v, initial speed v0 and time interval Δt.

[0032] The calculation model combined with deep neural network and multi-physical field coupling, specifically including:

[0033] S31, develop a deep neural network model for predicting and analyzing the stress response and potential damage of the bridge and ship structure under impact events, the training data includes historical impact event data, material performance parameters and structure response data, and the model outputs the predicted stress distribution map and damage probability evaluation:

[0034]

[0035] Where L represents the loss function, N is the number of training samples, y i is the actual stress response of the i-th sample, and X iis the corresponding input feature, Θ represents the parameters of the neural network, and f(·) represents the prediction function of the neural network model;

[0036] S32, a multi-physics coupling model is constructed, and the basic equations of fluid dynamics, material mechanics and structural dynamics are integrated to simulate the physical process under the impact event. The fluid dynamics part adopts the Navier-Stokes equation:

[0037]

[0038] where g represents the gravity acceleration vector, F ext is other external force;

[0039] The material mechanics part uses the stress-strain relationship:

[0040]

[0041] where: represents the tensor product, Φ is a function, represents the additional stress contribution of the material nonlinear behavior, and depends on the strain ∈, strain rate and material parameters Θ M ;

[0042] The structural dynamics part, which integrates nonlinear response and high-order effects, uses the following equation:

[0043]

[0044] where M, C, and K are the mass, damping and stiffness matrices respectively, u is the displacement vector, F(t) is the time-dependent external force vector, f ext is the body force density, and Ω is the volume of the structure.

[0045] S33, the deep neural network and the multi-physics coupling model are integrated through an algorithm framework. The damage probability evaluation provided by the deep neural network is used as the boundary condition and initial condition of the multi-physics coupling model, and the simulation results of the multi-physics coupling model are used to train and optimize the deep neural network model:

[0046]

[0047] where L(Θ, Θ M ) is a data-based loss function, Ψ(Θ, Θ M ) is a model complexity or regularization term, and λ is a weight parameter balancing the two terms.

[0048] Based on the impact event data and hydrogeological data in the mixed reality environment, combined with the received weather data set, a calculation model is used to simulate the impact event and render the impact process in real time in the mixed reality environment. Specifically, it includes:

[0049] S41, integrating the impact event dataset Data in the mixed reality environment Impact , hydrogeological dataset Data Hydro and received weather dataset Data Weather to provide initial conditions and environmental parameters for the simulation process;

[0050] S42, introducing fluid-structure interaction analysis methods and dynamic damage assessment algorithms when performing multi-physics coupled calculation models for impact event simulation.

[0051] The fluid-structure interaction analysis uses improved coupling formulas to simulate the dynamic influence of water flow on ship hulls and bridge structures:

[0052]

[0053] where ρ f , u f , p f and μ f represent the density, velocity field, pressure and dynamic viscosity of the fluid, respectively; F fsi represents the interaction force between the fluid and the structure;

[0054] The dynamic response and damage assessment of the structure use a deep learning-based damage detection algorithm that takes into account the complexity of material nonlinearity and impact load. The structural stress response σ s and strain response ∈ s are calculated by the following relationship:

[0055]

[0056] where E s represents the elastic modulus of the structural material, α s is the material nonlinearity factor, Φ is a function representing the material nonlinearity behavior, and Θ s is the material parameter set.

[0057] The dynamic damage assessment algorithm compares the structural response with historical damage data, uses a deep learning model to predict the potential damage location and extent of the structure, and optimizes it through the following loss function:

[0058]

[0059] where M represents the number of training samples, d j and represent the true damage extent and the model-predicted damage extent of the jth sample, respectively, and Θ damage represents the parameters of the damage assessment model.

[0060] The analysis simulates the results of the impact event, evaluates the impact of the impact on the structural integrity of the bridge and the ship, and specifically includes:

[0061] S51, after the simulation of the impact event is completed, the system automatically enters the result analysis stage, extracts the structural deformation variable, stress distribution σ(x,y,z) and identification of potential damage area from the simulation data;

[0062] S52, applying a structural integrity analysis algorithm, based on the stress distribution σ(x,y,z) obtained by simulation and the material strength characteristics, evaluating the safety of the structure under the action of a specific impact force, by comparing the stress value with the yield strength σ y and the maximum bearing strength σ u to determine the safety of the structure, the calculation formula is:

[0063]

