A nonlinear multidirectional irregular wave and internal wave making system and a control method thereof

Abstract

The application discloses a nonlinear multidirectional irregular wave and internal wave wave-making system and a control method thereof, and relates to the technical field of hydrodynamic experimental research. The system comprises a wave-making module, a monitoring unit and a processing module. The wave-making module is of a non-uniform height layering structure, and the height of the layering structure can be self-adaptively adjusted according to a target frequency band. The monitoring unit is used for collecting flow velocity and wave height in real time. The processing module is internally provided with a mode management module, a wave construction module, a displacement conversion module, an instruction output module, a closed-loop correction module and a monitoring calculation module. The method comprises the following steps: generating a target vertical velocity profile, inversely kinematically solving, driving output, real-time monitoring, closed-loop selective correction, special monitoring and automatic mode switching. The application realizes multi-mode high-precision wave making of surface waves, internal solitary waves, quasi-steady internal waves, broken internal waves and mixed waves, and solves the problems of vertical velocity distribution distortion, full-frequency band correction interference and lack of pertinence in internal wave simulation.

Description

Technical Field

[0001] This invention relates to the technical field of hydrodynamic experimental research, specifically to a nonlinear multidirectional irregular wave and internal wave generation system and its control method. Background Technology

[0002] In fields such as marine engineering and marine environment simulation, accurate wave generation technology for nonlinear multidirectional irregular waves and internal waves is the core support for conducting related experimental research. Existing wave generation systems and methods have many technical defects in practical applications, resulting in low wave generation accuracy, poor waveform stability, and insufficient targeting of internal wave simulation, which cannot meet the requirements of high-precision experiments. The specific technical problems are as follows: Vertical wave-generating plates often employ a uniform layered design with fixed plate heights, making it impossible to adaptively adjust according to the center frequency of the preset target frequency band. This results in insufficient waveform simulation accuracy in the target frequency band and susceptibility to noise interference from non-target frequency bands. Furthermore, they often focus solely on velocity parameter control, failing to independently constrain wave heights in the near-field, far-field, surface, and bottom layers of the wave-generating area, and lacking clear wave height thresholds. This easily leads to wave height distortion and exceeding limits, failing to guarantee wave morphological stability. Closed-loop correction uses a uniform correction mode across the entire frequency band, neglecting to specifically correct velocity deviations within the target frequency band. This results in noise and harmonic interference from non-target frequency bands, reducing wave-generating accuracy. Internal wave generation technology lacks specificity, failing to provide dedicated control based on the characteristics of different types of internal waves, such as quasi-steady-state interface internal waves and broken internal waves. This makes it difficult to maintain the stability of the density gradient interface morphology of quasi-steady-state interface internal waves and to adapt to the high-frequency harmonic components generated by broken internal waves, leading to poor internal wave simulation results and morphological distortion.

[0003] In view of the shortcomings of the existing technology, there is an urgent need for a system and method that can solve the above technical problems and realize the precise generation of nonlinear multidirectional irregular waves and internal waves. Summary of the Invention

[0004] The purpose of this invention is to provide a nonlinear multidirectional irregular wave and internal wave generation system and its control method. The technical problems this invention aims to solve are: existing wave generation systems have fixed vertical plate heights and uniform layering, making adaptive adjustment based on the target frequency band impossible, resulting in insufficient waveform simulation accuracy in the target frequency band; they lack independent constraints on near-field, far-field, surface, and bottom-layer wave heights, easily leading to wave height distortion or exceeding limits; closed-loop correction uses a unified correction mode across the entire frequency band, resulting in severe noise and harmonic interference in non-target frequency bands; and internal wave generation technology lacks specificity, failing to provide dedicated control based on the characteristics of different types of internal waves, such as quasi-steady-state interface internal waves and broken internal waves. This invention achieves high-precision wave generation of nonlinear multidirectional irregular waves and internal waves through non-uniform adaptive layering, partitioned wave height constraints, selective closed-loop correction for the target frequency band, and multi-mode state machine switching.

[0005] The objective of this invention can be achieved through the following technical solutions: This invention provides a nonlinear multidirectional irregular wave and internal wave generation system, comprising: Wave-generating module: It is composed of multiple independent wave-generating plates that are hinged together in sequence along the water depth direction. The multiple independent wave-generating plates are arranged in layers with non-uniform heights along the water depth direction, and the height of the plates can be continuously or progressively adjusted according to the center frequency of the preset target frequency band. Each wave-generating plate is connected to an independent actuator. Monitoring unit: includes a water depth velocity measurement unit and a friction wave height sensor unit, which are used to collect the actual horizontal flow velocity at each water depth location and the actual wave height in the near field, far field, surface and bottom layers in real time; Processing module: The processing module is electrically connected to the actuator, the water depth velocity measurement unit, and the friction wave height sensor unit, respectively. The processing module includes: Mode Management Module: Built-in state machine for multiple working modes, used to switch working modes according to user instructions or real-time monitoring data, and load the corresponding preset parameter set; Wave construction module: used to calculate the theoretical horizontal velocity time history at each water depth position at the wave generator based on the selected nonlinear wave theory model, and generate the target vertical velocity profile; Displacement conversion module: used to construct the transfer function matrix between the motion velocity of each plate and the velocity of the fluid in front of the plate. The off-diagonal elements of the function matrix represent the hydrodynamic coupling effect between adjacent plates, and wave height constraint conditions are introduced as equality or inequality constraints. The displacement history of each plate is obtained by solving the constrained linear equation system. Command output module: Its input end is connected to the output end of the displacement conversion module, and its output end is connected to each of the actuators, used to convert the displacement time history into driving commands; Closed-loop correction module: used to compare the actual horizontal flow velocity with the theoretical horizontal velocity, and the actual wave height with the preset wave height threshold, and outputs correction signals only for deviations within the target frequency band; Monitoring and calculation module: includes interface displacement change rate calculation unit, flow velocity shear calculation unit and turbulent kinetic energy estimation unit, used to activate and calculate characteristic parameters in real time in the corresponding working mode, so that the mode management module can determine the switching conditions.

[0006] Preferably, the multiple operating modes of the mode management module include surface wave generation mode, internal solitary wave generation mode, quasi-steady-state interface internal wave generation mode, broken internal wave generation mode, and mixed wave generation mode; the preset parameter set includes nonlinear wave theory type, frequency band filter passband range, PID control coefficient, wave height constraint threshold, control period, and correction intensity level.

[0007] Preferably, the wave construction module selects the corresponding nonlinear wave theory based on the current working mode identifier provided by the mode management module: when the working mode is surface wave generation mode, higher-order Stokes theory is used; when the working mode is internal isolated wave generation mode, MCC theory is used; when the working mode is quasi-steady interface internal wave generation mode, eKdV theory is used; when the working mode is broken internal wave generation mode, mKdV theory is used; when the working mode is mixed wave generation mode, higher-order Stokes theory is used for the surface wave component, and MCC theory or eKdV theory is used for the internal wave component.

[0008] Preferably, the wave construction module is also used to perform frequency band truncation and reconstruction processing on the preset target frequency band range, retaining the target frequency band components; the frequency band truncation and reconstruction processing adopts a bandpass filter or wavelet transform method to limit the frequency components in the input target wave parameters to the preset target frequency band range, and performs waveform reconstruction on the retained target frequency band components based on the nonlinear wave theory.

[0009] Preferably, the displacement conversion module considers the hydrodynamic coupling effect between adjacent plates in the following way: establishing a transfer function matrix between the motion velocity of each plate and the velocity of the fluid in front of the plate, wherein the off-diagonal elements of the transfer function matrix characterize the mutual influence between adjacent plates, and obtaining the mapping relationship between the motion velocity of each plate and the target vertical velocity profile by solving a system of linear equations; in the wave height constraint condition, the preset independent wave height thresholds for the near field, far field, surface layer, and bottom layer are set independently, wherein the near field wave height threshold is 1.0 to 1.2 times the target wave height, and the far field wave height threshold is 0.9 to 1.1 times the target wave height.

[0010] Preferably, the closed-loop correction module includes a deviation calculation unit, a frequency band selection unit, and a correction amount generation unit; the deviation calculation unit is used to calculate a first deviation between the actual horizontal flow velocity and the theoretical horizontal velocity, and a second deviation between the actual wave height and a preset wave height threshold; the frequency band selection unit is used to extract the deviation component located in the target frequency band from the first deviation through a bandpass filter; the correction amount generation unit uses a PID control algorithm, a fuzzy PID control algorithm, or a model predictive control algorithm to generate a correction signal based on the deviation component in the target frequency band, and outputs it to the displacement conversion module or the command output module.

[0011] This application provides a method for generating nonlinear multidirectional irregular waves and internal waves, including the following steps: S1: Based on the input target wave parameters and the current working mode, call the corresponding nonlinear wave theory model to generate the target vertical velocity profile; S2: By establishing a transfer function matrix that considers the hydrodynamic coupling effect between adjacent plates and introducing wave height constraints, the constrained linear equations are solved to obtain the displacement history of each plate. S3: Convert the displacement time history into drive commands and output them to each actuator to drive the plate movement; S4: Real-time acquisition of actual horizontal flow velocity and actual wave height, and calculation of deviations from target values ​​and preset thresholds; S5: When the deviation exceeds a preset threshold, the deviation component in the target frequency band is extracted to generate a correction signal, and the output path of the correction signal is selected according to the degree of deviation. S6: Activate the monitoring and calculation module according to the current working mode, calculate the feature parameters in real time, and automatically switch the working mode and load the corresponding preset parameter set when the preset switching conditions are met. S7: Repeat steps S4 to S6 until the preset wave generation time is reached.

[0012] The beneficial effects of this invention are as follows: The system employs independent wave-generating plates arranged in non-uniform height layers, with the plate height adaptively adjusted according to the center frequency of the preset target frequency band. This allows the vertical velocity distribution at the wave-generating boundary to accurately match the nonlinear distribution of the actual wave, while effectively suppressing noise interference from non-target frequency bands and improving the waveform simulation accuracy of the target frequency band.

[0013] By introducing preset independent wave height thresholds for the near field, far field, surface, and bottom layers as constraints for inverse kinematics solutions, we ensure that the wave height distribution at each location in the wave-generating area meets the experimental requirements, avoid wave height distortion or exceeding limits, and ensure wave morphology stability.

