SPH-based wave impact boundary force simulation analysis method

Abstract

The invention relates to the technical field of fluid mechanics analysis, in particular to a wave impact boundary force simulation analysis method based on SPH. According to the method, a divergence index is obtained according to the energy accumulation efficiency of each wave in the process from rolling to crushing, the spatial and temporal distribution characteristics of energy release during crushing and a wave height value; dividing all waves into different clusters; the critical time step length is adjusted according to the dispersion degree and divergence index of the waves in the clustering cluster where each wave is located, and the optimized critical time length is obtained; obtaining an interference sensitive area of each wave, and obtaining a modal coupling coefficient according to the correlation intensity and energy distribution of a pressure signal of the interference sensitive area in a time-frequency domain; and according to the optimized critical duration and the modal coupling coefficient, adjusting the reference artificial viscosity coefficient to obtain an optimized artificial viscosity coefficient, and performing wave impact boundary force simulation analysis by using an SPH algorithm. According to the method, the contradiction between the precision and efficiency of multi-scale wave impact boundary force simulation is effectively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of fluid mechanics analysis, and in particular to a wave impact boundary force simulation analysis method based on SPH. Background Art

[0002] In the field of marine engineering, accurately understanding the boundary force characteristics generated by waves impacting marine structures is a key basis for assessing structural safety and optimizing structural design. As a highly promising meshless numerical simulation method, Smoothed Particle Hydrodynamics (SPH) demonstrates significant advantages in simulating complex flow problems such as wave impacts, thanks to its ability to handle large free surface deformations and fluid-structure interaction without requiring mesh construction. Compared to traditional mesh-based methods such as the finite element method and the finite volume method, the SPH algorithm can more efficiently capture dynamic processes such as wave breaking and water splashing, providing more intuitive visualization results and richer quantitative data for in-depth exploration of the interaction mechanisms between waves and marine structures.

[0003] However, in the actual simulation of wave impact boundary forces, the scale of waves in the ocean varies significantly. From small-scale breaking waves near the shore to large-scale waves in the open sea, the energy transfer and force distribution characteristics of the impact structures vary greatly. However, existing SPH methods usually use fixed particle resolution and calculation parameters, which are difficult to adapt to the impact characteristics of waves of different scales. When simulating small-scale waves, if the particle resolution is not enough, details such as local pressure concentration will be missed; when simulating large-scale waves, high resolution will lead to a sharp increase in the number of particles, greatly increasing the computational cost, and even making it impossible to complete long-period simulations, seriously affecting the simulation accuracy and engineering applicability. Summary of the Invention

[0004] In order to solve the technical problem that the fixed parameters in the SPH method lead to low accuracy in the simulation of multi-scale wave impact boundary forces, the present invention aims to provide a wave impact boundary force simulation and analysis method based on SPH. The technical solution adopted is as follows: The present invention proposes a wave impact boundary force simulation analysis method based on SPH, which includes: Obtaining pressure and strain signals at test points on the surface of the structure under each wave impact, where the waves have different scales; The divergence index of each wave is obtained based on the energy accumulation efficiency of each wave from rolling to breaking, the spatiotemporal distribution characteristics of energy release during breaking, and the wave height value; all waves are divided into different clusters; the critical time step is adjusted based on the degree of disorder of the frequency domain energy distribution of the wave pressure signal and the degree of dispersion of the delay of the structural deformation relative to the pressure within each cluster, and the divergence index, to obtain the optimized critical time length of each wave; Based on the fluctuation degree of the pressure signal at each test point on the surface of the structure under each wave impact, the interference sensitive area of ​​each wave is obtained; based on the correlation strength and energy distribution of the pressure signal in the time-frequency domain of each wave interference sensitive area, the modal coupling coefficient of each wave is obtained; Obtain a baseline artificial viscosity coefficient for each scale of wave; adjust the baseline artificial viscosity coefficient according to the optimized critical time and the modal coupling coefficient to obtain the optimized artificial viscosity coefficient for each scale of wave, and use the SPH algorithm to perform wave impact boundary force simulation analysis.

[0005] Furthermore, obtaining the divergence index of each wave includes: Obtain the wave image of each wave at each moment during its test period, perform edge detection on the wave image to extract the wave boundary line, and calculate the fractal dimension of the wave image based on the spatial position distribution of the wave boundary line; For each wave in its test period, the ratio of the fractal dimension difference between the latter and the previous moments and the time interval is taken as the dimension change rate. All dimension change rates are curve-fitted to obtain the dimension curve. The peak point and the first maximum point of the dimension curve are recorded as the breaking moment and the rolling moment of each wave respectively. Get the wave height of each wave at the time of rolling up and breaking, as well as the arrival time of each wave after breaking; The ratio of the wave height at the moment of rolling and breaking of each wave is recorded as the energy accumulation value; The ratio of the fractal dimension of the wave image of each wave at the breaking moment to the arrival time after the breaking is recorded as the complex potential energy index; The ratio of the wave height of each wave at the breaking moment to the average wave height of all waves at the corresponding breaking moment is recorded as the probability of deformed wave. The divergence index of each wave is obtained according to the energy accumulation value, the complex potential energy index and the deformed wave possibility.

[0006] Furthermore, all waves are divided into different clusters, including: Calculate the mean pressure signal and the mean strain signal of all test points on the surface of the structure under each wave impact, and record them as the overall pressure signal and the overall strain signal of the corresponding wave respectively; The overall pressure signal is decomposed by wavelet packets to obtain the wavelet packet decomposition coefficients corresponding to each decomposition frequency band; the square sum of all wavelet packet decomposition coefficients in each decomposition frequency band is calculated as the energy representation value; the ratio of the energy representation value of each decomposition frequency band to the sum of the energy representation values ​​of all decomposition frequency bands is recorded as the energy proportion of each decomposition frequency band; the information entropy of the energy proportion of all decomposition frequency bands corresponding to the overall pressure signal is obtained, which is recorded as the modal energy entropy; The corresponding moments of the maximum points on the overall pressure signal and the overall strain signal of the same wave are recorded as the pressure analysis moment and the deformation analysis moment respectively; the time interval between each pressure analysis moment and its next adjacent deformation analysis moment is obtained respectively, and the ratio of the corresponding time intervals of all pressure analysis moments is taken as the deformation hysteresis duration; The mean of the time intervals between all adjacent maximum points on the overall pressure signal is calculated as the wave period; the ratio of the deformation lag time to the wave period is used as the hysteresis coefficient; the modal energy entropy and the hysteresis coefficient constitute a characteristic binary group of each wave; based on the Euclidean distance of the characteristic binary groups of different waves, all waves are clustered to obtain several clusters.