[0064] Wherein, SafetyFactor(x,y,z) represents the safety factor of the structure at point (x,y,z), the value greater than 1 indicates safety, and the value less than 1 indicates potential risk;

[0065] S53, for the identified potential damage area, a damage evaluation model is used to analyze the damage degree and possible influence, combined with the fatigue characteristics of the structural material and the cumulative damage theory, the cumulative effect of the structure under continuous or multiple impact is evaluated, and the damage degree is represented by damage index D i :

[0066]

[0067] Wherein, N is the number of impacts, Δσ n is the stress range generated by the nth impact, σ f is the fatigue limit of the material, and m is the fatigue index of the material;

[0068] S54, the system will generate a comprehensive evaluation report by comprehensively evaluating the safety factor and the damage index.

[0069] A bridge and ship impact simulation evaluation system based on mixed reality technology, comprising: a mixed reality environment creation module, an impact event data set construction module, a calculation model construction module combining a deep neural network and a multi-physical field coupling, an impact event simulation module and a simulation impact event result analysis module.

[0070] Advantages: the present application has the following advantages:

[0071] (1) The application creates an environment that contains both real-world elements and highly realistic virtual simulations through the application of mixed reality technology, greatly improving the realism and interactivity of simulation evaluation. Users can directly operate in the simulation environment, such as adjusting the impact angle and intensity, and observe the changes in structural response in real time, thereby gaining a deep understanding of the impact of impact events;

[0072] (2) The application can accurately simulate the dynamic response during the impact process, including the distribution of impact force, structural stress, and potential damage, by combining deep neural networks with multi-physics coupling computational models, providing detailed data support for the design and maintenance of bridges and ships;

[0073] (3) The application can render the impact process in real-time in a mixed reality environment, including visualization of structural deformation, stress distribution, and potential damage, improving the intuitiveness and interactivity of evaluation. By considering impact event parameters, environmental conditions, and structural material properties, comprehensive structural safety evaluation and maintenance recommendations can be provided. BRIEF DESCRIPTION OF DRAWINGS

[0074] Figure 1 The flowchart of the application. DETAILED DESCRIPTION

[0075] The application will be further described below in conjunction with the accompanying drawings.

[0076] As shown in the figure, the bridge and ship impact simulation evaluation method based on mixed reality technology of the application includes the following steps: Figure 1 S1, create a mixed reality environment that integrates the physical properties of actual bridges and ships and virtual impact event simulation: the mixed reality environment simulates the interaction between the real world and the virtual world using three-dimensional modeling technology and advanced physics engines, and integrates real-time hydrogeological data, weather data, and ship dynamic response systems inside the mixed reality environment; specifically including:

[0077] S11, use three-dimensional scanning technology to scan the target bridge and ship on site to collect accurate geometric structure data. The point cloud data obtained by scanning is represented by the function F(x, y, z) = 0, where x, y, and z represent the coordinates in three-dimensional space. According to the point cloud data, a Poisson reconstruction algorithm is used to construct a three-dimensional model of the bridge and ship:

[0078]

[0079]

[0080] where f is the implicit function of the reconstructed surface, V is the normal vector field of the point cloud, and represent the Laplacian and divergence, respectively; ​

[0081] S12, import the three-dimensional model into the mixed reality software platform, and apply the advanced physical engine to set the physical properties of the model. For the bridge model, the settings include material density, elastic modulus, tensile strength and compressive strength parameters. For the mechanical behavior of the material, it is represented by the three-dimensional form of Hooke's law:

[0082] σ = E · ε

[0083] where σ is the stress tensor, E is the elastic modulus tensor, and ε is the strain tensor;

[0084] For the ship model, the waterline area and ship resistance coefficient navigation related parameters are also set. For the hydrodynamic characteristics of the ship, the Navier-Stokes equation in fluid dynamics is used:

[0085]

[0086] where v is the fluid velocity field, p is the pressure field, ρ is the fluid density, μ is the dynamic viscosity, and F represents the external force;

[0087] S13, through the integrated environmental sensor network, real-time collection of water depth, flow rate and wave height, construction of hydrogeological data set, at the same time, real-time collection of wind speed, wind direction, temperature and humidity, construction of weather data. Among them, the construction of hydrogeological data set includes deploying underwater sonar sensor, flowmeter and wave height sensor in the specified water area, real-time monitoring and collecting water depth D, flow rate V w and wave height H w data;