[0014] By focusing on correcting velocity deviations within the target frequency band, deviations in non-target frequency bands are attenuated or ignored, thus avoiding interference from non-target frequency band noise and harmonics on the target waveform during full-band correction and improving wave generation accuracy.

[0015] Dedicated monitoring and control strategies are set up for quasi-steady-state interface internal waves and broken internal waves. For quasi-steady-state interface internal waves, the closed-loop correction intensity is automatically adjusted by monitoring the rate of change of interface displacement to maintain the stability of the density-skimming interface morphology. For broken internal waves, the closed-loop correction frequency band is automatically expanded by monitoring the flow velocity shear and turbulent kinetic energy to adapt to the high-frequency harmonic components generated by the breakage, ensuring the realism and morphological fidelity of the internal wave simulation. Attached Figure Description

[0016] To better understand and implement this application, the technical solution is described in detail below with reference to the accompanying drawings.

[0017] Figure 1 This is a schematic diagram of the nonlinear multidirectional irregular wave and internal wave generation system provided in Embodiment 1 of this application; Figure 2 This is a schematic diagram of the processing module architecture of the nonlinear multidirectional irregular wave and internal wave generation system according to Embodiment 1 of the present invention; Figure 3 This is a flowchart illustrating the nonlinear multidirectional irregular wave and internal wave generation method of Embodiment 1 of the present invention. Detailed Implementation

[0018] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, exemplary embodiments will be described in detail below, examples of which are illustrated in the accompanying drawings. In the following description, when referring to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application.

[0019] The terminology used in this application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. The singular forms “a,” “the,” and “the” used herein are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any and all possible combinations of one or more of the associated listed items.

[0020] The following detailed description of the specific implementation methods, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided in detail.

[0021] Example 1 Please see Figures 1-3 This embodiment provides a nonlinear multidirectional irregular wave and internal wave generation system and method for simulating nonlinear multidirectional irregular surface waves and specific types of internal waves such as internal isolated waves, quasi-steady-state interface internal waves, and broken internal waves in a water tank or flume. The nonlinear multidirectional irregular wave and internal wave generation system includes: a wave generation module, a monitoring unit, and a processing module; The wave-generating module is composed of eight independent wave-generating panels hinged sequentially along the water depth direction. These eight panels are arranged in non-uniform layers at different heights along the water depth direction: three panels are located near the water surface with heights of 5cm, 5cm, and 6cm; two panels are located near the density gradient layer, both with a height of 6cm; and three panels are located in other water depth areas with heights of 10cm, 12cm, and 15cm. The heights of the panels near the water surface and density gradient layer are smaller than those in other water depth areas.

[0022] The height of each sub-board can be determined according to the center frequency of the preset target frequency band. Continuous adaptive adjustment; the adjustment rules are as follows: Single mode (surface wave, internal solitary wave, quasi-steady interface internal wave, breaking internal wave): Scaling factor of the three partitions near the water surface Determined by the following piecewise linear function:

[0023] According to linear wave theory, the degree to which the amplitude of the horizontal velocity of a wave decreases with depth is determined by the wave number k, and the wave number and frequency satisfy a dispersion relation. (ω=2π) (hh is water depth); for high-frequency waves ( For waves ≥0.5Hz, the wave number is relatively large, and the velocity amplitude decreases exponentially with increasing water depth. The gradient is concentrated in the surface layer, therefore a smaller partition height is required to improve the vertical resolution of the surface layer; for low-frequency waves ( For frequencies ≤0.2 Hz, the wavenumber is relatively small, and the velocity profile tends to be uniform, allowing for a larger segmentation height. The scaling factors of 0.5 (high frequency) and 1.2 (low frequency) are selected based on the following: Under typical deep-water conditions (h≥2 m), the dimensionless wavenumber kh≈3.1 corresponds to a frequency of 0.5 Hz, with the velocity attenuating to approximately 0.9 times from the surface to 1 / 10 of the water depth; kh≈1.2 corresponds to a frequency of 0.2 Hz, with a slower velocity attenuation. 0.5 and 1.2 are the minimum and maximum values, respectively, that allow the segmentation height to cover the main velocity variation range, and these values ​​were determined through numerical simulations under typical operating conditions. Linear interpolation is used in the 0.2–0.5 Hz range to ensure that the scaling factor changes continuously with frequency, avoiding abrupt changes in vertical resolution.

[0024] Hybrid Wave Mode: Since the mixed wave contains both high-frequency surface waves (0.3-1.5Hz) and low-frequency internal waves (0.1-0.5Hz), it is impossible to simultaneously satisfy the optimal vertical resolution of both. This embodiment adopts a compromise configuration: the height of the three partitions near the water surface is taken as 0.8 times the default value (between halving the high-frequency component to 0.5 times and keeping the low-frequency component unchanged to 1.0 times), while the height of the remaining partitions remains at the default value; this 0.8 times factor is calibrated through numerical simulation of mixed wave generation, which can ensure that the high-frequency component of the surface wave is not distorted while taking into account the vertical coverage of the internal wave interface.

[0025] Each sub-board is connected to an independent servo motor actuator, which drives the corresponding sub-board to reciprocate independently in the horizontal direction, with a maximum stroke of ±0.3m.

[0026] The monitoring unit includes a water depth velocity measurement unit and a friction wave height sensor unit; The water depth velocity measurement unit is located 0.5m in front of the wave-making module. It consists of eight Doppler velocity sensors arranged at non-equidistant intervals along the water depth direction. The center depth of each sensor corresponds one-to-one with the center depth of the eight independent wave-making sub-plates, thus covering the entire water depth range from the surface to the bottom. This array is used to collect the actual horizontal velocity at each water depth location in real time, with a sampling frequency of 100Hz.

[0027] The wave height sensor unit consists of four capacitive wave height meters arranged along the wave propagation direction: two wave height meters are arranged at 1m (near field) and 5m (far field) from the wave generator, located at the surface (0.05m underwater) and the bottom (0.1m from the bottom of the pool), respectively; this array is used to collect the actual wave height at each location in the near field surface, near field bottom, far field surface, and far field bottom in real time, with a sampling frequency of 50Hz; The processing module uses an industrial PLC with an operating frequency of no less than 1GHz. It is electrically connected to each actuator, the water depth velocity measurement unit, and the friction wave height sensor unit to realize real-time signal reception and synchronous command output. The processing module includes a mode management module, a wave construction module, a displacement conversion module, a command output module, and a closed-loop correction module. The mode management module has a built-in state machine for five working modes, including surface wave generation mode, internal solitary wave generation mode, quasi-steady-state interface internal wave generation mode, broken internal wave generation mode, and mixed wave generation mode. The mode management module receives mode selection commands input by the user or automatically triggers mode switching based on real-time monitoring data. Each mode corresponds to a set of preset parameters, including: nonlinear wave theory type, frequency band filter passband range, PID control coefficient, wave height constraint threshold, control period, and correction intensity level. When switching modes, the processing module automatically loads the corresponding parameter set and smoothly adjusts the working state of each module.

[0028] Specifically, by using a built-in state machine with five working modes and a preset parameter set, and by automatically switching modes according to user commands or real-time monitoring data, it achieves intelligent adaptation and seamless switching between different wave types. The wave construction module receives target wave parameters and the current mode identifier at its input end, and calls the corresponding nonlinear wave theory calculation function according to the mode identifier. The parameters include wave height spectrum, period spectrum, direction spectrum, target frequency band center frequency, and specific internal wave type. The wave construction module first performs frequency band truncation and reconstruction processing on the preset target frequency band range: a configurable bandpass filter is used to filter out noise and spurious harmonic components in non-target frequency bands, retaining only the target frequency band components, and the waveform is reconstructed based on nonlinear wave theory for the retained target frequency band components. Then, the theoretical horizontal velocity time history at each water depth position at the wave generator is calculated and output as the target vertical velocity profile. Specifically, in the wave construction module, the nonlinear wave theory includes at least one of the higher-order Stokes theory, Boussinesq equation theory, or eKdV / MCC internal wave theory; the frequency band truncation and reconstruction processing uses a bandpass filter or wavelet transform method to limit the frequency components in the input target wave parameters to a preset target frequency band range, and reconstructs the waveform of the retained target frequency band components based on the nonlinear wave theory; The input of the displacement conversion module is connected to the output of the wave construction module and receives the wave height constraint threshold parameter of the current mode. The displacement conversion module establishes an 8×8 transfer function matrix T between the motion velocities of the eight plates and the fluid velocity in front of the plates. This matrix is ​​not a fixed constant but is dynamically adjusted according to the current wave frequency. Specifically, the off-diagonal elements of the matrix represent the hydrodynamic coupling effect between adjacent plates, and their values ​​are related to the wave frequency, the gap width between the plates, and the amplitude of the motion velocity. In this embodiment, a combination of offline CFD simulation and online interpolation is used to determine the matrix elements: for the frequency range of 0.1Hz to 1.5Hz, with a step size of 0.05Hz, the response of the fluid velocity in front of each plate under unit velocity excitation at different frequencies is calculated by CFD simulation to obtain a frequency-coupling coefficient lookup table. During actual operation, the processing module performs linear interpolation on the table according to the center frequency of the current target frequency band to obtain a transfer function matrix adapted to the current dynamic conditions in real time. For example, the adjacent coupling coefficient is about 0.25 at low frequencies, about 0.15 at mid frequencies, and about 0.08 at high frequencies. This scheme uses a frequency-adaptive matrix to accurately describe the dynamic coupling between the plates under different wave conditions, avoiding modeling errors caused by fixed constant matrices. Then, the displacement conversion module introduces independently configurable wave height constraints: the near-field wave height threshold is 1.1 times the target wave height, the far-field wave height threshold is 1.0 times the target wave height, the surface wave height threshold is 1.05 times the target wave height, and the bottom wave height threshold is 0.9 times the target wave height. By solving a system of linear equations, the motion velocity history of each plate is obtained, and then the displacement history of each plate is obtained by integration. The system of linear equations is constructed based on a pre-established transfer function matrix, which characterizes the mapping relationship between the motion velocity of each plate and the velocity of the fluid in front of the plate.