[0007] Furthermore, obtaining the optimized critical duration of each wave includes: Obtaining a critical time step; using the ratio of the modal energy entropy to the hysteresis coefficient as the force-structure coupling factor of each wave; and obtaining the coefficient of variation of the force-structure coupling factors of all waves in the cluster in which each wave is located; negatively correlating the coefficient of variation of each wave with the product of the divergence index of each wave when it reaches the structure and normalizing the result to obtain a time adjustment coefficient for each wave; The critical time step is weighted using the time adjustment coefficient to obtain the optimized critical time length of each wave.

[0008] Furthermore, obtaining the interference sensitive area includes: The pressure signals of all test points on the surface of the structure under each wave impact are interpolated and fused to generate a spatial pressure field. The spatial pressure field is numerically differentiated to obtain a pressure gradient field. The Poincaré index of the singular point in the pressure gradient field is calculated, and the singular point with a non-zero index is recorded as a force field singular point. All force field singularities are divided into several clusters based on the Euclidean distance between the singularities in the force field. The area enclosed by the convex hull boundary of the cluster corresponding to the maximum number of singularities in the cluster is recorded as the interference sensitive area of ​​each wave.

[0009] Furthermore, obtaining the modal coupling coefficient includes: Decompose the coupled pressure signal of the interference-sensitive area of ​​each wave into different component signals, calculate the mutual information entropy between each component signal and all preset single-scale wave reference signals, and perform normalization processing. Select the component signal with the mutual information entropy normalization result greater than or equal to the preset threshold as the analysis signal; The ratio of the integral of the square of the amplitude of each analysis signal to the sum of the integrals of the square of the amplitudes of all analysis signals is taken as the energy contribution value of each analysis signal; The coupled pressure signal in the interference-sensitive area is decomposed to obtain component signals of different frequencies. The two component signals with the largest frequency difference are selected as target signals, and the two target signals are converted into the time-frequency domain to obtain wavelet coefficients. Based on the wavelet coefficients, the mutual wavelet spectrum between the two target signals and the wavelet power spectrum of each target signal are calculated. The wave coherence coefficient is obtained by taking the ratio of the square of the modulus of the mutual wavelet spectrum and the product of the wavelet power spectra of the two target signals. The two analysis signals are linearly superimposed in the time domain to generate a composite signal. The difference between the integral of the square of the composite signal's amplitude and the sum of the energy contributions of the two target signals is used as the numerator, and the sum of the energy contributions of the two target signals is used as the denominator to obtain the ratio, which is used as the coupling ratio. The modal coupling coefficient of each wave is obtained according to the coupling ratio and the wave correlation coefficient.

[0010] Furthermore, obtaining the optimized artificial viscosity coefficient of each scale wave includes: The product of the ratio of the optimized critical duration of each wave to the critical time step and the modal coupling coefficient is used as the duration correction coefficient of each wave; the average of the duration correction coefficients of waves of the same scale is recorded as the final correction coefficient; The reference artificial viscosity coefficient of each scale wave is weighted by using the sum of the constant 1 and the final correction coefficient to obtain the optimized artificial viscosity coefficient of the corresponding scale wave.

[0011] Furthermore, the method for decomposing the coupled pressure signal into different component signals is a variational mode decomposition algorithm.

[0012] Furthermore, the coupled pressure signal of the interference sensitive area of ​​each wave is an average signal of the pressure signals of all test points in the interference sensitive area of ​​each wave under the impact of the corresponding wave.

[0013] Furthermore, the preset threshold is 0.3.

[0014] The present invention has the following beneficial effects: In an embodiment of the present invention, the characteristic discreteness of all waves in the cluster where each wave is located represents the stability of the coupling between force and structure at the wave scale, reflects the significance of the temporal difference of force transmission between waves, and the divergence index reflects that the boundary force is unevenly distributed on the surface of the structure and changes dynamically. The above two factors are combined to dynamically adjust the time step in the SPH algorithm, so that when small-scale waves hit, the fine step is automatically enabled to accurately capture the instantaneous force peak, and when large-scale waves hit, the step is relaxed to efficiently simulate the cumulative effect, solving the contradiction between the small-scale leakage of key features and the large-scale resource consumption under a fixed step size; the interference sensitive area provides an accurate spatial target for interference correction, and the pressure signal of the interference sensitive area of ​​each wave is in the time-frequency domain. The wave energy cascade is analyzed based on the correlation intensity and energy distribution in the wave model, the nonlinear interference intensity is accurately quantified to guide the dynamic correction of the inter-particle force, and it is simplified to linear superposition in weak coupling to avoid over-adjustment; according to the optimized critical time and modal coupling coefficient, the benchmark artificial viscosity coefficient is adjusted, and the optimized artificial viscosity coefficient of each scale wave is adaptively determined to ensure that small-scale transient details are not lost due to enhanced viscosity and large-scale energy transfer is not distorted due to the relaxation of step size. Finally, a full-chain optimization of "scale identification-time adaptation-interference elimination" is formed within the SPH framework, which solves the contradiction between accuracy and efficiency of multi-scale wave impact boundary force simulation and provides high-fidelity and high-reliability numerical support for the wave-resistant design of marine structures. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0016] Figure 1 A flowchart of a wave impact boundary force simulation analysis method based on SPH according to one embodiment of the present invention; Figure 2 A schematic diagram of a computer device for simulating and analyzing wave impact boundary forces based on SPH provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0017] To further illustrate the technical means and effectiveness of the present invention in achieving its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, describes in detail the specific implementation, structure, features, and effectiveness of a wave impact boundary force simulation and analysis method based on SPH proposed in accordance with the present invention. In the following description, different references to "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.

