Large-scale test pool wave flow testing device and method

Abstract

The invention discloses a large-scale test pool wave current testing device and method, and the method comprises the steps: forming a transverse movement foundation through a C-shaped guide rail, and enabling the transverse movement to be used for transverse movement; transverse movement is driven by a winding and unwinding reel, meanwhile, a power cable is wound and unwound, and after the power cable is in place, the power cable is locked by an electromagnetic brake located on the transverse movement frame. And after reaching a test area, the transverse linear module realizes transverse movement control during test, the longitudinal linear module controls longitudinal displacement, and the vertical lifting mechanism controls lifting of the instrument mounting rack, so that wave velocity and wave height measurement in the test area is finally realized, and an intelligent decision instruction is output through the driving control module for equipment control. Real-time intelligent decision making is carried out, invalid data collection is reduced, the experiment efficiency is improved, and real-time feedback capacity is provided for transient characteristic research in the dynamic evolution process.

Description

Technical Field

[0001] The present invention relates to the technical field of wave and current testing, and in particular to a wave and current testing device and method for a large-scale test pool. Background Art

[0002] In water tank experiments (such as water conservancy projects, ocean engineering, and ship hydrodynamic testing), accurate measurement of wave and current parameters is a core requirement for studying water flow characteristics and verifying model performance. Traditional wave and current detection instruments (such as current meters and wave sensors) usually need to be fixed at a specific location in the water tank, manually deployed, and the instrument posture and depth are adjusted, and positioning is achieved by relying on mechanical brackets or buoy systems. However, water tank experiments often require data from multiple locations and working conditions, and manual operation has the following problems:

[0003] Inefficiency: Each adjustment requires reinstalling the bracket or moving the buoy, which is time-consuming and labor-intensive, especially in large pools or complex working conditions. Large positioning errors: Manual adjustments rely on visual or simple ruler positioning, which makes it difficult to ensure the precise match between the instrument and the target coordinates, affecting data consistency. Lack of intelligent methods: The real-time and correctness of the data cannot be guaranteed. It is very likely that the measurement data will be erroneous due to problems with the sensor or other reasons, resulting in a waste of time.

[0004] In the existing technology, the core of the wave and current detection instrument fixing and adjustment method based on manual operation relies on the operator to manually complete the installation, movement and positioning of the instrument. The structural composition includes: a fixed bracket / buoy device: a physical support structure composed of a metal bracket, a float, a counterweight, etc., used to carry the wave and current detection instrument (such as an ADCP current meter, a pressure wave sensor); a manual adjustment component: a mechanical component including a height-adjustable telescopic rod, an angle adjustment knob, a locking bolt, etc., used to coarsely adjust the instrument's water immersion depth and horizontal posture; a simple positioning tool: a scale ruler on the side wall of the pool, a handheld laser rangefinder or a visual marking point to assist manual judgment of the instrument's horizontal position and depth.

[0005] The above structure has the following limitations: single instrument deployment requires the collaboration of multiple personnel, the adjustment process is time-consuming (usually more than 30 minutes), and it is difficult to adapt to the needs of multi-point, high-frequency experiments. Positioning accuracy relies on experience, and the error of manual visual alignment of the scale generally exceeds ±5cm. The instrument attitude adjustment accuracy is insufficient (e.g., inclination error >3°), affecting data reliability. Dynamic adaptability is poor, and the fixed bracket / buoy cannot move dynamically with waves or currents, making it difficult to achieve continuous trajectory measurement or real-time working condition response. Summary of the Invention

[0006] The purpose of the present invention is to provide a large-scale test pool wave and current testing device and method, which can make real-time intelligent decisions, reduce invalid data collection, improve experimental efficiency, and provide real-time feedback capabilities for the study of transient characteristics of dynamic evolution processes.

[0007] To achieve the above object, the present invention provides the following solutions:

[0008] A large-scale test pool wave and current testing device, comprising:

[0009] C-shaped track, fixedly connected to the bottom of the bridge beam through a connecting plate;

[0010] The transverse frame is connected to the C-type track and is equipped with a transverse linear module and an electromagnetic brake;

[0011] The longitudinal frame is connected to the lower part of the transverse frame and is equipped with a longitudinal linear module and a built-in drive control module. The longitudinal linear module controls the longitudinal displacement;

[0012] The vertical frame is connected to the bottom of the longitudinal frame and has a built-in battery pack;

[0013] A vertical lifting mechanism connected to the bottom of the vertical frame;

[0014] An instrument mounting frame, connected to the end of the vertical lifting mechanism, is used to install measuring instruments, including ADCP and wave height sensor;

[0015] The retractable reel is installed on both sides of the C-shaped track and is used for cable retraction and retraction.