[0029] Specifically, the displacement conversion module considers the hydrodynamic coupling effect between adjacent plates in the following way: a transfer function matrix is ​​established between the motion velocity of each plate and the velocity of the fluid in front of the plate. The off-diagonal elements of the transfer function matrix represent the mutual influence between adjacent plates. The mapping relationship between the motion velocity of each plate and the target vertical velocity profile is obtained by solving a system of linear equations. In the wave height constraint condition, the preset independent wave height thresholds for the near field, far field, surface layer, and bottom layer are set independently. The near field wave height threshold is 1.0 to 1.2 times the target wave height, and the far field wave height threshold is 0.9 to 1.1 times the target wave height.

[0030] The input terminal of the instruction output module is connected to the output terminal of the displacement conversion module, and the output terminal is connected to each servo motor actuator. This module converts the displacement time history into an analog voltage signal (±10V), which is then amplified to drive the movement of each sub-board. This module supports dynamic adjustment of the control cycle and switches the execution frequency according to the instructions of the mode management module. The input of the closed-loop correction module is connected to the water depth velocity measurement unit and the friction wave height sensor unit, respectively. The output is selected to be connected to the displacement conversion module or the command output module according to the degree of deviation. This module is used to compare the actual horizontal velocity with the theoretical horizontal velocity and the actual wave height with the preset wave height threshold, and outputs a correction signal only for deviations within the target frequency band. Specifically, the closed-loop correction module includes a deviation calculation unit, a frequency band selection unit, and a correction amount generation unit. The deviation calculation unit calculates a first deviation between the actual horizontal flow velocity and the theoretical horizontal velocity, and a second deviation between the actual wave height and a preset wave height threshold for the corresponding region. The frequency band selection unit extracts the deviation component within the target frequency band from the first deviation using a bandpass filter with configurable parameters. The correction amount generation unit uses a PID control algorithm with online modifiable coefficients to generate a correction signal based on the deviation component within the target frequency band. When the amplitude of the correction signal corresponds to a deviation less than or equal to 1.5 times the preset threshold, the signal is output to the displacement conversion module for fine-tuning the displacement time history. When the deviation exceeds 1.5 times the preset threshold, the signal is directly output to the command output module and superimposed on the current drive command to achieve rapid response.

[0031] Furthermore, the processing module also integrates a monitoring and computing module, including: Interface displacement change rate calculation unit: Based on the velocity difference between two adjacent sensors above and below the density cascade in the water depth velocity measurement unit, the vertical displacement of the density cascade interface is inverted and its change rate is calculated; this unit is activated in the quasi-steady-state interface internal wave mode. Velocity shear calculation unit: Calculates the velocity difference between the upper and lower layers at the density jump layer by dividing the jump layer thickness to obtain the velocity shear value; this unit is activated in the broken internal wave mode. Turbulent kinetic energy estimation unit: Estimates turbulent kinetic energy in real time by performing high-pass filtering and root mean square calculation on the velocity fluctuation signal; This unit is activated in the broken internal wave mode; After each unit is activated, the corresponding characteristic parameters are calculated in real time and fed back to the mode management module. Specific trigger thresholds are set: when the interface displacement change rate is less than 0.01 m / s, the feedback signal triggers the mode management module to maintain the quasi-steady-state interface internal wave mode and synchronously adjust the closed-loop correction intensity; when the flow velocity shear value exceeds 0.5 s... -1 Or turbulent kinetic energy exceeds 0.01m 2 / s 2At that time, the feedback signal triggers the mode management module to adjust the parameters of the broken internal wave mode and expand the closed-loop correction frequency band range; After the system starts, the mode management module loads the corresponding mode parameter set according to the user input and distributes the parameters to each sub-module. Each sub-module operates according to the above functions, while the monitoring and calculation module calculates the characteristic parameters in real time and feeds them back to the mode management module. When the preset switching conditions are met, the mode management module automatically triggers mode switching and reconfigures the parameters of each sub-module to ensure the continuity and stability of wave generation control.

[0032] This embodiment provides a method for generating nonlinear multidirectional irregular waves and internal waves, including the following steps: Step S1: Parameter input, mode setting, and target vertical velocity profile generation The processing module receives target wave parameters input by the user, including: wave type (surface wave, internal isolated wave, quasi-steady interface internal wave, breaking internal wave, or mixed wave), and target wave height H. s Spectral peak period T p Target frequency band center frequency Frequency band range [f_L, f_H], water depth h, density stratification parameters (upper layer density ρ1, lower layer density ρ2, and stratification thickness δ are required for internal wave mode).

[0033] The mode management module automatically selects the corresponding working mode based on the wave type and loads the preset parameter set for that mode; the parameter configurations for the five working modes are as follows: Surface wave mode: Employs high-order Stokes theory, with the order automatically selected based on wave steepness (second order for steepness less than 0.1, third order for steepness between 0.1 and 0.2, and fifth order for steepness greater than or equal to 0.2). The filter passband range is 0.4Hz to 0.6Hz. The PID control coefficient is K. p =0.8, K i =0.05, K_d=0.2. The wave height constraint thresholds are: the surface wave height does not exceed 1.2 times the target wave height, and the bottom wave height does not exceed 1.0 times the target wave height. The control cycle is 0.02 seconds.

[0034] Internal solitary wave mode: MCC theory is employed. The filter passband range is 0.4Hz to 0.6Hz. The PID control coefficient is K. p =0.6, K i =0.03, K_d=0.15. The wave height constraint thresholds are: near-field wave height not exceeding 1.1 times the target wave height, far-field wave height not exceeding 1.0 times the target wave height, surface wave height not exceeding 1.05 times the target wave height, and bottom-layer wave height not exceeding 0.9 times the target wave height; the control period is 0.02 seconds.

[0035] Quasi-steady-state interface internal wave mode: Employing eKdV theory. The filter passband range is 0.3Hz to 0.5Hz. The PID control coefficient is K. p =0.4, K i =0.02, K_d=0.1. The wave height constraint threshold is the same as that of the inner solitary wave mode, but each threshold is relaxed by 10% (i.e., 1.21 times for the near field, 1.1 times for the far field, 1.155 times for the surface layer, and 0.99 times for the bottom layer); the control period is 0.02 seconds.

[0036] Broken internal wave mode: Employing mKdV theory. The filter passband range is 0.4Hz to 1.2Hz. The PID control coefficient is K. p =1.2、K i =0.05, K_d=0.4. Wave height constraint threshold is temporarily disabled (wave height exceeding the limit is allowed); control cycle is shortened to 0.01 seconds.

[0037] Hybrid wave mode: A combination of higher-order Stokes theory and MCC theory is used to calculate the surface wave and internal wave components separately and then linearly superimpose them. The filter employs a dual-channel design: the surface wave channel has a passband of 0.3Hz to 1.5Hz, and the internal wave channel has a passband of 0.1Hz to 0.5Hz. The PID control uses a dual-loop structure: the surface wave loop coefficient K... p =0.8, K i =0.05, K_d=0.2, internal loop coefficient K p =0.5, K i =0.02, K_d=0.1. Wave height constraints are set independently in layers: the surface wave height threshold is for surface waves (not exceeding 1.2 times the target wave height), and the bottom layer wave height threshold is for internal waves (not exceeding 0.9 times the target wave height); the control period is 0.02 seconds.

[0038] The processing module also considers the center frequency of the target frequency band. Adjust the panel height of the wave generation module. For non-hybrid wave modes, calculate the scaling factor using the piecewise linear function described above. Multiply the height of the three partitions near the water surface by For the mixed wave mode, we directly take α=0.8.

[0039] Then, the wave construction module calls the corresponding nonlinear wave theory according to the current mode identifier to truncate and reconstruct the preset target frequency band range; a configurable bandpass filter (or wavelet transform) is used to strictly limit the frequency components in the input target wave parameters to the target frequency band, and all frequency components outside the target frequency band (such as low-frequency swaying of the pool, high-frequency vibration of machinery, spurious harmonics, etc.) are set to zero or attenuated by at least 40dB; the target vertical velocity profile generated by this processing reduces the spurious components of non-target frequency bands, thereby avoiding the excitation of non-target frequency band noise by subsequent closed-loop control at the wave generation command level.

[0040] Taking the internal solitary wave mode as an example, the process is as follows: Frequency band cutoff: A zero-phase digital bandpass filter is used, with a passband frequency range of 0.4Hz to 0.6Hz and a stopband attenuation of -40dB. A Fast Fourier Transform (FFT) is performed on the input wavefront spectrum to zero out the frequency components outside the passband, followed by an Inverse Fourier Transform (IFFT) to reconstruct the time-domain wavefront sequence. A Hanning window is selected as the window function to suppress the Gibbs phenomenon. Waveform reconstruction: Based on MCC theory, given an upper water depth h1 = 1.0 m and a lower water depth h2 = 1.0 m, the reduced gravitational acceleration g′ = g(ρ2 - ρ1) / ρ2 = 0.196 m / s² 2 Solving the MCC equations yields the isolated waveform within the steady state:

[0041] Amplitude , wave number Calculated phase velocity ,in ; Velocity profile calculation: Horizontal velocity at various water depths at the wavemaker (x=0): Upper layer (-h1≤z≤0):

[0042] Lower layer (-h1-h2≤z<-h1):

[0043] in .

[0044] The center depth coordinates z of the 8 sub-plates i Substituting (j=1..8) into the equation, we obtain 8 velocity-time curves u. i (t), duration 60s, time step 0.01s, is used as the target vertical velocity profile output.

[0045] For other working modes, only the corresponding nonlinear wave theory and preset parameters need to be changed; the rest of the calculation process remains the same. Surface wave mode: adopts higher-order Stokes theory, and considers the direction spectrum discretized into 36 direction components, calculated separately and then superimposed; Quasi-steady-state interface internal wave mode: using eKdV theory, with the same parameters as above but different nonlinear term coefficients; Broken internal wave mode: using mKdV theory, the passband is extended to 0.4-1.2Hz, and higher-order nonlinear terms are considered during waveform reconstruction; Hybrid wave mode: Calculate the velocity components as surface waves and internal waves respectively, and then superimpose them linearly. Note that the total velocity does not exceed the actuator's capacity limit. Specifically, by adopting a non-uniform height partition and adaptively adjusting the partition height according to the center frequency of the target frequency band, the wave-generating board can have a higher vertical resolution within the target frequency band, effectively suppressing noise and harmonic interference in non-target frequency bands and improving the waveform reproduction of the target wave; through frequency band truncation and reconstruction processing, only the target frequency component is retained, avoiding waveform distortion and wave generation errors caused by false high-frequency components.