[0018] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0019] The following describes in detail a specific solution of a wave impact boundary force simulation analysis method based on SPH provided by the present invention with reference to the accompanying drawings.

[0020] Example 1:

[0021] This paper proposes a wave impact boundary force simulation analysis method based on SPH, please refer to Figure 1 , which shows a flowchart of a wave impact boundary force simulation analysis method based on SPH provided by one embodiment of the present invention, the method comprising: Step S1: Obtain the pressure signal and strain signal of each wave impact at the test point on the surface of the structure. Waves have different scales.

[0022] Combining wave tank experiments with field observations, a high-speed camera was mounted on a fixed platform to ensure full coverage of the target waters. A wave height meter array and acoustic Doppler current meters were deployed simultaneously in the target waters. The test period for a single wave was from the moment the leading edge of the wave entered the high-speed camera's field of view until the trailing edge of the wave completely left it.

[0023] During the test period, a high-speed camera continuously captured wave images at a sampling rate of 1000 frames per second. A hardware-triggered synchronization device ensured strict alignment of the sampling times of the high-speed camera, wave height meter array, and acoustic Doppler current meter, enabling the synchronous acquisition of multiple parameters. Specifically, the wave height meter array collected the wave surface elevation at each moment, recording it as the wave height value. Based on the wave height values ​​at all moments during the test period, the average time interval between all adjacent maximum points was used as the wave period, and the average horizontal spacing between adjacent wave crests in the wave height meter spatial distribution was recorded as the wavelength. The acoustic Doppler current meter was used to measure the wave propagation velocity, which is equal to the ratio of wavelength to wave period. Each wave corresponds to a wave period, propagation velocity, and wavelength.

[0024] In one implementation of the present invention, small-scale waves have a wavelength less than 5 meters, large-scale waves have a wavelength greater than 20 meters, and medium-scale waves have a wavelength between 5 and 20 meters. At least 30 independent samples are collected for each wave size.

[0025] An array of micro-pressure sensors is placed on the surface of the structure, with each sensor located as a test point. Based on a high-speed camera image sequence, the movement trajectories of the leading and trailing edges of the wave are tracked using optical flow or feature point matching to determine the precise moment the wave reaches the structure. When the wave reaches the structure, the pressure and strain history signals of each test point under each wave impact are synchronously collected. The pressure history signal corresponds to a time period covering the time interval from the start of each wave impact to the complete dissipation of the pressure. The strain history signal is triggered synchronously with the pressure history signal and extends until the vibration of the structure decays to a steady state, ensuring that the entire dynamic response process of the structure is included.

[0026] In one implementation of the embodiment of the present invention, the acquisition frequency of the pressure sensor is set to 1000 Hz.

[0027] Step S2: According to the energy accumulation efficiency of each wave from rolling to breaking, the spatiotemporal distribution characteristics of energy release during breaking, and the wave height value, the divergence index of each wave is obtained; all waves are divided into different clusters; according to the discrete degree and divergence index of the frequency domain energy distribution disorder of the pressure signal of the waves in the cluster where each wave is located and the delay degree of the structural deformation relative to the pressure, the critical time step is adjusted to obtain the optimized critical time of each wave.

[0028] When high-energy waves impact structures, their kinetic and potential energies are converted into more complex fluid motions. This energy diffuses in three dimensions through turbulent eddies and wave breaking, and boundary forces are no longer locally concentrated. High interface complexity leads to more microscale eddies when waves break, dispersing boundary forces over a wider surface area. Large wave heights are typically associated with long wavelengths, which enhances diffraction effects and results in a more dispersed force field. Therefore, by combining energy accumulation efficiency, the spatiotemporal distribution of energy released during breaking, and wave height, we analyze the degree of dispersion of the spatial distribution of wave boundary forces on the surface of structures and derive a divergence index.

[0029] Waves in the same cluster correspond to waves of a certain scale. The characteristic discreteness of all waves in the cluster to which each wave belongs represents the stability of the coupling between force and structure at the wave scale, and reflects the significance of the temporal difference in force transmission between waves. In order to avoid smoothing or missed detection, the time step should be smaller. The divergence index reflects that the boundary force is distributed unevenly and dynamically on the surface of the structure. When the distribution changes rapidly, the coarse time scale cannot capture the instantaneous transfer of spatial loads, and the time step should be smaller. The above two factors are combined to dynamically adjust the time step in the SPH algorithm, so that when small-scale waves hit, the fine step is automatically enabled to accurately capture the instantaneous force peak, and when large-scale waves hit, the step is relaxed to efficiently simulate the cumulative effect, thus resolving the contradiction between missing key features at small scales and consuming resources at large scales under fixed step sizes.

[0030] Step S3: Based on the fluctuation degree of the pressure signal at each test point on the surface of the structure under each wave impact, the interference sensitive area of ​​each wave is obtained; based on the correlation strength and energy distribution of the pressure signal in the time-frequency domain of each wave interference sensitive area, the modal coupling coefficient of each wave is obtained.

[0031] Updating the time step is a fine-grained depiction of the force-structure coupling of single-scale waves. However, in reality, complex ocean environments are subject to the simultaneous action of multiple-scale waves. For example, nearshore areas are subject to the combined effects of long-period waves in the open ocean and short-period breakers near the shore. Existing SPH algorithms lack an effective multi-scale coupling mechanism, making it difficult to accurately simulate the mutual interference of waves of different scales impacting structures and the superposition of boundary forces. For example, when waves of different scales impact together, the forces are not simply added together; they interfere with each other, much like a large wave carrying smaller waves, which complicates the forces. Interference-sensitive zones provide precise spatial targets for interference correction. The correlation strength and energy distribution of the pressure signals in each wave's interference-sensitive zone in the time-frequency domain analyze the presence of wave energy cascades. The nonlinear interference intensity is accurately quantified to guide the dynamic correction of inter-particle forces. This means that when strong coupling occurs, nonlinear interactions are strengthened to restore the energy cascade effect, while when weak coupling occurs, linear superposition is simplified to avoid over-adjustment.