[0088] The measurement of water depth D is determined by the relationship between sound wave reflection time T r and sound speed c, which combines water temperature T w , salinity S and depth:

[0089]

[0090]

[0091] The measurement of flow rate V w is calculated by the difference between the transmitted frequency f e and the received frequency f r :

[0092]

[0093] where θ is the angle between the flowmeter and the flow direction;

[0094] The measurement of wave height H w is based on wave period T w and wave speed C wEstimation:

[0095]

[0096] where E w is the wave energy per unit area, p w is the density of water, and g is the acceleration due to gravity;

[0097] All the collected water depth D, flow velocity V w and wave height H w data include a unique identifier ID and a timestamp T stamp ; and the data from different sensors are integrated using the weighted average method to construct the hydrogeological dataset Data Hydro :

[0098] Data Hydro (ID, T stamp , D, V w , H w ) = SensorData Hydro .

[0099] In this embodiment, the weather dataset construction includes deploying an anemometer and a temperature and humidity sensor at a specified location to monitor and collect wind speed V a , wind direction D a , air temperature T and humidity H data in real time;

[0100] The measurement of wind speed V a and wind direction D a combines the wind speed profile in the atmospheric boundary layer, which is given by the logarithmic wind speed profile formula:

[0101]

[0102] where z is the measurement height, u * is the friction velocity, K is the von Karman constant, d is the displacement height, and z0 is the roughness length;

[0103] The determination of wind direction uses the influence of atmospheric stability and surface roughness to analyze, combined with the dynamics of the atmospheric boundary layer, by analyzing the change of wind direction under different stability conditions to obtain the expression of wind direction:

[0104]

[0105] where represents the rate of change of wind direction with height z, f is the Coriolis parameter related to the Earth's rotation and latitude, V g is the geostrophic wind speed, V a is the actual wind speed, and s is the static stability parameter, reflecting the stability of the atmosphere, and AD terrainis the terrain-induced wind deflection, estimated according to the relationship between the terrain features and the wind direction;

[0106] The relationship between air temperature T and humidity H is calculated through a physical description of humidity, the absolute humidity H a and the air temperature T and the relative humidity H r :

[0107]

[0108] All collected wind speed V a , wind direction D a , air temperature T and humidity H data each data point includes a unique identifier ID and a timestamp T stamp ; and using the weighted average method, integrating data from different sensors, to build a weather dataset Data Weather :

[0109] Data Weather (ID, T stamp , V a , D a , T, H) = SensorData Weather .

[0110] S14, Hydrogeological data set and weather data set data are transmitted to the mixed reality platform in real time through the API interface, and are updated synchronously with the three-dimensional model;

[0111] S15, integrate dynamic response algorithm, used to simulate the response of the ship when subjected to impact force:

[0112]

[0113] Where M is the mass matrix, C is the damping matrix, K is the stiffness matrix, x is the displacement vector, and F(t) is the time-varying external force vector, including the force generated by the impact event.

[0114] S2, the input device receives the user's impact angle, speed and force of the impact event in the mixed reality environment, and constructs an impact event dataset; specifically including:

[0115] S21, deploy a user interface in the mixed reality environment, and the user inputs the impact angle a, impact speed v and impact force F of the impact event through the input device; wherein the impact angle a is defined as the angle between the ship's moving direction and the bridge's transverse direction, with a value range of 0° to 180°, directly input by the user; the impact speed v represents the speed of the ship at the moment of contact with the bridge, with a unit of meters per second, directly input by the user; the impact force F is calculated by the acceleration a:

[0116]

[0117] where v0 is the initial velocity, which is not zero before the instant of contact;

[0118] The force F is calculated by the following equation:

[0119]

[0120] S22, the user input data is packaged into an impact event data set:

[0121] Data Impact = {ID, T stamp , α, v, F(m, v, v0, Δt)};

[0122] where F(m, v, v0, Δt) represents that the force F is a function of mass m, impact velocity v, initial velocity v0, and time interval Δt.