[0046] Step S2: Solving the inverse kinematics The displacement conversion module receives the target vertical velocity profile U_target(t) (an 8-dimensional vector) and the wave height constraint threshold parameters of the current mode.

[0047] Determining the transfer function matrix: Based on the wave frequency of the current operating mode, the current 8×8 transfer function matrix T is obtained by interpolation from a pre-stored frequency-coupling coefficient lookup table. This lookup table is built based on offline CFD simulation, covering a frequency range of 0.1Hz to 1.5Hz, and calculates the fluid velocity response in front of each plate under unit velocity excitation at each frequency point with a step size of 0.05Hz, thereby obtaining the frequency-related coupling coefficients. During actual operation, the processing module determines the current target frequency band center frequency based on the... Linear interpolation is performed on the lookup table to update the values ​​of each element in matrix T in real time.

[0048] Taking a typical operating condition as an example, the influence of frequency on matrix elements is illustrated: when At 0.2Hz (low frequency band), the coupling coefficient between adjacent boards is approximately 0.25; when At 0.6Hz (mid-frequency band), the adjacent coupling coefficient is approximately 0.15; when At 1.2Hz (high frequency band), the adjacent coupling coefficient is approximately 0.08. The diagonal elements also vary slightly with frequency, fluctuating between 0.92 and 0.98. Therefore, the transfer function matrix T is not a fixed constant, but is dynamically adjusted in real time according to the current wave generation frequency, thus accurately characterizing the hydrodynamic coupling effect between the plates under different wave conditions.

[0049] This matrix represents the speed of the plate movement. with the fluid velocity in front of the plate Linear mapping relationship between them: = · Through this frequency adaptive mechanism, this scheme avoids modeling errors caused by a fixed constant matrix under high-frequency or low-frequency conditions, and ensures the accuracy of inverse kinematics solutions across the entire frequency band. For strongly nonlinear wave conditions (such as internal solitary wave mode and broken internal wave mode), simply relying on the linear transfer function matrix T to solve the problem will lead to irreversible structural distortion of the vertical velocity profile. To resolve the fundamental contradiction between the linear approximation and the nonlinear target profile, this step introduces a nonlinear iterative compensation mechanism during the solution process: First, let's look at the system of linear equations. = · The analytical solution is used as the initial approximation. ^0. Subsequently, the processing module calls a pre-built nonlinear mapping database—which calibrates the high-order hydrodynamic coupling effects under different nonlinear characteristic parameters (such as dimensionless wave height H / h, wave steepness H / L, and density gradient thickness ratio) through offline CFD simulation—to obtain the nonlinear residual correction ΔU_NL(V_plate) corresponding to the current operating condition. The objective function considering nonlinear compensation is constructed as follows:

[0050] The Levenberg-Marquardt algorithm is used to iteratively optimize the objective function until the relative residual of the velocity profile is less than 5%. For highly nonlinear conditions such as solitary waves within the MCC, the vertical structural distortion caused by the linear matrix T can usually be eliminated after 3 to 5 iterations, achieving a precise closed loop from the nonlinear theoretical target to the physical plate displacement. Introducing wave height constraints: A wave height threshold vector H_max is set based on the current mode. Taking the internal isolated wave mode as an example:

[0051] These correspond to the near-field surface layer, near-field bottom layer, far-field surface layer, and far-field bottom layer, respectively. The relationship between wave height and the amplitude of the segmented plate motion, A, is H = S·A, where S is a 6×8 sensitivity matrix (6 measurement points, 8 segments); this matrix characterizes the wave height response generated by each segmented plate at different measurement points based on its unit motion amplitude; its elements are determined using a two-layer architecture of "offline CFD calibration as the primary reference and online RLS adaptive compensation as the secondary reference," as detailed below: Offline CFD calibration to obtain the initial reference matrix For the frequency range of 0.1Hz to 1.5Hz, CFD simulations of sinusoidal motion with unit displacement amplitude were performed on each of the eight sub-boards in a step size of 0.05Hz. The steady-state wave height amplitude at six wave height measurement points was recorded, and a frequency-dependent sensitivity reference matrix was constructed. (f) The CFD model setup is consistent with that described in Example 1. This reference matrix covers the entire frequency band and serves as the initial value for online operation, ensuring that the wave height prediction error is within ±5% under most normal operating conditions.

[0052] Online RLS adaptive compensation is only used to correct minor sensitivity drifts caused by slow time-varying factors such as water temperature stratification and pool wall reflection accumulation due to prolonged wave generation. In actual operation, the processing module uses the current plate displacement and measured wave height as inputs in each control cycle, and employs a recursive least squares algorithm with a forgetting factor to correct the sensitivity matrix online.

[0053] in Let y(k) be the current plate displacement vector, y(k) be the measured wave height vector, λ = 0.99 be the forgetting factor, and P be the value of P. The covariance matrix is ​​used. The RLS algorithm only fine-tunes the sensitivity coefficients near the current operating frequency. The correction range is strictly limited to ±10% of the reference value. Corrections exceeding this range will be truncated and recorded in the anomaly log to prevent incorrect corrections caused by measurement noise.

[0054] In inverse kinematics solving and wave height constraint calculation, the sensitivity matrix S actually used is taken as the reference matrix. With RLS compensation item sum: ; This dual-layer architecture ensures the physical accuracy of the sensitivity matrix across the entire frequency band (relying on CFD calibration reference) while also possessing the ability to adapt to slow environmental changes (relying on constrained online RLS correction). It avoids the respective shortcomings of pure offline calibration lacking adaptability and pure online identification lacking generalization ability. The two work together with a clear distinction between primary and secondary functions.

[0055] The wave height constraint is transformed into a linear inequality relating the amplitude of the plate movement: ,because (ω is the wave angular frequency), which can be further converted into a constraint on the velocity; Solve the constrained system of linear equations, where ≥ 0 are Lagrange multipliers and satisfy the complementary relaxation condition. The objective function is constructed using the Lagrange multiplier method:

[0056] Solve the KKT conditions to obtain the optimal velocity solution. If the inequality constraints cannot be satisfied simultaneously, a softening approach is adopted: the objective function is modified as follows:

[0057] Among the penalty factors =100, where Indicates taking The larger of the values ​​in parentheses corresponds to a penalty term of zero when the wave height constraint is satisfied. First, attempt a strict constraint solution; if convergence fails after 10 iterations, switch to a softening process to ensure uninterrupted wave generation. Under softening, the wave height overshoot is typically less than 5% and the duration does not exceed two wave cycles. Finally, trapezoidal integration was used to obtain the displacement history of each plate, and it was checked whether it exceeded the maximum travel ±0.3m.

[0058] The displacement time history is obtained by integration: trapezoidal numerical integration is used.

[0059] Initial displacement Check if the maximum displacement of each plate exceeds ±0.3m. If it does, scale the velocity sequence proportionally. Specifically, by introducing a transfer function matrix to consider the hydrodynamic coupling between adjacent plates, the motion of the plates is made to better match the actual velocity distribution of the flow field, avoiding the problems of discontinuous flow velocity and vertical profile distortion caused by independent motion of the plates; by introducing independent wave height constraints for each zone, the wave heights in the near field, far field, surface, and bottom layers can be strictly controlled to ensure that the wave morphology is stable and the data is reliable.

[0060] Step S3: Output driver instructions The instruction output module converts the displacement time history into an analog voltage signal: The amplitude is limited to ±10V; this signal is amplified by a power amplifier to drive the servo motor actuator; the processing module simultaneously reads the actual displacement fed back by the encoder, forming an internal position closed loop of the actuator to ensure that the tracking error is less than 0.1mm. The control cycle is set according to the instructions of the mode management module: 0.02s (50Hz) for normal mode and 0.01s (100Hz) for broken internal wave mode.

[0061] Specifically, by accurately converting displacement time history into drive commands, the actuator's movement is ensured to be smooth and shock-free, improving wave generation repeatability and stability; high-frequency control cycles are adopted to enhance the system's ability to track high dynamic waveforms such as nonlinear waves and broken internal waves.

[0062] Step S4: Real-time monitoring and deviation calculation After wave generation is started, the monitoring unit continuously collects data: Water depth velocity measurement unit: 8 sensors sample at 100Hz to obtain the actual horizontal velocity vector. .

[0063] Along-field wave height sensor unit: 4 sensors (near-field surface, near-field subsurface, far-field surface, and far-field subsurface) sample at 50Hz to obtain the actual wave height vector. .

[0064] In each control cycle, the deviation calculation unit of the closed-loop correction module performs the following: First deviation: ,in Indexed by time from the array stored in step S1 Read ( (This refers to the wave cycle).

[0065] Second deviation: ,in This is the wave height threshold vector for the current mode.

[0066] Specifically, by collecting flow velocity and wave height data in real time and calculating the double deviation, accurate feedback is provided for closed-loop control, enabling the system to quickly identify waveform deviations and correct them in a timely manner. Step S5: Closed-loop selective correction During wave generation, the closed-loop correction module monitors the actual horizontal flow velocity in real time. Compared with theoretical speed The first deviation between - and actual wave height With preset wave height threshold The second deviation between = - When the infinity norm of the first deviation exceeds 0.02 m / s or the infinity norm of the second deviation exceeds 0.01 m, the closed-loop correction process is initiated. The frequency band selection unit first applies a digital bandpass filter (IIR second-order Butterworth filter, the passband range is determined by the current operating mode, for example, 0.4Hz~0.6Hz for the internal solitary wave mode, with a stopband attenuation of -40dB) to the first deviation signals at the eight water depth positions, and extracts the velocity deviation components located in the target frequency band. The deviation components in non-target frequency bands are attenuated by at least 20dB to ensure that the correction effect is focused only on the target wave components of actual interest, and to avoid the amplification of low-frequency swaying in the pool or high-frequency mechanical vibrations in the closed loop.

[0067] The correction generation unit uses a PID control algorithm to calculate the speed correction. :

[0068] in , , The PID control coefficients are preset for the current operating mode (e.g., 0.6, 0.03, and 0.15 for the internal solitary wave mode). The integral term uses a limited accumulation to prevent integral saturation, and the derivative term is implemented through backward differential. This correction is used to compensate for the dynamic error between the plate motion and the target vertical velocity profile.