[0032] Step S4: Obtain the baseline artificial viscosity coefficient for each scale of waves; adjust the baseline artificial viscosity coefficient according to the optimized critical time and modal coupling coefficient to obtain the optimized artificial viscosity coefficient for each scale of waves, and use the SPH algorithm to simulate and analyze the wave impact boundary force.

[0033] Based on the interference quantification results, the particle interaction parameters in the SPH algorithm are dynamically adjusted to achieve accurate simulation of boundary forces. In order to match the coupling mechanism of the actual wave scale, a collaborative mechanism of the time step and the coupling coefficient is established, and the artificial viscosity coefficient is jointly regulated to ensure that small-scale transient details are not lost due to enhanced viscosity and large-scale energy transfer is not distorted due to the relaxation of the step size. Ultimately, a full-chain optimization of "scale identification-time adaptation-interference elimination" is formed within the SPH framework, which solves the contradiction between accuracy and efficiency in the simulation of multi-scale wave impact boundary forces, and provides high-fidelity and high-reliability numerical support for the wave-resistant design of marine structures.

[0034] Preferably, in some possible implementations of the present invention, the method for obtaining the divergence index includes: obtaining the wave image of each wave at each moment in its test period, performing edge detection on the wave image to extract the wave boundary line, and calculating the fractal dimension of the wave image based on the spatial position distribution of the wave boundary line; for each wave in its test period, for each two adjacent moments, taking the ratio of the fractal dimension difference between the latter moment and the previous moment and the time interval as the dimension change rate, performing curve fitting on all dimension change rates to obtain a dimension curve; recording the peak point of the dimension curve and the corresponding moment of the first maximum point as each moment in turn. The breaking and rolling moments of each wave are determined; the wave height values ​​of each wave at the rolling and breaking moments, as well as the arrival time after breaking, are obtained; the ratio of the wave height values ​​of each wave at the rolling and breaking moments is recorded as the energy accumulation value; the ratio of the fractal dimension of the wave image of each wave at the breaking moment to the arrival time after breaking is recorded as the complex potential energy index; the ratio of the wave height value of each wave at the breaking moment to the average wave height value of all waves at the corresponding breaking moment is recorded as the abnormal wave possibility; based on the energy accumulation value, the complex potential energy index and the abnormal wave possibility, the divergence index of each wave is obtained. Among them, this embodiment uses the box counting method to calculate the fractal dimension, and the random walk method or differential box counting method can also be used.

[0035] It's important to note that the fractal dimension reflects the complexity of the boundary line where waves meet the air, and the dimensionality change rate reflects the dynamic characteristics of the boundary line's morphological evolution. When a wave reaches the critical breaking state, the motion of water particles at the wave crest accelerates dramatically, leading to a sudden change in the topological structure of the contact surface and a significant increase in the fractal dimension change rate. Therefore, the dimensionality change rate is a key characteristic value for determining the transition between wave breaking stages. The specific method for obtaining the dimensionality curve is to construct a two-dimensional space with time as the horizontal axis and the dimensionality change rate as the vertical axis. The dimensionality change rate of each wave at two adjacent moments during its test period is mapped into the two-dimensional space to obtain corresponding scatter points. These scatter points are then fitted using the least squares method, and the resulting fitted curve is recorded as the dimensionality curve. The moment corresponding to the first maximum point of the dimensionality curve represents the wave entering the curling stage, at which the interface begins to bend but does not break. The moment corresponding to the peak point represents the wave entering the breaking stage, at which the interface violently curls and breaks into a gas-liquid mixture.

[0036] The energy accumulation value represents the accumulated energy of a wave from its inception. A higher energy accumulation value indicates no breakup or energy loss has occurred since the wave began to roll, indicating greater wave energy accumulation. Furthermore, when a high-energy wave impacts a structure, its kinetic and potential energy are converted into more complex fluid motion. This energy diffuses in three dimensions through turbulent eddies and wave breaking, reducing the localized boundary force and resulting in a more dispersed distribution of boundary forces on the structure's surface. A higher complex potential energy index indicates greater complexity at each wave-air interface. Furthermore, a shorter time after a wave breaks to reach the structure's boundary—in other words, greater wave potential energy—indicates higher interface complexity and greater wave potential energy. Furthermore, higher interface complexity generates more microscale eddies when the wave breaks, distributing boundary forces over a wider surface area and resulting in a more dispersed distribution of boundary forces on the structure's surface. A higher freak wave probability indicates a greater amplification of the boundary force upon wave breaking. Large wave heights are typically associated with longer wavelengths, which enhances diffraction effects and a more dispersed force field, resulting in a more dispersed distribution of boundary forces on the structure's surface. Therefore, the energy accumulation value, complex potential energy index and distorted wave probability are all positively correlated with the divergence index. In the embodiment of the present invention, the product of the energy accumulation value, complex potential energy index and distorted wave probability of each wave is used as the divergence index.

[0037] It is important to note that the amplitude at the peak of the dimensionality curve is the maximum amplitude of all the data on the dimensionality curve; the arrival time after each wave is broken is the ratio of the distance from the wave's position at the moment of breaking to the structure boundary to the wave propagation speed. A 3D laser scanner is used to obtain the 3D coordinates of the structure surface, and the shortest distance from the wave boundary to the structure boundary is used as the distance from the wave boundary to the structure boundary.