[0016] A large-scale test pool wave and current testing method, applied to the large-scale test pool wave and current testing device described above, comprises the following steps:

[0017] The C-type guide rail forms the basis for lateral movement, which is used for lateral movement. The lateral movement is driven by the reel, which retracts and releases the power cable at the same time. After it is in place, it is locked by the electromagnetic brake located on the lateral frame. After reaching the test area, the lateral movement control during the test is realized by the lateral linear module, the longitudinal linear module controls the longitudinal displacement, and the vertical lifting mechanism controls the lifting and lowering of the instrument mounting frame, ultimately realizing the measurement of wave velocity and wave height in the test area, and outputting intelligent decision-making instructions through the drive control module to control the equipment.

[0018] Preferably, the drive control module outputs intelligent decision instructions to control the device, specifically including:

[0019] The drive control module has an embedded edge computing unit to achieve real-time data access between the wave height sensor array and ADCP, and uses a dual-channel transmission mechanism to reduce latency and ensure real-time response. A wave height time series prediction model is established based on the LSTM model to achieve wave height time series prediction and abnormal alarm. The ADCP velocity field is reconstructed through an autoencoder, and the spatial distribution differences of the reconstruction error are calculated to identify the non-steady-state structure of the turbulent vortex, and a velocity dynamic monitoring model is established. A parameter dynamic adjustment strategy is designed based on CNN. By dynamically adjusting model parameters and fusing multi-model outputs, intelligent decision-making instructions are output in real time, ultimately achieving adaptive control of the ocean simulation environment.

[0020] Preferably, a wave height time series prediction model is established based on the LSTM model to realize wave height time series prediction and abnormal alarm, specifically including:

[0021] The current wave height data is collected and preprocessed in real time, and the wave height data of the Tth second is maintained as the model input; the model outputs the standardized wave height prediction sequence of the τth second, and the physical quantity prediction value is obtained by denormalization. After obtaining the actual measurement value Y at the τth second, calculate the absolute residual When the residual error at any forecast time point exceeds the preset threshold, an alarm is triggered immediately.

[0022] Preferably, the dynamic adjustment strategy of CNN design parameters specifically includes:

[0023] A CNN model is constructed based on the dynamic characteristics of waves and water flow. The wave height sensor array data is converted into a space-time matrix and input into a CNN model with void convolution to extract the characteristics of narrowband waves and focused waves, and output the parameters of main frequency and wave steepness in real time. The vertical flow velocity profile data of ADCP is encoded into a pseudo-color image, and the attention mechanism is used to enhance the CNN model's ability to capture vortex boundaries and quantify the indicators of vorticity and turbulent kinetic energy. Based on the real-time analysis results, a dynamic parameter adjustment strategy is designed.

[0024] Preferably, by dynamically adjusting model parameters and fusing multiple model outputs, real-time output of intelligent decision instructions specifically includes:

[0025] First, the wave main frequency / wave steepness parameters output by the CNN model and the vorticity / turbulent kinetic energy indicators output by the ADCP are received in real time and integrated into a multi-dimensional state vector. Based on the preset test objectives, a dynamic adjustment rule library is constructed, including: threshold trigger mechanism, PID feedback control, reinforcement learning strategy, and embedded safety boundary constraints. The adjustment amount is calculated according to the rule library, and intelligent decision-making instructions are generated to drive the equipment execution.

[0026] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the large-scale test pool wave and current testing method as described above is implemented.

[0027] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0028] (1) The present invention proposes a large-scale test pool wave and current testing device, which realizes the autonomous deployment and precise positioning of wave and current detection instruments in the pool through the coordinated control of an automated mobile platform and a high-precision positioning module, solving the problems of cumbersome operation and poor positioning accuracy caused by manual installation and adjustment of existing wave and current detection instruments in the pool. It makes real-time intelligent decisions through machine learning, reduces invalid data collection, improves experimental efficiency, and provides real-time feedback capabilities for the study of transient characteristics of dynamic evolution processes, providing an efficient, reliable, and intelligent large-scale test pool measurement and instrument deployment integrated solution.