[0069] Select the output path of the correction signal based on the degree of deviation: When the infinity norm of the first deviation is ≤0.03m / s and the infinity norm of the second deviation is ≤0.015m (i.e., the deviation does not exceed 1.5 times the preset threshold), the speed correction amount will be... The output is sent to the displacement conversion module and superimposed on the input of the current plate velocity solution process. Step S2 is then re-executed to smoothly adjust the displacement history of each plate. This path has a small correction amplitude and a stable response, making it suitable for fine calibration when the wave pattern is basically stable. Specifically, by correcting the deviation within the target frequency band, the noise amplification and waveform jitter problems caused by full-band correction are avoided, significantly improving wave generation accuracy; a dual path of smooth correction and fast compensation is adopted to balance control stability and response speed, with smooth correction of small deviations and rapid suppression of large deviations.

[0070] Step S6: Dedicated monitoring and automatic mode switching This step activates the corresponding monitoring and calculation module based on the current working mode, thus achieving automatic mode switching.

[0071] Scenario 1: Monitoring and Control in Quasi-steady-state Interface Internal Wave Mode When the current mode is the quasi-steady-state interface internal wave mode, the interface displacement change rate calculation unit is activated: Based on the flow velocity difference between adjacent flow velocity sensors (corresponding to the depths of the 4th and 5th partition plates) on the upper and lower levels of the mezzanine. Combined with the thickness of the interlayer Inversion of interface displacement .

[0072] Calculate the rate of change (Sliding window average, window length 1s).

[0073] when After 10 seconds, a feedback signal is sent to the mode management module: maintain the quasi-steady-state mode, and simultaneously adjust the PID coefficients of the closed-loop correction module to... , , The deviation threshold is relaxed to This reduces the intensity of correction and maintains the stability of the interface morphology.

[0074] Scenario 2: Switching from internal solitary wave / quasi-steady-state mode to broken internal wave mode When the current mode is an internal solitary wave mode or a quasi-steady-state mode, activate the velocity-shear calculation unit and the turbulent kinetic energy estimation unit (to predict breakup conditions): Flow rate shear ,in , The velocity data were taken from the flow velocity sensors located above and below the mezzanine.

[0075] Turbulent kinetic energy By high-pass filtering (cutoff frequency) of the flow velocity pulsation signal The root mean square is then calculated.

[0076] when or At that time, the feedback signal trigger mode management module immediately switches to the broken internal wave mode: The passband of the frequency band selection unit is extended to 0.4-1.2Hz; PID coefficients adjusted to , , ; The control cycle has been shortened to 0.01s; Temporarily disable wave height constraint threshold (allow wave height to exceed limit).

[0077] The switching process is completed within one control cycle without interrupting wave generation.

[0078] Scenario 3: Recovery from broken internal wave mode to internal solitary wave mode After the crushing is completed, that is and After 20 seconds, the mode management module automatically reverts to the internal solitary wave mode, restoring the original parameters (passband 0.4-0.6Hz, PID coefficient). , , (Control cycle 0.02s, wave height constraint re-enabled).

[0079] Other mode switching instructions: Surface wave mode and internal isolated wave mode do not switch automatically; user must select manually. Hybrid wave mode is an independent mode and does not automatically switch with other modes. All mode switching can be manually triggered by the user. When manually switching, automatic monitoring is temporarily disabled for 30 seconds, after which monitoring automatically resumes. Specifically, the system uses a dedicated monitoring unit to identify the internal wave state in real time, thereby maintaining the quasi-steady-state internal wave interface, automatically identifying broken internal waves and expanding the frequency band. This enables the system to adapt to different internal wave evolution stages without the need for manual intervention to switch modes.

[0080] Step S7: Loop execution and wave generation termination The above steps S1–S6 are executed cyclically in the current control cycle; after each control cycle is completed, the processing module determines whether the preset wave generation duration has been reached: If it is not achieved, continue to the next cycle; If the target has been reached, the termination process is executed: the instruction output module outputs zero voltage, each board slowly resets to zero at a speed of 0.1m / s, the monitoring channel is shut down, all real-time recorded data (including 8-channel flow rate, 4-channel wave height, 8-channel board displacement, 8-channel correction signal, etc.) are saved to the solid-state drive, and a log file is generated.

[0081] This embodiment achieves high-precision multi-mode wave generation of surface waves, internal isolated waves, quasi-steady-state interface internal waves, broken internal waves, and mixed waves through a non-uniform adaptive wave generation plate, target frequency band selective control, partitioned wave height constraints, dual-loop closed-loop correction, and dedicated internal wave monitoring and switching strategies. By forming a dual synergy between frequency band truncation and reconstruction and closed-loop frequency band selective correction, the noise of non-target frequency bands is attenuated step by step in the entire control link, avoiding the problem of noise being amplified by the closed loop in traditional full-band correction. When the wave generation is stable, the energy of the wave height spectral density of non-target frequency bands is reduced by about an order of magnitude.

[0082] Example 2 This embodiment provides a nonlinear multidirectional irregular wave and internal wave generation system and method, specifically involving the collaborative wave generation of surface waves and internal waves under a mixed wave mode. Based on the same system hardware structure as Embodiment 1, this embodiment designs specialized parameter configuration and control strategies for the mixed wave mode, which is used to simultaneously simulate the coexistence wave field of nonlinear multidirectional irregular surface waves and internal isolated waves in a pool or tank. This solves the problem that single wave mode simulation cannot reproduce the coupling phenomenon of surface waves and internal waves in the actual ocean, and improves the realism and applicability of the wave generation scenario.

[0083] The system hardware in this embodiment is basically the same as that in embodiment 1, including a wave generation module, a monitoring unit, and a processing module. The difference lies in the mode management module, wave construction module, displacement conversion module, and closed-loop correction module inside the processing module. The following special configurations are made for the mixed wave mode to ensure that the surface wave and internal wave are controlled in a coordinated manner and do not interfere with each other.

[0084] When the user selects the hybrid waveform mode, the mode management module automatically loads the following preset parameter set, and also supports the user to fine-tune some parameters according to actual needs: Nonlinear wave theory type: adopts a dual-theory superposition mode, with the surface wave component using the fifth-order Stokes theory; the internal wave component can be specified by the user, selecting either the MCC theory or the eKdV theory to adapt to different internal wave scenario requirements. Filter passband: Dual-channel independent filtering is used to avoid mutual interference between the two waveform components. The surface wave channel passband is set to 0.3Hz to 1.5Hz, covering the conventional frequency band of multi-directional irregular surface waves; the inner wave channel passband is set to 0.1Hz to 0.5Hz, matching the low-frequency characteristics of internal solitary waves and quasi-steady-state interface inner waves. PID control coefficients: Dual-loop independent PID control is adopted to adapt to the control requirements of two waveforms. The surface wave loop coefficients are set to... , , Emphasis is placed on high-frequency response accuracy to ensure the stability of the surface wave waveform; the internal wave loop coefficient is set to... , , It focuses on maintaining the low-frequency shape and avoiding distortion of the inner wave interface; Wave height constraint thresholds: set independently in layers to match the distribution characteristics of the two wave types. The surface wave height threshold is for surface waves and does not exceed 1.2 times the target wave height of the surface wave; the bottom wave height threshold is for internal waves and does not exceed 0.9 times the target wave height of the internal wave; the wave height thresholds for the intermediate water depths in the near and far fields are obtained by linear interpolation of the surface and bottom thresholds to ensure the continuity of wave height constraints across the entire water depth. Control cycle: uniformly set to 0.02 seconds, taking into account the requirements of high-frequency response of surface waves and low-frequency control of internal waves, and ensuring the coordinated synchronization of the two loops.

[0085] Correction intensity level: The dual-loop independent correction and superimposed output mode is adopted. The surface wave loop and the inner wave loop can independently adjust the correction intensity according to their respective deviations without affecting each other. At the same time, the superimposed correction signal can be adapted to the upper limit of the actuator capability. Wave construction module: Simultaneously and independently calculates the velocity components of surface waves and internal waves, and then obtains the total theoretical horizontal velocity time history at each water depth in front of the wave generator through linear superposition. This ensures that the superposition of the two wave components does not produce waveform distortion and satisfies the actuator motion constraints. Specific configuration is as follows: Independent calculation: The velocity calculation modules corresponding to surface waves and internal waves are called separately to generate the velocity history of the two wave types independently, avoiding mutual interference during the calculation process.

[0086] Superposition rules: A linear superposition method is adopted. The total velocity is the algebraic sum of the surface wave velocity and the internal wave velocity. Before superposition, it is necessary to check whether the total velocity amplitude exceeds the upper limit of the actuator's capability. If it exceeds the limit, it is scaled proportionally. Among them, the accuracy of the internal wave is given priority. The scaling factor of the surface wave is 0.8 to 1.0, and the scaling factor of the internal wave is not less than 0.9.

[0087] Output specifications: Generate total velocity time-history curves corresponding to the center depths of 8 sub-plates, with a duration of 60s and a time step of 0.01s, to ensure compatibility with subsequent displacement conversion modules; Displacement conversion module: Receives the total target velocity profile output by the wave construction module. And the wave height constraint threshold parameter for the hybrid wave mode. The transfer function matrix T follows the frequency adaptation mechanism of Example 1: based on the current superposition frequency characteristics of surface waves and internal waves, the current 8×8 matrix is ​​obtained by interpolation from a pre-stored frequency-coupling coefficient lookup table, rather than a fixed constant matrix. The off-diagonal elements of the matrix vary with frequency—the adjacent coupling coefficient is approximately 0.25 in the low-frequency band, approximately 0.15 in the mid-frequency band, and approximately 0.08 in the high-frequency band—thus accurately characterizing the dynamic hydrodynamic coupling effect between the plates in the hybrid wave field.

[0088] Wave height constraint adaptation: Constraints are strictly executed according to the hierarchical thresholds set by the mode management module. The surface constraint is for the target wave height of the surface wave, the bottom constraint is for the target wave height of the inner wave, and the intermediate water depth constraint is obtained through linear interpolation. During the solution process, the wave height limits of the two wave types are satisfied at the same time to avoid wave generation failure caused by a single wave type exceeding the threshold.