[0038] Preferably, in some possible implementations of the embodiments of the present invention, the clustering method includes: respectively calculating the mean signal of the pressure signal and the mean signal of the strain signal of all test points on the surface of the structure under each wave impact, and recording them as the overall pressure signal and the overall strain signal of the corresponding wave in turn; performing wavelet packet decomposition on the overall pressure signal to obtain the wavelet packet decomposition coefficients corresponding to each decomposition frequency band; calculating the sum of the squares of all wavelet packet decomposition coefficients in each decomposition frequency band as the energy representation value; recording the ratio of the energy representation value of each decomposition frequency band to the sum of the energy representation values ​​of all decomposition frequency bands as the energy proportion of each decomposition frequency band; obtaining information on the energy proportion of all decomposition frequency bands corresponding to the overall pressure signal entropy, recorded as modal energy entropy; the corresponding moments of the maximum points on the overall pressure signal and the overall strain signal of the same wave are recorded as the pressure analysis moment and the deformation analysis moment respectively; the time interval between each pressure analysis moment and its next adjacent deformation analysis moment is obtained respectively, and the ratio of the corresponding time intervals of all pressure analysis moments is used as the deformation lag time; the mean of the time intervals between all adjacent maximum points on the overall pressure signal is calculated as the wave period; the ratio of the deformation lag time to the wave period is used as the lag coefficient; the modal energy entropy and the lag coefficient constitute the characteristic binary group of each wave; based on the Euclidean distance of the characteristic binary groups of different waves, all waves are clustered to obtain several clusters.

[0039] It should be noted that small-scale waves are susceptible to disturbances such as wind and turbulence, producing complex fluctuations with multiple frequency bands superimposed, dispersing their energy across a wide frequency band. Large-scale waves are primarily driven by gravity, with their energy concentrated in the primary frequency band. Mesoscale waves have both high- and low-frequency components, with their energy concentrated in several significant frequency bands. Therefore, when the modal energy entropy is large, the waves are small-scale; when the modal energy entropy is small, the waves are large-scale; and when the modal energy entropy is compromised, the waves are in a multimodal mixed state. Small-scale waves act for a short time, allowing local deformation of the structure to follow immediately. In other words, the boundary forces of small-scale waves are nearly synchronous with the structural deformation. However, the long-lasting loads of large-scale waves induce global vibration or inertial effects, requiring longer periods of time for displacement accumulation, causing the structural deformation caused by large-scale waves to lag behind the boundary forces. The modal energy entropy reflects the frequency-domain characteristics of the wave itself, while the hysteresis coefficient reflects the time-domain coupling relationship between the wave and the structure. Together, they encompass the dynamic characteristics of the wave and the structural response behavior.

[0040] In one implementation of the embodiment of the present invention, a K-means clustering algorithm is used to cluster waves, where K is 3.

[0041] Preferably, in some possible implementation methods of the embodiments of the present invention, the method for obtaining the optimized critical duration includes: obtaining the critical time step; taking the ratio of the modal energy entropy to the hysteresis coefficient as the force-structure coupling factor of each wave; obtaining the coefficient of variation of the force-structure coupling factors of all waves in the cluster in which each wave is located; negatively correlating and normalizing the product of the coefficient of variation of each wave and the divergence index when each wave reaches the structure to obtain the time adjustment coefficient of each wave; and using the time adjustment coefficient to weight the critical time step to obtain the optimized critical duration of each wave.

[0042] It should be noted that a larger force-structure coupling factor for each wave indicates transient strong coupling between the wave and the structure, and the wave may be a small-scale wave. Conversely, a larger force-structure coupling factor indicates a gradual weak coupling between the wave and the structure, and the wave may be a large-scale wave. A larger coefficient of variation of the force-structure coupling factor for all waves within the cluster to which each wave belongs indicates a more unstable force-structure coupling at the wave scale, and significant temporal variability in force transmission between waves. To avoid smoothing out transient impact peaks or missing short-term resonance effects, the time step should be smaller. A larger divergence index indicates that the boundary force is non-uniformly distributed and dynamically changes on the surface of the structure. When the distribution changes rapidly, the coarse time scale cannot capture the instantaneous transfer of spatial loads, and the time step should be smaller. Therefore, both the coefficient of variation and the divergence index are negatively correlated with the optimization critical time. In an embodiment of the present invention, the coefficient of variation of the force-structure coupling factor of all waves in the cluster to which each wave belongs is negatively correlated with the product of the divergence index of each wave when it reaches the structure and normalized to obtain a time adjustment coefficient; the time adjustment coefficient is weighted with the critical time step to obtain the optimized critical duration of each wave.

[0043] In this embodiment, the data to be measured is used as the exponent of an exponential function with a natural constant as the base to achieve negative correlation and normalization of the data to be measured. Negative correlation mapping and Sigmoid function normalization can also be achieved by taking the inverse, which is not limited here.

[0044] It should be noted that the critical time step in the SPH algorithm is calculated by first calculating the numerical stability upper limit of the safe ratio of the particle motion length to ensure non-supersmoothness according to the CFL condition, and then combining the calibration accuracy of key physical characteristics such as the pressure peak and period of typical wave impact, and finally taking the minimum value of the two constraints. The minimum value is the critical value for ensuring numerical stability and physical accuracy. The CFL time step determined by the CFL condition is calculated as follows: the ratio of the wave particle smooth length as the numerator and the sum of the sound speed and the propagation speed as the denominator is calculated, and the product of this ratio and the Courant number is used as the CFL time step, where the Courant number is 0.2. The calibration time step determined by the key physical characteristics of typical wave impact is calculated as follows: the ratio of the wave period to 50 and the ratio of the time it takes for the pressure signal to rise from static pressure to peak to 10 are calculated, and the minimum of the two ratios is recorded as the calibration time step. The minimum of the CFL time step and the calibration time step is recorded as the critical time step for each wave.

[0045] Preferably, in some possible implementation methods of the embodiments of the present invention, the method for obtaining the interference sensitive area includes: interpolating and fusing the pressure signals of all test points on the surface of the structure under each wave impact to generate a spatial pressure field; numerically differentiating the spatial pressure field to obtain a pressure gradient field; calculating the Poincaré index of the singularity in the pressure gradient field, and recording the singularity with a non-zero index as a force field singularity; dividing all the force field singularities into a number of clusters based on the Euclidean distance between the force field singularities, and selecting the area enclosed by the convex hull boundary of the cluster whose number of singularities in the cluster is greater than a preset number threshold as the interference sensitive area of ​​each wave.