[0029] (2) The large-scale test pool wave and current testing device proposed in this invention is specifically used for the installation and movement of wave and current detection instruments in the pool. This device replaces manual handling and adjustment with an automated mobile platform, and combines high-precision positioning and attitude control technology to significantly reduce the complexity of manual operation. At the same time, it significantly improves the positioning accuracy of the instrument, solves the measurement data distortion caused by human error in traditional methods, and provides reliable support for dynamic continuous measurement. It also makes real-time intelligent decisions through machine learning, reduces invalid data collection, improves experimental efficiency, and provides real-time feedback capabilities for the study of transient characteristics of dynamic evolution processes. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0031] Figure 1 A structural diagram of a large-scale test pool wave and current testing device provided by the present invention;

[0032] Figure 2 This is a diagram of the lateral movement mechanism of the present invention;

[0033] Figure 3 A top view of the wave and current testing device for a large test pool according to the present invention;

[0034] Figure 4 This is a flowchart for realizing intelligent decision-making of the present invention;

[0035] Among them, 1-transverse frame, 2-C-type track, 3-horizontal linear module, 4-longitudinal frame, 5-longitudinal linear module, 6-vertical frame, 7-vertical lifting mechanism, 8-instrument mounting frame, 9-measuring instrument, 10-reel. DETAILED DESCRIPTION

[0036] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0037] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0038] Example 1

[0039] like Figure 1-Figure 3 As shown, the present invention provides a large-scale test pool wave and current testing device, comprising:

[0040] C-type track 2, fixed to the bottom of the bridge girder by welding or bolting through a connecting plate;

[0041] The transverse frame 1 is connected to the C-shaped track 2 and is equipped with a transverse linear module 3 and an electromagnetic brake;

[0042] The longitudinal frame 4 is connected to the lower side of the transverse frame 1 and is equipped with a longitudinal linear module 5 and a built-in drive control module. The longitudinal linear module 5 controls the longitudinal displacement.

[0043] The vertical frame 6 is connected to the bottom of the longitudinal frame 6 and has a built-in battery pack;

[0044] A vertical lifting mechanism 7 is connected to the bottom of the vertical frame 6;

[0045] An instrument mounting frame 8 is connected to the end of the vertical lifting mechanism 7 and is used to mount a measuring instrument 9, which includes an ADCP and a wave height sensor;

[0046] The retractable reel 10 is arranged on both sides of the C-shaped track 2 and is used for retracting and releasing the cable.

[0047] Specifically, the working principle of the large-scale test pool wave and current test device is as follows:

[0048] Connected to the underside of the measuring bridge's girder via connecting plates welded or bolted to form a point-to-point connection, the C-shaped guide rails 2 form the foundation for lateral movement, enabling wide-range lateral movement. This wide-range movement is driven by a retractable reel 10, which simultaneously retracts and releases the power cable. Once in position, it is locked by an electromagnetic brake located on the traversing frame. Upon reaching the test area, the transverse linear module 3 controls lateral movement during testing, the longitudinal linear module 5 controls longitudinal displacement, and the vertical lifting mechanism 7 controls the raising and lowering of the instrument mounting frame 8, enabling three-dimensional measurement within the test area. A rotational drive control can be provided at the upper end of the instrument mounting frame 8 for local angle control, as needed.

[0049] Example 2

[0050] The present invention provides a large-scale test pool wave and current testing method, which is applied to the large-scale test pool wave and current testing device mentioned above, and comprises the following steps:

[0051] The C-shaped guide rail 2 forms a lateral movement basis for lateral movement; the lateral movement is driven by the reel 10, which retracts and releases the power cable at the same time. After it is in place, it is locked by the electromagnetic brake located on the lateral movement frame; after reaching the test area, the lateral movement control during the test is realized by the lateral linear module 3, the longitudinal linear module 8 controls the longitudinal displacement, and the vertical lifting mechanism 7 controls the lifting and lowering of the instrument mounting frame 8, ultimately realizing the wave velocity and wave height measurement in the test area, and outputting intelligent decision-making instructions through the drive control module to control the equipment.

[0052] Specifically, the testing method of the present invention includes:

[0053] 1. Preliminary Preparation

[0054] First, the test parameters are determined according to the test requirements, including the target position (longitudinal coordinates of the pool), water depth conditions, the arrangement of the wave height sensor array (such as linear or array type), and the movement speed and sampling frequency of the ADCP. Then the equipment is installed and calibrated. The wave height sensor is modularly installed on the lifting device, the sensor placement height is adjusted according to the water depth, and the ADCP navigation module is installed in the middle of the track. The probe is adjusted to the target depth (such as 0.2 times the water depth from the bottom of the pool or full water depth layered measurement). Then the system self-test is started to verify the module movement accuracy, calibrate the wave height sensor and ADCP coordinates, and finally, after inputting the parameters through the control system, the device will move to the predetermined position along the track and perform the final positioning by controlling the horizontal and vertical linear modules and the lifting mechanism.

[0055] 2. Wave height test

[0056] Wave height testing is divided into two modes: linear testing and array testing. In the linear test, the sensor is placed in a straight line through the horizontal and vertical linear modules (horizontal or vertical placement can be selected as needed) and accurately positioned to the target coordinates. The lifting device fine-tunes the sensor height to adapt to the wave measurement range. After starting the wave generation system, the single-point continuous sampling time must cover at least 10 typical wave cycles. If an array test is used, it is mainly to measure three-dimensional waves. The xyz directions of the three-dimensional waves all change, so an array test is used. At the same time, the lifting device is linked to adjust the sensor height according to the change in water depth to achieve efficient collection of three-dimensional wave field data.