[0089] Closed-loop correction module: Configured as a dual-loop structure, the surface wave loop and the inner wave loop operate independently and output collaboratively. The core is to design different correction strategies based on the characteristic differences of the two waveforms. The specific configuration is as follows: Dual-loop independent design: The surface wave loop focuses on high-frequency deviation correction to adapt to the high-frequency characteristics of surface waves; the inner wave loop focuses on low-frequency deviation correction to ensure the stability of the inner wave interface morphology. The correction parameters and correction paths of the two loops can be adjusted independently without interference.

[0090] Deviation calculation: Calculate the velocity deviation and wave height deviation of surface waves and internal waves respectively, and independently determine whether the correction process needs to be initiated.

[0091] Correction signal superposition: The correction signals generated by the two loops are linearly superimposed and output. After superposition, it is necessary to check whether the signal amplitude exceeds the upper limit of the actuator. If it exceeds the limit, it is rescaled according to the weight allocation, with the surface wave weight being 0.4 and the inner wave weight being 0.6, giving priority to ensuring the correction effect of the inner wave.

[0092] This embodiment provides a nonlinear multidirectional irregular wave and internal wave generation method in a hybrid wave mode. Based on the above system configuration, it realizes the coordinated generation of surface waves and internal waves, specifically including the following steps: Step S1: Input of hybrid wave parameters and generation of dual-theoretical target velocity profiles The processing module receives the mixed wave parameters input by the user, determines the specific parameters of the surface wave and the internal wave, and completes mode loading, plate height adjustment, and dual-wavelength velocity component calculation and superposition. The specific process is as follows: Parameter Input: The processing module receives user-input hybrid wave-related parameters, which are clearly categorized and unambiguous, specifically including: Surface wave parameters: target wave height Spectral peak period Spectral parameters (main wave direction) directional distribution coefficient ), target frequency band center frequency ; Internal wave parameters: target wave height ,cycle Density stratification parameters, upper layer density Lower layer density Thickness of the layer Internal wave type; General parameters: Total water depth And the relative phase of surface waves and internal waves.

[0093] Mode Loading and Panel Height Adjustment: The mode management module loads the preset parameter set of the hybrid wave mode. Since the hybrid wave includes a wide frequency band (0.3-1.5Hz for high-frequency surface waves and 0.1-0.5Hz for low-frequency internal waves), it is necessary to balance the vertical resolution of high-frequency surface waves and the coverage of low-frequency internal waves. Unlike the step-by-step adjustment based on the center frequency of a single frequency band in Example 1, a compromise configuration is adopted in the hybrid wave mode: the height of the three panels near the water surface is taken as 0.8 times the default value (between 0.5 times for high-frequency waves and 1.0 times for low-frequency waves), while the height of the remaining panels remains at the default value.

[0094] Dual-wave velocity component calculation: The wave construction module calculates the velocity components of surface waves and internal waves separately, ensuring calculation accuracy and waveform integrity. Surface wave velocity component calculation: Fifth-order Stokes theory is used, considering directional spectrum characteristics, to simulate multi-directional irregular surface waves. First, the directional spectrum is discretized into 36 directional components (uniformly distributed, each direction spaced 10° apart). The frequency range of each directional component (0.3Hz~1.5Hz) is discretized as follows: Each frequency point ensures that the frequency resolution meets the wave generation accuracy requirements. For each frequency-direction component... Calculate the wave-generating plate ( , ) water depth The horizontal velocity at that point is given by the following formula: in, This is the JONSWAP spectrum. for Type directional distribution function, Satisfy dispersion relation ( (Acceleration due to gravity) For random phase (range 0 ~ ), For frequency intervals, For directional spacing. By taking the center depth of the 8 sub-plates and substituting it into the formula, the surface wave velocity time history can be obtained. ( ).

[0095] Internal wave velocity component calculation: The calculation is performed using the corresponding nonlinear theory based on the user-specified internal wave type. If an internal solitary wave is selected, the MCC theory is used; if a quasi-steady-state interface internal wave is selected, the eKdV theory is used. The specific calculation process follows the internal wave velocity calculation method in step S1 of Example 1, yielding the internal wave velocity time history. ( It should be noted that the propagation direction of the internal wave is the same as that of the main surface wave by default. Users can set it to the opposite direction according to their needs. In this case, the phase parameter needs to be adjusted in the velocity calculation.

[0096] Velocity component superposition: The total target velocity profile is a linear superposition of surface wave velocity and internal wave velocity, as shown in the following formula: After superposition, it is necessary to check whether the total velocity amplitude exceeds the actuator's maximum capability (in this embodiment, the actuator's maximum speed is ±0.6m / s, corresponding to a displacement of ±0.3m and a frequency of 2Hz). If it exceeds the upper limit, the surface wave and internal wave components are scaled proportionally. The scaling principle is: prioritize the accuracy of the internal wave, with the surface wave scaling factor set at 0.8 to 1.0 and the internal wave scaling factor set at no less than 0.9, to ensure that the internal wave interface morphology is not significantly affected.

[0097] Output target velocity profile: Output 8 total velocity time-history curves The duration is 60 seconds, the time step is 0.01 seconds, and the data is sent to the displacement conversion module. Specifically, a hybrid wave velocity profile is generated by independent calculation using dual theories and linear superposition to ensure that the surface wave and internal wave components do not interfere with each other, and the waveform superposition is natural and distortion-free; the velocity components are scaled according to priority to ensure the stability of the internal wave interface morphology while taking into account the accuracy of the surface wave.

[0098] Step S2: Solving the inverse kinematics The displacement conversion module receives the total target velocity profile. The wave height constraint threshold parameter for the mixed wave mode is also considered. Since the mixed wave includes strongly nonlinear surface waves (fifth-order Stokes) and strongly nonlinear internal waves (MCC or eKdV), the linear transfer function matrix T cannot accurately characterize the mapping relationship between the plate motion and the nonlinear flow field. To resolve this fundamental contradiction, this step employs the same nonlinear iterative compensation mechanism as in Example 1. First, based on the superposition frequency characteristics of the current mixed waves, the linear transfer function matrix T is obtained by interpolation from the frequency-coupling coefficient lookup table. Then, the linear equations are solved to obtain an approximate solution for the initial plate separation velocity. Then, the nonlinear mapping database is invoked to obtain the current nonlinear characteristic parameters (surface wave steepness). / Relative wave height of internal waves Nonlinear residual correction amount corresponding to / h, density gradient thickness ratio δ / h) Construct an objective function that comprehensively considers velocity fitting error and layered wave height constraints:

[0099] Among them, wave height constraint threshold A layered independent thresholding method was adopted: the threshold for the surface layer (near-field and far-field surfaces) was 1.2 times the target surface wave height, and the threshold for the bottom layer (near-field and far-field bottom layers) was 0.9 times the target internal wave height. The Levenberg-Marquardt algorithm was used iteratively until the relative residual of the velocity profile was less than 5%. After convergence, the displacement time history was obtained by trapezoidal integration of the velocity time history of each plate, and it was verified whether it exceeded the actuator's maximum stroke ±0.3m. This nonlinear iterative solution process ensures that the vertical velocity structures of both surface waves and internal waves in the mixed wave field can accurately reproduce the nonlinear theoretical predictions, achieving a complete closed loop from theoretical generation to physical realization.

[0100] Constructing the transfer function matrix The matrix structure is consistent with that of Example 1, ensuring that the motion coupling relationship between the plates remains unchanged.

[0101] Define the threshold for layered wave height constraints: surface layer ( ) wave height threshold (For surface wave target wave height), corresponding to near-field and far-field surface measurement points; bottom layer ( ) wave height threshold (For internal wave target wave height), corresponding to near-field and far-field bottom measurement points; the wave height threshold at intermediate water depths (such as near a strata divide) is obtained by linear interpolation of the surface and bottom thresholds, using the following formula: To ensure the continuity of the full-depth wave height constraint, the intermediate depth constraint is only used as an additional constraint when solving the inverse kinematics, and no corresponding wave height sensor is set up for real-time monitoring.

[0102] Solving the constrained linear equation system using the Lagrange multiplier method, with the objective function being: ( (Based on the measured wave height prediction values ​​from four wave height sensors), the motion velocity time history of each segment is obtained by solving the equation, and the displacement time history is obtained by integration. Send to the instruction output module; Specifically, by using layered independent wave height constraints, surface waves and internal waves can satisfy their respective limits, thus preventing one wave type from exceeding the limit in the mixed wave field and affecting the other wave type, thereby improving the stability of the mixed wave simulation.

[0103] Step S3: Output driver instructions The instruction output module will output the plate displacement time history obtained by the displacement conversion module. This is converted into the corresponding voltage control signal, and the conversion formula is as follows: The output is sent to the drive unit of the wave generation module, which drives the 8 sub-boards to move according to the preset displacement time history, thus initially realizing the hybrid wave generation; Specifically, the dual-path drive output ensures the simultaneous and accurate output of the high-frequency components of the surface wave and the low-frequency components of the internal wave, thus achieving a true reproduction of the wave field where surface waves and internal waves coexist.

[0104] Step S4: Real-time monitoring and dual-loop deviation calculation The monitoring unit collects wave field data from the water tank in real time. The deviation calculation unit of the closed-loop correction module calculates the deviation for both surface waves and internal waves, providing a basis for subsequent dual-loop correction. The specific process is as follows: Data Acquisition: The monitoring unit uses eight flow velocity sensors to collect the actual flow velocity at each water depth. ( Four wave height sensors respectively collected the actual wave heights of the near-field surface, near-field subsurface, far-field surface, and far-field subsurface. ( The acquisition frequency is consistent with the control cycle (0.02 seconds) to ensure data real-time performance.

[0105] Surface wave deviation calculation: The actual flow velocity is used as the basis for this calculation. Surface wave component in target velocity By comparing time-by-time data, the surface wave velocity deviation was obtained. Simultaneously, the measured surface wave height values ​​(near-field surface and far-field surface) are compared with the surface wave height threshold. By comparison, the surface wave height deviation was obtained. .

[0106] Internal wave deviation calculation: The actual flow velocity is used as the basis for this calculation. With the internal wave component in the target velocity By comparing time-by-time data, the internal wave velocity deviation was obtained. Simultaneously, the measured values ​​of bottom wave height (near-field bottom layer and far-field bottom layer) are compared with the bottom wave height threshold. By comparison, the internal wave height deviation was obtained. ,in 0.9× For example, when the internal wave target height is 0.15m. 0.135m.