[0046] It should be noted that this embodiment uses radial basis functions for interpolation and fusion, and numerically differentiates the spatial pressure field using the central difference method. The Poincaré index of the pressure gradient field describes the rotational characteristics of the gradient field, which is the center of concentration of interference energy, with higher pressure fluctuation amplitude than the surrounding area. Interference-sensitive areas provide precise spatial targets for interference correction. This solution uses the K-means clustering algorithm for clustering, with the K value determined by the elbow method.

[0047] Preferably, in some possible implementations of the embodiments of the present invention, the method for obtaining the modal coupling coefficient includes: decomposing the coupled pressure signal of the interference-sensitive area of ​​each wave into different component signals, calculating the mutual information entropy between each component signal and all preset single-scale wave reference signals and performing normalization processing, selecting the component signal with a mutual information entropy normalization result greater than or equal to a preset threshold as the analysis signal; taking the ratio of the integral of the square of the amplitude of each analysis signal to the sum of the integrals of the square of the amplitudes of all analysis signals as the energy contribution value of each analysis signal; decomposing the coupled pressure signal of the interference-sensitive area to obtain component signals of different frequencies, and selecting the component signal with the largest frequency difference. The two largest component signals are recorded as target signals, and the two target signals are converted into time-frequency domain respectively to obtain wavelet coefficients; based on the wavelet coefficients, the mutual wavelet spectrum between the two target signals and the wavelet power spectrum of each target signal are calculated; the wave coherence coefficient is obtained by taking the ratio of the square of the modulus value of the mutual wavelet spectrum and the product of the wavelet power spectrum of the two target signals; the two analysis signals are linearly superimposed in the time domain to generate a synthetic signal; the difference between the integral of the square of the amplitude of the synthetic signal and the sum of the energy contribution values ​​of the two target signals is taken as the numerator, and the sum of the energy contribution values ​​of the two target signals is taken as the denominator to obtain the ratio as the coupling ratio; according to the coupling ratio and the wave correlation coefficient, the modal coupling coefficient of each wave is obtained.

[0048] It should be noted that, in an embodiment of the present invention, the mean signal of the overall pressure signal of all waves in each cluster is calculated and recorded as a preset single-scale wave reference signal, and three clusters correspond to three preset single-scale wave reference signals. This embodiment uses the empirical mode decomposition algorithm to decompose the coupled pressure signal, and selects the wavelet transform algorithm to convert the target signal into the time-frequency domain. Each component signal obtained by the signal decomposition of the coupled pressure signal corresponds to the wave contribution of a single scale. The low mutual information entropy represents that the component signal is almost independent of the reference signal, indicating that the component signal may be noise or interference and does not need to be analyzed. If the wave correlation coefficient is larger, it indicates that there is an energy cascade, that is, nonlinear energy transfer from large-scale waves to medium-scale waves, and from medium-scale waves to small-scale waves. The cascade efficiency decreases as the scale difference increases.

[0049] When the difference between the integral of the square of the amplitude of the composite signal and the sum of the energy contribution values ​​of the two target signals is greater than 0, it means that the coupling has produced energy amplification, such as large waves pushing small waves, resulting in a stronger impact; conversely, it means that the coupling has produced energy cancellation, such as small waves breaking and consuming the energy of large waves; the larger the wave coherence coefficient, the more active the nonlinear energy transfer between the two-scale waves. The modal coupling coefficient reflects the strength of the nonlinear interference caused by the active coupling of the two-scale waves. If the modal coupling coefficient of each wave is larger, the wave is a strong coupling interference. At this time, the energy is nonlinearly superimposed. The SPH model needs to specifically strengthen the nonlinear interaction between the two-scale particles to accurately restore the complex impact of large waves with small waves; conversely, the wave is a weak coupling interference, which can be approximated by linear superposition. The model can be simplified to a linear superposition correction to avoid excessive adjustment and the introduction of new errors. Therefore, the coupling ratio and the wave correlation coefficient are both positively correlated with the modal coupling coefficient. In an embodiment of the present invention, the product of the coupling ratio of each wave and the wave correlation coefficient is used as the modal coupling coefficient.

[0050] In one implementation of the embodiment of the present invention, the preset threshold is set to 0.3.

[0051] In an embodiment of the present invention, a time-frequency analysis is performed on the collected mixed wave pressure signal, and the dominant frequency and corresponding wave height with the highest energy proportion are determined through Hilbert transform or wave height meter calibration, which are recorded as dominant parameters; the wave closest to the dominant parameter is selected from the single-scale waves, and the matching criterion is the minimum Euclidean norm. The single-scale wave that successfully matches is recorded as the reference wave; SPH numerical simulation is performed on the reference wave, and the artificial viscosity coefficient is initialized to a range of 0.01 to 1. The artificial viscosity coefficient is iteratively adjusted through bisection until the following convergence conditions are met, such as the relative error of the pressure peak is less than 5%. The optimal solution artificial viscosity system is used as the reference artificial viscosity coefficient for the wave of that scale. It should be noted that the single-scale wave can be waves of all scales collected in step S1, or waves of each scale collected by impact, and the scale division is the same as the scale division standard in step S1.

[0052] Preferably, in some possible implementations of the embodiments of the present invention, the method for obtaining the optimized artificial viscosity coefficient includes: taking the product of the ratio of the optimized critical duration of each wave to the critical time step and the modal coupling coefficient as the duration correction coefficient of each wave; recording the average of the duration correction coefficients of waves of the same scale as the final correction coefficient; and using the sum of a constant 1 and the final correction coefficient to weight the baseline artificial viscosity coefficient of waves of each scale to obtain the optimized artificial viscosity coefficient of waves of the corresponding scale.

[0053] It should be noted that the duration correction coefficient binds the fineness of the time scale to the strength of the coupling interference, making the adjustment of the viscosity coefficient scene-adaptive. Small-scale waves require a fine time step, that is, a small critical time step, to capture high-frequency impact details. The optimization critical time is small. If the viscosity is too high, it will smooth out the instantaneous force peak. Therefore, only a small increase in viscosity is needed to match weak nonlinearities and avoid destroying details. Large-scale waves use a coarse time step to simulate cumulative effects. The optimization critical time is large. When the coupling is strong, the energy cascade is active, such as when large waves push small waves. The viscosity needs to be greatly increased to simulate nonlinear energy transfer and compensate for the loss of coupling details caused by the coarse time step.