[0123] S3, a computational model combining deep neural networks and multi-physical field coupling is constructed, the deep neural network is used to predict and analyze the impact of impact force on bridge and ship structure, and the multi-physical field coupling is used to combine the interaction of fluid dynamics, material mechanics and structural dynamics, so as to simulate the distribution of impact force, structural stress and potential damage; Specifically includes:

[0124] S31, develop a deep neural network model for predicting and analyzing the stress response and potential damage of bridge and ship structure under impact events, the training data includes historical impact event data, material performance parameters and structural response data, and the model outputs the predicted stress distribution map and damage probability evaluation:

[0125]

[0126] where L represents the loss function, N is the number of training samples, y i is the actual stress response of the i-th sample, X i is the corresponding input feature, including impact angle, velocity, force, etc., Θ represents the parameters of the neural network, and (f(·) represents the prediction function of the neural network model;

[0127] S32, construct a multi-physical field coupling model, integrate the basic equations of fluid dynamics, material mechanics and structural dynamics, simulate the physical process under impact events, and the fluid dynamics part adopts Navier-Stokes equation:

[0128]

[0129] where g represents the gravity acceleration vector, F ext is other external force;

[0130] The material mechanics part uses the stress-strain relationship:

[0131]

[0132] where: denotes the tensor product, Φ is a function representing the additional stress contribution of the material nonlinear behavior, which can depend on strain ∈, strain rate and material parameters Θ M ;

[0133] The structural dynamics part, which integrates the nonlinear response and high-order effects, uses the following equation:

[0134]

[0135] where M, C, and K are the mass, damping, and stiffness matrices, respectively, u is the displacement vector, F(t) is the time-dependent external force vector, f ext is the body force density, and Ω is the volume of the structure.

[0136] The S33, deep neural network and multi-physical field coupling model are integrated through an algorithmic framework, allowing interactive data exchange and feedback between the two, achieving comprehensive analysis and simulation of the impact process. The damage probability assessment provided by the deep neural network serves as the boundary conditions and initial conditions for the multi-physical field coupling model, and the simulation results of the multi-physical field coupling model are used to train and optimize the deep neural network model:

[0137]

[0138] where L(Θ, Θ M ) is the data-based loss function, Ψ(Θ, Θ M ) is the model complexity or regularization term, and λ is the weight parameter balancing the two terms.

[0139] S4, based on the impact event data and hydrogeological data in the mixed reality environment, combined with the received weather data set, using a computational model to simulate the impact event and render the impact process in real time in the mixed reality environment; Specifically includes:

[0140] S41, integrate the impact event data set Data Impact , the hydrogeological data set Data Hydro and the received weather data set Data Weather , provide initial conditions and environmental parameters for the simulation process;

[0141] S42, when performing the multi-physical field coupling calculation model for impact event simulation, an improved fluid-structure interaction analysis method and dynamic damage assessment algorithm are introduced;

[0142] Fluid-structure interaction analysis employs improved coupling formulas to simulate the dynamic effects of water flow on ship and bridge structures:

[0143]

[0144] where ρ f , u f , p f and μ f represent the density, velocity field, pressure and dynamic viscosity of the fluid respectively; F fsi represents the interaction force between the fluid and the structure;

[0145] The dynamic response and damage assessment of the structure employs a deep learning-based damage detection algorithm, which takes into account the complexity of material nonlinearity and impact load, and calculates the structural stress response σ s and strain response ∈ s through the following relationship:

[0146]

[0147] where E s represents the elastic modulus of the structural material, α s is the material nonlinearity factor, Φ is a function representing the material nonlinear behavior, and Θ s is the set of material parameters;

[0148] The dynamic damage assessment algorithm compares the structural response with historical damage data, and uses a deep learning model to predict the potential damage location and extent of the structure, which is optimized through the following loss function:

[0149]

[0150] where M represents the number of training samples, d j and represent the true damage extent and the model-predicted damage extent of the jth sample, respectively, and Θ damage represents the parameters of the damage assessment model.

[0151] S5, analyze the results of the simulated impact event, assess the impact on the integrity of the bridge and ship structures, including structural deformation, stress distribution and potential damage; specifically including:

[0152] S51, after the completion of the simulation of the impact event, the system automatically enters the result analysis stage, extracts the structural deformation, stress distribution σ(x, y, z) and identification of potential damage areas from the simulation data;

[0153] S52, applying a structural integrity analysis algorithm, based on the simulated stress distribution σ(x, y, z) and material strength characteristics, to evaluate the safety of the structure under the action of a specific impact force, by comparing the stress value with the yield strength σ y and the maximum bearing strength σ u to determine the safety of the structure, the calculation formula is:

[0154]

[0155] Among them, SafetyFactor(x, y, z) represents the safety factor of the structure at point (x, y, z), and the value greater than 1 indicates safety, and the value less than 1 indicates potential risk;