[0107] Deviation preprocessing: The flow velocity deviation between the two loops is denoised (using sliding window averaging with a window length of 0.05 seconds) to remove abnormal deviation values ​​and ensure the stability of the correction signal; Specifically, the deviation is calculated independently through dual loops, and the surface wave and internal wave deviations do not interfere with each other, and are accurately fed back separately.

[0108] Step S5: Dual-loop closed-loop selective correction It includes two independent correction loops to handle surface wave deviation and internal wave deviation respectively. Finally, the correction signals from the two loops are superimposed and output to achieve precise control of the hybrid wave. The specific process is as follows: Surface wave loop correction: Frequency band selection: for the pre-processed surface wave velocity deviation Apply a bandpass filter (passband 0.3–1.5 Hz) to extract the deviation component within the surface wave frequency band. This eliminates low-frequency interference and high-frequency noise.

[0109] Correction quantity generation: A PID control algorithm is used, and preset coefficients are substituted. , , Calculate the surface wave correction velocity The formula is: The integral term uses a limited summation (limited value ±0.1) to avoid integral saturation; the differential term uses backward difference, as shown in the formula: ( (seconds), improving high-frequency response speed.

[0110] Output path selection: Determine if the deviation is within the allowable range. (1.5 times the threshold) and Then Output to the inverse kinematics solver module to smooth the plate displacement path; otherwise, It is directly output to the drive command output module to achieve rapid correction and avoid surface wave waveform distortion.

[0111] Internal wave loop correction: Frequency band selection: for the pre-processed internal wave velocity deviation Apply a bandpass filter to extract the deviation component within the inner frequency band. Eliminate high-frequency interference.

[0112] Correction quantity generation: A PID control algorithm is used, and preset coefficients are substituted. , , Calculate the internal wave correction velocity The formula is: The integral and differential terms are handled in the same way as the surface wave loop to ensure a stable correction signal and avoid drastic fluctuations at the internal wave interface.

[0113] Output path selection: Determine if the deviation is within the allowable range. (1.5 times the threshold) and Then Output to the inverse kinematics solution module; otherwise, output directly to the drive command output module to quickly correct the internal wave deviation.

[0114] Correction of signal superposition and amplitude limiting: The total correction signal is a linear superposition of the correction signals from the two loops, as shown in the formula: After superposition, it is necessary to check whether the total correction signal amplitude exceeds the actuator's upper limit (±0.6m / s). If it does, the signal should be rescaled according to the weight allocation: surface wave correction signal weight 0.4, internal wave correction signal weight 0.6. Since the internal wave interface morphology has a greater impact on wave generation accuracy and the internal wave correction is more difficult than the surface wave correction, the internal wave weight is higher to prioritize the internal wave correction effect; ensuring the stability of the internal wave interface morphology while also considering the accuracy of the surface wave; after scaling, Output to the corresponding module to complete one correction; Specifically, by using dual-loop independent PID correction, the surface wave focuses on high-frequency tracking while the internal wave focuses on low-frequency shape maintenance, thus solving the problem of simultaneous high-precision control of mixed wave fields; the correction amount is superimposed according to weights to prioritize the stability of the internal wave, and the overall wave generation effect is closer to the real marine environment.

[0115] Step S6: Dedicated Monitoring and Mode Switching In hybrid wave mode, the processing module adds a monitoring and calculation module to simultaneously monitor the characteristic parameters of surface waves and internal waves. Based on the monitoring results, it triggers corresponding parameter adjustments or mode switching to ensure the stability of the hybrid wave field, as detailed below: Surface wave dedicated monitoring: Real-time calculation of the effective wave height of surface waves using surface measurement points along the path of the wave height sensor unit. Spectral peak period Compare this with the target parameters input by the user. If and If the surface wave reaches the target accuracy, the correction strength of the surface wave loop is appropriately reduced (by setting the correction strength to the target accuracy). Adjust to 0.6); if the surface wave deviates from this range again (i.e., any error exceeds the limit), the original correction intensity will be automatically restored (the original correction intensity will be adjusted to 0.6). Adjusted to 0.8).

[0116] Internal wave dedicated monitoring: Activate the corresponding monitoring unit according to the internal wave type specified by the user; If the internal wave type is quasi-steady-state interface internal wave: activate the interface displacement change rate calculation unit, based on the velocity difference between adjacent velocity sensors above and below the jump layer. Combined with the thickness of the interlayer Inversion of interface displacement Calculate the rate of change .when After 10 seconds, the internal wave is determined to have entered a quasi-steady state, and the PID coefficients of the internal wave loop are adjusted to... , , This reduces the intensity of the correction and maintains the stability of the interface morphology.

[0117] If the internal wave type is an internal solitary wave: activate the velocity shear and turbulent kinetic energy estimation units, and calculate the velocity shear. and turbulent kinetic energy .when or When the internal wave is about to break, the internal wave circuit parameters are adjusted: the passband is extended to 0.4–1.2 Hz, and the PID coefficient is adjusted to... , , Simultaneously disable internal wave height constraint; when the crushing ends ( and (Continue for 20 seconds) to restore the original parameters of the inner wave loop.

[0118] Special Mode Switching: If users need to simulate the coexistence of surface waves and internal breaking waves for extended periods, they can manually force the dual-loop strategy in the internal breaking wave mode. This operation overrides the automatic switching logic in scenario two, ensuring the system does not automatically recover even after the breaking end conditions are met. In this mode, the internal wave loop passband is fixed at 1.2Hz, while the surface wave loop maintains its original parameters until manually switched back by the user. This mode is suitable for special experiments requiring continuous observation of the interaction between internal breaking waves and surface waves. Specifically, through dedicated monitoring and dynamic parameter adjustment of hybrid wave fields, the system can operate stably under various combined conditions of surface waves + internal isolated waves / quasi-steady state / broken internal waves, making it applicable to a wider range of scenarios.

[0119] Step S7: Loop execution and wave generation termination With a control cycle of 0.02 seconds, steps S1 to S6 are executed cyclically, data is collected in real time, deviations are calculated, and corrections are performed until the preset wave generation time is reached. The processing module then automatically terminates wave generation, shuts down the drive unit and the monitoring unit, and completes the mixed wave generation process.

[0120] In summary, this embodiment achieves precise coordinated wave generation of nonlinear multidirectional irregular surface waves and internal waves on the basis of existing hardware through targeted configuration of internal modules of the processing module and dual-loop collaborative control strategy. It not only solves the limitations of single wave type simulation, but also takes into account the wave generation accuracy and stability, and can provide reliable wave generation technology support for experimental research related to ocean mixed wave fields.

[0121] Example 3 This embodiment provides a nonlinear multidirectional irregular wave and internal wave generation system and method, specifically applied to the deterministic generation of rogue waves in the laboratory for testing the survivability of marine structures under extreme sea conditions. This embodiment is based on the same system hardware as Embodiment 1. The difference between this embodiment and Embodiment 1 is that the maximum stroke of the wave generation module is extended to ±0.5m; a rogue wave generation unit is added inside the processing module; a wave crest tracking sensor is added to the monitoring unit; and the wave generation method uses phase focusing to generate transient rogue waves.

[0122] The system hardware in this embodiment is basically the same as that in Embodiment 1, including a wave generation module, a monitoring unit, and a processing module, with the following differences: Wave generation module: Similar to Example 1, it consists of 8 independent wave generation plates arranged at non-uniform heights along the water depth direction; in order to generate high-peak distorted waves, the maximum stroke of each plate is extended from ±0.3m to ±0.5m, and the rated torque and power of the servo motor actuator are increased accordingly.

[0123] Monitoring Unit: Based on the water depth velocity measurement unit and the friction wave height sensor unit in Example 1, a wave crest tracking sensor is added: A high-speed capacitive wave height meter with a sampling frequency of 200Hz is arranged at the expected focal position of the distorted wave to accurately capture the instantaneous wave height and waveform of the distorted wave and serve as a feedback signal for closed-loop correction.

[0124] Processing Module: Based on the processing module of Example 1, a new distorted wave generation unit is added; this unit is connected to the wave construction module and is used to calculate the phase-focused wavefront sequence according to the distorted wave parameters input by the user, and then the wave construction module converts it into the target vertical velocity profile.

[0125] This embodiment provides a method for generating distorted waves, including the following steps: Step S1: Inputting distorted wave parameters and synthesizing wavefront sequences The processing module receives the distorted wave parameters input by the user, including: focused wave height. (Target distortion wave peak height, e.g., 0.6m); Focusing position (Distance from wave generator, e.g., 5.0m); Focusing time (Focusing time from the start of wave generation, e.g., 10.0s); Frequency bandwidth (e.g., 0.3Hz to 1.2Hz); Background wave type (optional: no background wave, JONSWAP spectrum random wave, regular wave); Water depth h (e.g., 2.0m); The distorted wave generation unit discretizes the frequency bandwidth into =100 equally spaced frequency components ( =1,2,…, =1,2,…,N). Amplitude of each component The amplitude is allocated according to the root mean square value of the background spectrum (if it is an isolated transient wave, a triangular envelope is used, with the amplitude being the largest at the center frequency and decreasing linearly towards both ends).

[0126] Calculate the wavenumber for each frequency component. It satisfies the linear dispersion relation:

[0127] in = , For gravitational acceleration; set phase:

[0128] So that in At this point, the phase of all components is an integer multiple of 2π (i.e., they are in phase).

[0129] Wavefront time history at the synthesized wave plate (x=0) :

[0130] If background random waves are included, then a random wavefront sequence of the target spectrum is generated according to the method in Example 1. and with Linear superposition, while adjusting the superposition coefficient to achieve the desired total wave height at the focal point. This embodiment uses an isolated transient distortion wave without background waves as an example.

[0131] Step S2: Calculation of target vertical velocity profile The wave construction module is based on the wave surface time history The theoretical horizontal velocity time history at various water depths near the wavemaker is calculated based on the fifth-order Stokes wave theory to fully characterize the strong nonlinear characteristics of the distorted wave. Horizontal velocity at water depth z. for:

[0132] in = , and These are third-order and fifth-order nonlinear velocity correction terms, given by the Stokes expansion. The depth coordinates of the centers of the eight partition plates are... Substituting the values, we obtain 8 velocity-time-duration curves. (t), duration 30s (focusing point at 10s), time step 0.01s, as the target vertical velocity profile. (t) Output.