[0054] If the scale of the newly collected wave belongs to the scale range of waves of a certain scale, the optimized artificial viscosity coefficient of waves of this scale will be used for analysis when conducting wave impact boundary force simulation analysis.

[0055] When using the SPH algorithm to simulate wave impact boundaries, the three-dimensional space containing wave generation, impact and wave-breaking zones is first delineated, and the ocean scene and structure size are input to determine the range; then, seawater particles with mass, density and initial velocity are filled in the fluid zone and initialized according to the wave type and distribution rules. The particle resolution determines the fineness; fixed particles are then used to construct a structural model, with key parts densely distributed, and the structural shape and surface features are input; then, the particle interaction rules are set, and the motion is calculated through kernel functions, state equations and force balance, and the fluid characteristics are input to obtain the calculation model; finally, the simulation is started to allow wave particles to impact the structure to generate boundary forces. After processing in the wave-breaking zone, the force output time history curve and distribution cloud map are statistically analyzed to provide data for the project.

[0056] So far, the present invention is completed.

[0057] Example 2:

[0058] The present invention also proposes a computer device schematic diagram of a wave impact boundary force simulation and analysis device based on SPH, please refer to Figure 2 The computer device includes a memory 501, a processor 502, and a computer program 503 stored in the memory 501 and running on the processor 502, wherein when the processor 502 executes the computer program 503, the computer device can execute any one of the SPH-based wave impact boundary force simulation analysis methods introduced above.

[0059] In addition, an embodiment of the present application also protects a device, which may include a memory and a processor, wherein the memory stores executable program code, and the processor is used to call and execute the executable program code to execute an SPH-based wave impact boundary force simulation analysis method provided in an embodiment of the present application.

[0060] In this embodiment, the device can be divided into functional modules based on the above-described method examples. For example, each functional module can be mapped to a specific functional module, or two or more functions can be integrated into a single processing module. The integrated module can be implemented in hardware. It should be noted that the module division in this embodiment is illustrative and represents only a logical functional division. In actual implementation, other division methods may be used.

[0061] In the case of dividing the modules into modules corresponding to their functions, the device may further include a communication module, a signal analysis module, a complexity analysis module, a positioning module, etc. It should be noted that all relevant contents of the various steps involved in the above method embodiment can be referred to the functional description of the corresponding functional modules and will not be repeated here.

[0062] It should be understood that the device provided in this embodiment is used to execute the above-mentioned SPH-based wave impact boundary force simulation analysis method, and thus can achieve the same effect as the above-mentioned implementation method.

[0063] In the case of an integrated unit, the device may include a processing module and a storage module. When the device is applied to a device, the processing module may be used to control and manage the operation of the device. The storage module may be used to support the device in executing mutual program codes, etc.

[0064] The processing module may be a processor or controller that implements or executes the various exemplary logic blocks, modules, and circuits disclosed herein. The processor may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a digital signal processor (DSP) and a microprocessor, and the like. The storage module may be a memory.

[0065] Example 3:

[0066] This embodiment also provides a computer-readable storage medium, which stores computer program code. When the computer program code is run on a computer, the computer executes the above-mentioned related method steps to implement a wave impact boundary force simulation analysis method based on SPH provided in the above embodiment.

[0067] Example 4:

[0068] This embodiment also provides a computer program product. When the computer program product is run on a computer, it enables the computer to execute the above-mentioned related steps to implement the wave impact boundary force simulation analysis method based on SPH provided in the above embodiment.

[0069] Among them, the device, computer-readable storage medium, computer program product or chip provided in this embodiment are all used to execute the corresponding methods provided above. Therefore, the beneficial effects that can be achieved can refer to the beneficial effects in the corresponding methods provided above, and will not be repeated here.

[0070] In the embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of modules or units is only a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0071] It should be noted that the order in which the embodiments of the present invention are described above is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0072] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments.

Claims

1. A wave impact boundary force simulation analysis method based on SPH, characterized in that: The method includes: Obtaining pressure and strain signals at test points on the surface of the structure under each wave impact, where the waves have different scales; The divergence index of each wave is obtained based on the energy accumulation efficiency of each wave from rolling to breaking, the spatiotemporal distribution characteristics of energy release during breaking, and the wave height value; all waves are divided into different clusters; the critical time step is adjusted based on the degree of disorder of the frequency domain energy distribution of the wave pressure signal and the degree of dispersion of the delay of the structural deformation relative to the pressure within each cluster, and the divergence index, to obtain the optimized critical time length of each wave; Based on the fluctuation degree of the pressure signal at each test point on the surface of the structure under each wave impact, the interference sensitive area of ​​each wave is obtained; based on the correlation strength and energy distribution of the pressure signal in the time-frequency domain of each wave interference sensitive area, the modal coupling coefficient of each wave is obtained; Obtain a baseline artificial viscosity coefficient for each scale of wave; adjust the baseline artificial viscosity coefficient according to the optimized critical time and the modal coupling coefficient to obtain the optimized artificial viscosity coefficient for each scale of wave, and use the SPH algorithm to perform wave impact boundary force simulation analysis.

2. The SPH-based wave impact boundary force simulation analysis method according to claim 1 is characterized in that: The step of obtaining the divergence index of each wave includes: Obtain the wave image of each wave at each moment during its test period, perform edge detection on the wave image to extract the wave boundary line, and calculate the fractal dimension of the wave image based on the spatial position distribution of the wave boundary line; For each wave in its test period, the ratio of the fractal dimension difference between the latter and the previous moments and the time interval is taken as the dimension change rate. All dimension change rates are curve-fitted to obtain the dimension curve. The peak point and the first maximum point of the dimension curve are recorded as the breaking moment and the rolling moment of each wave respectively. Get the wave height of each wave at the time of rolling up and breaking, as well as the arrival time of each wave after breaking; The ratio of the wave height at the moment of rolling and breaking of each wave is recorded as the energy accumulation value; The ratio of the fractal dimension of the wave image of each wave at the breaking moment to the arrival time after the breaking is recorded as the complex potential energy index; The ratio of the wave height of each wave at the breaking moment to the average wave height of all waves at the corresponding breaking moment is recorded as the probability of deformed wave. The divergence index of each wave is obtained according to the energy accumulation value, the complex potential energy index and the deformed wave possibility.