[0156] S53, for the identified potential damage area, a damage evaluation model is used to analyze the damage degree and possible influence, combined with the fatigue characteristics of the structural material and the cumulative damage theory, to evaluate the cumulative damage effect of the structure under continuous or multiple impact actions, and the damage degree is represented by damage index D i :

[0157]

[0158] Among them, N is the number of impacts, Δσ n is the stress range generated by the nth impact, σ f is the fatigue limit of the material, and m is the fatigue index of the material;

[0159] S54, the system will generate a comprehensive evaluation report based on the evaluation results of the safety factor and the damage index, and the report will list the safety status, damage degree and recommended maintenance or reinforcement measures of each key structural component in detail.

[0160] S6, the user adjusts the parameters in real time during the mixed reality environment simulation process, observes the impact effect under different conditions, and evaluates and optimizes the design of the bridge and the ship based on the simulation results;

[0161] S7, record and store each bridge and ship impact simulation process and results.

[0162] The application also provides a bridge and ship impact simulation evaluation system based on mixed reality technology, comprising:

[0163] The mixed reality environment creation module integrates the actual bridge and ship physical properties and the virtual impact event simulation: the mixed reality environment uses three-dimensional modeling technology and advanced physics engine to simulate the interaction between the real world and the virtual world, and the mixed reality environment is internally integrated with real-time hydrogeological data, weather data and ship dynamic response system;

[0164] Impact event dataset construction module: In the mixed reality environment, the input device receives user input on the impact angle, speed, and force of the impact event, and constructs an impact event dataset;

[0165] Deep neural network and multi-physics coupling calculation model construction module: Deep neural networks are used to predict and analyze the impact of impact forces on bridge and ship structures, while multi-physics coupling is used to combine the interactions of fluid dynamics, material mechanics, and structural dynamics to simulate the distribution of impact forces, structural stress, and potential damage.

[0166] Impact event simulation module: Based on the impact event data and hydrogeological data in the mixed reality environment, combined with the received weather dataset, the calculation model is used to simulate the impact event and render the impact process in real-time in the mixed reality environment.

[0167] Simulation impact event result analysis module: Evaluate the impact of the impact on the integrity of the bridge and ship structure, including structural deformation, stress distribution, and potential damage; adjust parameters in real-time during the simulation process in the mixed reality environment, observe the impact effect under different conditions, and evaluate and optimize the design of the bridge and ship based on the simulation results.

[0168] Example 1:

[0169] In the offshore bridge project, the challenge of how to improve the safety and stability of the bridge when it encounters a ship impact is faced. In order to solve this problem, the method of the present application is used, in order to achieve higher preventive measures and coping strategies.

[0170] In the design stage of the offshore bridge project, a mixed reality environment is created that contains both real bridge models and virtual impact event simulations. In this environment, not only can the structural response of the bridge under different impact conditions be observed, but also the impact angle, speed, and force parameters can be adjusted in real-time to achieve more accurate simulation results.

[0171] First, the offshore bridge was scanned in the field using three-dimensional scanning technology, collecting accurate geometric structure data of the bridge, and using these data to construct a three-dimensional model of the bridge in the mixed reality environment. Then, a series of parameters for the simulated impact event were input through the input device, including the preset impact angle, speed, and force.

[0172] Using the deep neural network and multi-physics coupling calculation model, a large cargo ship directly impacting the pier at different speeds was simulated. The deep neural network model predicted the potential damage to the bridge structure, while the multi-physics coupling model provided detailed views of the impact force distribution, structural stress, and deformation.

[0173] During the simulation process, it was found that in a specific simulation impact event, when the ship collided with the pier at a speed of 15 nautical miles per hour, the safety factor of the specific pier was lower than the standard safety threshold, indicating that the bridge may suffer serious damage in this case.

[0174] After conducting dozens of simulation tests with different parameters, the data showed that in the case of a ship impact at 15 nautical miles per hour, the most severe structural stress was concentrated at the connection of the pier, with stress values reaching 150% of the material yield strength. The simulation results also showed that the maximum displacement caused by the impact force reached 5 cm, exceeding the design safety standard. Through the damage assessment of the deep learning model, it was predicted that under this impact condition, there was a 40% probability of moderate to severe structural damage to the bridge.