[0133] Step S3: Displacement Transformation The displacement conversion module receives the target vertical velocity profile. (t), since the distorted wave has extremely strong nonlinear characteristics, this step adopts the same nonlinear iterative compensation mechanism as in Example 1.

[0134] First, based on the center frequency of the distorted wave The linear transfer function matrix T is obtained by interpolation from the frequency-coupling coefficient lookup table, and the initial approximation is solved. Then, the nonlinear mapping database is invoked to obtain the current nonlinear characteristic parameters (dimensionless focused wave height). Nonlinear residual corrections corresponding to / h, maximum wave steepness, etc. Construct an objective function considering nonlinear compensation and wave height constraint:

[0135] In this embodiment, the wave height constraint threshold H_max,i is relaxed to 2.0 times the focused wave height. The Levenberg-Marquardt algorithm is used iteratively until the relative residual of the velocity profile is less than 5%. After the iteration converges, the velocity history of the plates is obtained by trapezoidal integration, and the displacement history is checked to see if it exceeds the maximum travel ±0.5m.

[0136] Step S4: Output of driver instructions The command output module converts the displacement time history into a voltage signal (conversion coefficient 50V / m), limits it to ±10V, and drives the movement of each sub-board after power amplification; the control cycle is 0.02s.

[0137] Step S5: Peak Tracking and Iterative Learning Correction This step uses a wave crest tracking sensor to obtain the actual wave height of the focal point and employs iterative learning control (ILC) offline to correct the initial phase of each frequency component, thereby compensating for focusing deviations caused by actual physical factors such as water pool reflection and nonlinear dispersion. The specific process is as follows: At the moment of focus Within 2 seconds before and after (i.e.) -2s≤ ≤ +2s, the closed-loop correction module calculates the deviation between the actual wave height and the target focused wave height:

[0138] If | |>0.05 Then, the error is recorded, and iterative learning control is used to correct the phase parameters for the next experiment: Add a correction amount

[0139] Where γ is the learning rate (0.2 in this embodiment), the wavefront sequence is resynthesized and steps S1-S4 are repeated until the deviation is ≤5%; after correction, the corrected displacement time history is used for formal experiments.

[0140] Step S6: Other Monitoring and Mode Switching This embodiment generates isolated distorted waves and does not involve switching internal wave modes. However, if the user requires it, the monitoring and calculation module in Embodiment 1 can still be activated.

[0141] Step S7: Loop execution and wave generation termination If iterative learning correction is used, repeat steps S1-S5 until the accuracy requirements are met; then perform formal wave generation: drive the plate movement with the corrected displacement history to generate distorted waves; after wave generation, reset each plate and save the data.

[0142] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A nonlinear multidirectional irregular wave and internal wave generation system, characterized in that, include: Wave-generating module: It is composed of multiple independent wave-generating plates that are hinged together in sequence along the water depth direction. The multiple independent wave-generating plates are arranged in layers with non-uniform heights along the water depth direction, and the height of the plates can be continuously or progressively adjusted according to the center frequency of the preset target frequency band. Each wave-generating plate is connected to an independent actuator. Monitoring unit: includes a water depth velocity measurement unit and a friction wave height sensor unit, which are used to collect the actual horizontal flow velocity at each water depth location and the actual wave height in the near field, far field, surface and bottom layers in real time; Processing module: The processing module is electrically connected to the actuator, the water depth velocity measurement unit, and the friction wave height sensor unit, respectively. The processing module includes: Mode Management Module: Built-in state machine for multiple working modes, used to switch working modes according to user instructions or real-time monitoring data, and load the corresponding preset parameter set; Wave construction module: used to calculate the theoretical horizontal velocity time history at each water depth position at the wave generator based on the selected nonlinear wave theory model, and generate the target vertical velocity profile; Displacement conversion module: used to construct the transfer function matrix between the motion velocity of each plate and the velocity of the fluid in front of the plate. The off-diagonal elements of the function matrix represent the hydrodynamic coupling effect between adjacent plates, and wave height constraint conditions are introduced as equality or inequality constraints. The displacement history of each plate is obtained by solving the constrained linear equation system. Command output module: Its input end is connected to the output end of the displacement conversion module, and its output end is connected to each of the actuators, used to convert the displacement time history into driving commands; Closed-loop correction module: used to compare the actual horizontal flow velocity with the theoretical horizontal velocity, and the actual wave height with the preset wave height threshold, and outputs correction signals only for deviations within the target frequency band; Monitoring and calculation module: includes interface displacement change rate calculation unit, flow velocity shear calculation unit and turbulent kinetic energy estimation unit, used to activate and calculate characteristic parameters in real time in the corresponding working mode, so that the mode management module can determine the switching conditions.

2. The nonlinear multidirectional irregular wave and internal wave generation system according to claim 1, characterized in that, The multiple operating modes of the mode management module include surface wave generation mode, internal solitary wave generation mode, quasi-steady-state interface internal wave generation mode, broken internal wave generation mode, and mixed wave generation mode; the preset parameter set includes nonlinear wave theory type, frequency band filter passband range, PID control coefficient, wave height constraint threshold, control period, and correction intensity level.

3. The nonlinear multidirectional irregular wave and internal wave generation system according to claim 1, characterized in that, The wave construction module selects the corresponding nonlinear wave theory based on the current working mode identifier provided by the mode management module: when the working mode is surface wave generation mode, higher-order Stokes theory is used; when the working mode is internal isolated wave generation mode, MCC theory is used; when the working mode is quasi-steady interface internal wave generation mode, eKdV theory is used; when the working mode is broken internal wave generation mode, mKdV theory is used; when the working mode is mixed wave generation mode, higher-order Stokes theory is used for the surface wave component, and MCC theory or eKdV theory is used for the internal wave component.

4. The nonlinear multidirectional irregular wave and internal wave generation system according to claim 1, characterized in that, The wave construction module is also used to perform frequency band truncation and reconstruction processing on the preset target frequency band range, retaining the target frequency band components; the frequency band truncation and reconstruction processing adopts a bandpass filter or wavelet transform method to limit the frequency components in the input target wave parameters to the preset target frequency band range, and performs waveform reconstruction on the retained target frequency band components based on the nonlinear wave theory.

5. The nonlinear multidirectional irregular wave and internal wave generation system according to claim 1, characterized in that: The displacement conversion module considers the hydrodynamic coupling effect between adjacent plates in the following way: a transfer function matrix is ​​established between the motion velocity of each plate and the velocity of the fluid in front of the plate. The off-diagonal elements of the transfer function matrix represent the mutual influence between adjacent plates. The mapping relationship between the motion velocity of each plate and the target vertical velocity profile is obtained by solving a system of linear equations. In the wave height constraint condition, the preset independent wave height thresholds for the near field, far field, surface layer, and bottom layer are set independently. The near field wave height threshold is 1.0 to 1.2 times the target wave height, and the far field wave height threshold is 0.9 to 1.1 times the target wave height.

6. The nonlinear multidirectional irregular wave and internal wave generation system according to claim 1, characterized in that: The closed-loop correction module includes a deviation calculation unit, a frequency band selection unit, and a correction amount generation unit. The deviation calculation unit is used to calculate a first deviation between the actual horizontal flow velocity and the theoretical horizontal velocity, and a second deviation between the actual wave height and a preset wave height threshold. The frequency band selection unit is used to extract the deviation component located in the target frequency band from the first deviation through a bandpass filter. The correction amount generation unit uses a PID control algorithm, a fuzzy PID control algorithm, or a model predictive control algorithm to generate a correction signal based on the deviation component in the target frequency band, and outputs it to the displacement conversion module or the command output module.

7. A method for generating nonlinear multidirectional irregular waves and internal waves using the system described in any one of claims 1-6, characterized in that, Includes the following steps: S1: Based on the input target wave parameters and the current working mode, call the corresponding nonlinear wave theory model to generate the target vertical velocity profile; S2: By establishing a transfer function matrix that considers the hydrodynamic coupling effect between adjacent plates and introducing wave height constraints, the constrained linear equations are solved to obtain the displacement history of each plate. S3: Convert the displacement time history into drive commands and output them to each actuator to drive the plate movement; S4: Real-time acquisition of actual horizontal flow velocity and actual wave height, and calculation of deviations from target values ​​and preset thresholds; S5: When the deviation exceeds a preset threshold, the deviation component in the target frequency band is extracted to generate a correction signal, and the output path of the correction signal is selected according to the degree of deviation. S6: Activate the monitoring and calculation module according to the current working mode, calculate the feature parameters in real time, and automatically switch the working mode and load the corresponding preset parameter set when the preset switching conditions are met. S7: Repeat steps S4 to S6 until the preset wave generation time is reached.

8. The nonlinear multidirectional irregular wave and internal wave generation method according to claim 7, characterized in that: In step S5, the step of selecting the output path of the correction signal according to the degree of deviation is as follows: when the deviation is less than or equal to the first threshold, the correction signal is output to the displacement conversion module to re-solve the displacement time history; when the deviation is greater than the first threshold, the correction signal is directly output to the instruction output module and superimposed on the current driving instruction.

9. The nonlinear multidirectional irregular wave and internal wave generation method according to claim 7, characterized in that, In step S6, the preset switching conditions include: when the rate of change of interface displacement is lower than the first rate threshold in the quasi-steady-state interface internal wave mode, the closed-loop correction intensity is reduced; when the flow velocity shear exceeds the first shear threshold or the turbulent kinetic energy exceeds the first kinetic energy threshold in the internal solitary wave mode or quasi-steady-state mode, the mode is switched to the broken internal wave mode and the correction frequency band is expanded and the control gain is increased; when the flow velocity shear is lower than the second shear threshold and the turbulent kinetic energy is lower than the second kinetic energy threshold in the broken internal wave mode, the mode is restored to the internal solitary wave mode.

10. The nonlinear multidirectional irregular wave and internal wave generation method according to claim 7, characterized in that, In S1, the current working mode includes surface wave generation mode, internal isolated wave generation mode, quasi-steady interface internal wave generation mode, broken internal wave generation mode and mixed wave generation mode; when the working mode is mixed wave generation mode, the velocity components of surface wave and internal wave are calculated separately and then linearly superimposed to obtain the total target velocity profile.

Images