3. The SPH-based wave impact boundary force simulation analysis method according to claim 1 is characterized in that: The above method divides all waves into different clusters, including: Calculate the mean pressure signal and the mean strain signal of all test points on the surface of the structure under each wave impact, and record them as the overall pressure signal and the overall strain signal of the corresponding wave respectively; The overall pressure signal is decomposed by wavelet packets to obtain the wavelet packet decomposition coefficients corresponding to each decomposition frequency band; the square sum of all wavelet packet decomposition coefficients in each decomposition frequency band is calculated as the energy representation value; the ratio of the energy representation value of each decomposition frequency band to the sum of the energy representation values ​​of all decomposition frequency bands is recorded as the energy proportion of each decomposition frequency band; the information entropy of the energy proportion of all decomposition frequency bands corresponding to the overall pressure signal is obtained, which is recorded as the modal energy entropy; The corresponding moments of the maximum points on the overall pressure signal and the overall strain signal of the same wave are recorded as the pressure analysis moment and the deformation analysis moment respectively; the time interval between each pressure analysis moment and its next adjacent deformation analysis moment is obtained respectively, and the ratio of the corresponding time intervals of all pressure analysis moments is taken as the deformation hysteresis duration; The mean of the time intervals between all adjacent maximum points on the overall pressure signal is calculated as the wave period; the ratio of the deformation lag time to the wave period is used as the hysteresis coefficient; the modal energy entropy and the hysteresis coefficient constitute a characteristic binary group of each wave; based on the Euclidean distance of the characteristic binary groups of different waves, all waves are clustered to obtain several clusters.

4. The SPH-based wave impact boundary force simulation analysis method according to claim 3 is characterized in that: The step of obtaining the optimized critical duration of each wave includes: Obtaining a critical time step; using the ratio of the modal energy entropy to the hysteresis coefficient as the force-structure coupling factor of each wave; and obtaining the coefficient of variation of the force-structure coupling factors of all waves in the cluster in which each wave is located; negatively correlating the coefficient of variation of each wave with the product of the divergence index of each wave when it reaches the structure and normalizing the result to obtain a time adjustment coefficient for each wave; The critical time step is weighted using the time adjustment coefficient to obtain the optimized critical time length of each wave.

5. The SPH-based wave impact boundary force simulation analysis method according to claim 1 is characterized in that: The obtaining of the interference sensitive area of ​​each wave includes: The pressure signals of all test points on the surface of the structure under each wave impact are interpolated and fused to generate a spatial pressure field. The spatial pressure field is numerically differentiated to obtain a pressure gradient field. The Poincaré index of the singular point in the pressure gradient field is calculated, and the singular point with a non-zero index is recorded as a force field singular point. All force field singularities are divided into several clusters based on the Euclidean distance between the singularities in the force field. The area enclosed by the convex hull boundary of the cluster corresponding to the maximum number of singularities in the cluster is recorded as the interference sensitive area of ​​each wave.

6. The SPH-based wave impact boundary force simulation analysis method according to claim 1 is characterized in that: The obtaining of the modal coupling coefficient of each wave includes: Decompose the coupled pressure signal of the interference-sensitive area of ​​each wave into different component signals, calculate the mutual information entropy between each component signal and all preset single-scale wave reference signals, and perform normalization processing. Select the component signal with the mutual information entropy normalization result greater than or equal to the preset threshold as the analysis signal; The ratio of the integral of the square of the amplitude of each analysis signal to the sum of the integrals of the square of the amplitudes of all analysis signals is taken as the energy contribution value of each analysis signal; The coupled pressure signal in the interference-sensitive area is decomposed to obtain component signals of different frequencies. The two component signals with the largest frequency difference are selected as target signals, and the two target signals are converted into the time-frequency domain to obtain wavelet coefficients. Based on the wavelet coefficients, the mutual wavelet spectrum between the two target signals and the wavelet power spectrum of each target signal are calculated. The wave coherence coefficient is obtained by taking the ratio of the square of the modulus of the mutual wavelet spectrum and the product of the wavelet power spectra of the two target signals. The two analysis signals are linearly superimposed in the time domain to generate a composite signal. The difference between the integral of the square of the composite signal's amplitude and the sum of the energy contributions of the two target signals is used as the numerator, and the sum of the energy contributions of the two target signals is used as the denominator to obtain the ratio, which is used as the coupling ratio. The modal coupling coefficient of each wave is obtained according to the coupling ratio and the wave correlation coefficient.

7. The SPH-based wave impact boundary force simulation analysis method according to claim 4 is characterized in that: The optimized artificial viscosity coefficient of each scale wave is obtained, including: The product of the ratio of the optimized critical duration of each wave to the critical time step and the modal coupling coefficient is used as the duration correction coefficient of each wave; the average of the duration correction coefficients of waves of the same scale is recorded as the final correction coefficient; The reference artificial viscosity coefficient of each scale wave is weighted by using the sum of the constant 1 and the final correction coefficient to obtain the optimized artificial viscosity coefficient of the corresponding scale wave.

8. The SPH-based wave impact boundary force simulation analysis method according to claim 6 is characterized in that: The method for decomposing the coupled pressure signal of the interference sensitive area of ​​each wave into different component signals is a variational mode decomposition algorithm.

9. The SPH-based wave impact boundary force simulation analysis method according to claim 6, characterized in that: The coupled pressure signal of the interference sensitive area of ​​each wave is the average signal of the pressure signals of all test points in the interference sensitive area of ​​each wave under the impact of the corresponding wave.

10. The SPH-based wave impact boundary force simulation analysis method according to claim 6, characterized in that: The preset threshold is 0.3.

Images