[0175] To verify the accuracy of the simulation results, a review of all bridge impact events that occurred in the past five years was conducted. Through comparative analysis, it was found that the simulation results were highly consistent with the actual impact events in terms of damage degree and location, proving the effectiveness and accuracy of the simulation method.

[0176] The present invention creates an environment that contains both real-world elements and highly realistic virtual simulations through the application of mixed reality technology, greatly improving the realism and interactivity of simulation evaluation. Users can directly manipulate the simulation environment, such as adjusting the impact angle and force, and observe the real-time changes in structural response, thereby gaining a deeper understanding of the impact of impact events.

[0177] The present invention combines deep neural networks with multi-physics coupling computational models to accurately simulate dynamic responses during impact, including impact force distribution, structural stress, and potential damage, providing detailed data support for bridge and ship design and maintenance.

[0178] The present invention can render the impact process in real time in a mixed reality environment, including visualization of structural deformation, stress distribution, and potential damage, improving the intuitiveness and interactivity of evaluation. By considering impact event parameters, environmental conditions, and structural material properties, comprehensive structural safety evaluation and maintenance recommendations can be provided.

Claims

1. A bridge and ship impact simulation and evaluation method based on mixed reality technology, characterized in that, Includes the following steps: Creating a mixed reality environment that integrates the physical properties of actual bridges and ships with virtual impact event simulations, specifically including: S11. Conduct on-site scanning of the target bridge and ship to collect their geometric structure data, and construct 3D models of the bridge and ship based on the point cloud data: Where f is the implicit function for reconstructing the surface, and V is the normal vector field of the point cloud. and Let Laplace operator and divergence be represented respectively; S12. Set the physical properties of the model: For the bridge model, set the material density, elastic modulus, tensile strength and compressive strength parameters; for the ship model, set the navigation parameters. S13. Construct hydrogeological and weather datasets through an integrated environmental sensor network; S14. Transmit hydrogeological and weather datasets to the mixed reality platform in real time and update them synchronously with the 3D model. S15. Integrated dynamic response algorithm to simulate the ship's response to impact forces: Where M is the mass matrix, C is the damping matrix, K is the stiffness matrix, and x is the displacement vector. This is a time-varying vector of external forces. In a mixed reality environment, the input device receives the user's information about the impact angle, velocity, and force of an impact event, and constructs an impact event dataset. Constructing a computational model that combines deep neural networks with multiphysics coupling, specifically including: S31. Develop a deep neural network model to predict and analyze the stress response and potential damage of bridge and ship structures under impact events. The training data includes historical impact event data, material property parameters, and structural response data. The model outputs a predicted stress distribution map and a damage probability assessment. Where L represents the loss function, and N is the number of training samples. It is the actual stress response of the i-th sample. These are the corresponding input features. The parameters of the neural network, Represents the prediction function of a neural network model; S32. Construct a multiphysics coupled model, integrating the fundamental equations of fluid dynamics, materials mechanics, and structural dynamics, to simulate the physical processes under impact events. The fluid dynamics part adopts the Navier-Stokes equations. in, Represents the gravitational acceleration vector. Other external forces; The mechanics of materials section uses stress-strain relationships: Where, : denotes tensor product, It is a function representing the additional stress contribution to the nonlinear behavior of a material, which depends on strain. strain rate and material parameters ; In the structural dynamics section, considering both nonlinear response and higher-order effects, the following equations are used: Where M, C, and K are the mass, damping, and stiffness matrices, respectively. It is a displacement vector. It is a time-dependent external force vector. It is the volumetric force density. It is the volume of the structure; S33, deep neural networks, and multiphysics coupling models are integrated through an algorithmic framework. The damage probability assessment provided by the deep neural network serves as the boundary and initial conditions for the multiphysics coupling model, and the simulation results of the multiphysics coupling model are used to train and optimize the deep neural network model. in, It is a loss function based on data. It is a model complexity or regularization term. It is the weight parameter that balances the two terms; Based on impact event data and hydrogeological data in a mixed reality environment, combined with received weather datasets, a computational model is used to simulate impact events and render the impact process in real time in a mixed reality environment. Analyze the results of simulated impact events to assess the impact on the structural integrity of bridges and ships; During the mixed reality environment simulation, parameters are adjusted in real time to observe the impact effects under different conditions, and the design of bridges and ships is evaluated and optimized based on the simulation results; Record and store the process and results of each bridge and ship impact simulation.

2. The bridge and ship impact simulation and evaluation method based on mixed reality technology according to claim 1, characterized in that, The construction of the hydrogeological dataset includes the deployment of underwater sonar sensors, current meters, and wave height sensors to monitor and collect water depth D and current velocity in real time. and wave height Data; the construction of the weather dataset includes deploying anemometers and temperature and humidity sensors to monitor and collect wind speed data in real time. ,wind direction Temperature (T) and humidity (H) data.

3. The bridge and ship impact simulation and evaluation method based on mixed reality technology according to claim 1, characterized in that, The input device in the mixed reality environment receives the user's information about the impact angle, velocity, and force of the impact event, and constructs an impact event dataset; specifically including: S21. Deploy the user interface in the mixed reality environment and input the impact angle of the impact event. Impact velocity v and impact force F; S22. The user-input data is encapsulated into a shock event dataset: Wherein, ID represents a unique identifier; For timestamps; The force F represents the mass m, the impact velocity v, and the initial velocity. and time interval The function.

4. The bridge and ship impact simulation and evaluation method based on mixed reality technology according to claim 1, characterized in that, The method, based on impact event data and hydrogeological data from a mixed reality environment, combined with received weather datasets, employs a computational model to simulate impact events and renders the impact process in real-time within the mixed reality environment; specifically including: S41. Integrating impact event datasets from mixed reality environments. Hydrogeological dataset and the received weather dataset This provides initial conditions and environmental parameters for the simulation process; S42. When performing impact event simulation using a multiphysics coupled calculation model, introduce fluid-structure interaction analysis methods and dynamic damage assessment algorithms.

5. The bridge and ship impact simulation and evaluation method based on mixed reality technology according to claim 4, characterized in that, The fluid-structure interaction analysis employs an improved coupling formula to simulate the dynamic effects of water flow on the ship hull and bridge structure: in, , , and These represent the fluid's density, velocity field, pressure, and dynamic viscosity, respectively. This represents the interaction force between the fluid and the structure; The dynamic response and damage assessment of the structure employ a deep learning-based damage detection algorithm, taking into account the complexity of material nonlinearity and impact load, and the structural stress response. and strain response Calculated using the following relationship: in, This represents the elastic modulus of a structural material. For material nonlinearity factor, It is a function representing the nonlinear behavior of materials. This is a set of material parameters.

6. The bridge and ship impact simulation and evaluation method based on mixed reality technology according to claim 4, characterized in that, The dynamic damage assessment algorithm compares the structural response with historical damage data, uses a deep learning model to predict the potential location and extent of damage to the structure, and is optimized using the following loss function: Where M represents the number of training samples. and Let represent the actual damage level and the model-predicted damage level of the j-th sample, respectively. These represent the parameters of the damage assessment model.

7. The bridge and ship impact simulation and evaluation method based on mixed reality technology according to claim 4, characterized in that, The analysis of the simulated impact event results assesses the impact on the structural integrity of bridges and ships, specifically including: S51. After the simulated impact event is completed, the system automatically enters the result analysis stage, extracting structural deformation and stress distribution from the simulation data. And identification of potential damage areas; S52. Apply structural integrity analysis algorithms based on simulated stress distribution. By comparing stress values ​​with the material's yield strength and material strength properties, the safety of a structure under specific impact forces can be assessed. and maximum bearing strength The safety of a structure is determined by the following formula: in, This represents the safety factor of the structure at point (x, y, z). A value greater than 1 indicates safety, while a value less than 1 indicates potential risk. S53. For identified potential damage areas, a damage assessment model is used to analyze the degree of damage and its potential impact. Combining the fatigue characteristics of structural materials and the theory of cumulative damage, the cumulative damage effect of the structure under continuous or repeated impacts is evaluated. The degree of damage is expressed using a damage index. express: Where N is the number of impacts, It is the stress range generated by the nth impact. is the fatigue limit of the material, and m is the fatigue index of the material; Based on the evaluation results of S54, the comprehensive safety factor, and the damage index, the system will generate a comprehensive evaluation report.

8. A bridge and ship impact simulation and evaluation system based on mixed reality technology, the system being used to implement the bridge and ship impact simulation and evaluation method based on mixed reality technology as described in claim 1, characterized in that, include: The system includes modules for creating mixed reality environments, constructing impact event datasets, building computational models that combine deep neural networks with multiphysics coupling, simulating impact events, and analyzing the results of simulated impact events.

Images