[0057] 3. Flow velocity section test

[0058] Velocity testing relies on a track system to drive the ADCP module at a constant speed (e.g., 0.3 m / s) along the measurement direction. Laser positioning corrects the trajectory to ensure straight-line motion. During navigation, the ADCP probe depth is adjusted to obtain a vertical velocity profile. For special needs, such as those in the underlying boundary layer, movement can be paused and high-frequency sampling (≥20 Hz) initiated. This method completely solves the problem of ADCP yaw on traditional boats while avoiding the safety risks of manual intervention.

[0059] 4. Combination Testing

[0060] For wave-current coupling experiments, wave height sensor subarrays can be deployed adjacent to the ADCP underway module, enabling synchronous acquisition of wave height and current velocity data through timestamp alignment. The device also supports rapid reconfiguration of test configurations, for example, switching between a dense array (0.2m spacing) and a sparse, wide-area array (2m spacing) within 30 minutes by replacing modular brackets.

[0061] 5. Data processing and effect verification

[0062] During the test, the human-machine interface displays the wave height time history curve, velocity vector field, and device motion trajectory in real time. Abnormal conditions (such as sensor signal loss) trigger automatic pause protection. In the post-processing stage, a coordinate mapping algorithm is used to convert the mobile measurement data into a wave height / velocity field on a fixed spatial grid, and a report is generated comparing the results with traditional methods. This device reduces the time required to adjust the height of the wave height meter, shortens ADCP section testing time, and reduces errors in wave height spatial positioning.

[0063] 6. Principle of intelligent decision-making: Figure 4 The flowchart for intelligent decision-making is shown. The specific principles are as follows:

[0064] (1) An edge computing unit is embedded in the drive control module to achieve real-time data access to the wave height sensor array and ADCP. To reduce latency, a dual-channel transmission mechanism is designed. Feature data is transmitted to the edge unit via Gigabit Ethernet, and an independent control instruction channel is simultaneously opened to ensure real-time response of the motion module. After hardware integration, time synchronization calibration must be completed to ensure that the deviation between the sensor data timestamp, module position information and edge computing clock is at the millisecond level;

[0065] (2) A training set was constructed based on historical test data, and a bimodal machine learning model was developed: a LSTM network was used to establish a wave height time series prediction model, which input the waveform data of the past 10 seconds and output the predicted value of the next 1 second. An alarm was triggered when the difference between the actual measured value and the predicted residual exceeded 3 times the standard deviation.

[0066] The development of a wave height time series prediction model involves first acquiring and preprocessing historical time series data from the test pool's wave height sensor (including cleaning, filtering, and normalization). Training sample pairs are constructed based on a specified input window (T seconds in the past, e.g., 10-second data points) and output window (τ seconds in the future, e.g., 1-second prediction points). Subsequently, an LSTM neural network (consisting of an LSTM layer and an output layer) is designed and trained to minimize the mean squared error (MSE) between the predicted values ​​and the actual future wave heights. A validation set is used to optimize hyperparameters and prevent overfitting. Finally, the model's performance is evaluated on a test set.

[0067] The specific process of using this model to predict and trigger an alarm is as follows: real-time acquisition and preprocessing of current wave height data, maintaining the latest T-second data as model input; the model outputs a standardized wave height prediction sequence for the next τ seconds, and the predicted value of the physical quantity is obtained by denormalization. When the actual measurement value Y arrives in the next τ seconds, calculate the absolute residual When the residual error at any prediction time point exceeds a preset threshold (k times the baseline standard deviation σ_benchmark, calculated from the model's prediction residuals on an independent normal dataset), an alarm is immediately triggered. An autoencoder is designed to reconstruct the ADCP velocity field. By calculating the spatial distribution of the reconstruction error, non-stationary structures such as turbulent vortices can be identified. After model training, a lightweight deployment is implemented, achieving real-time inference speeds exceeding 30 fps on edge units.

[0068] At the same time, a flow velocity dynamic detection model is established;

[0069] (3) Construct a CNN based on the dynamic characteristics of waves and currents: convert the wave height sensor array data into a spatiotemporal matrix (row = sensor position, column = time series), input it into a CNN model with dilated convolution, extract features such as narrowband waves and focused waves, and output parameters such as main frequency and wave steepness in real time. Encode the vertical velocity profile data of ADCP into a pseudo-color image, and use the attention mechanism to enhance the model's ability to capture vortex boundaries and quantify indicators such as vorticity and turbulent kinetic energy. Based on the real-time analysis results, design a dynamic parameter adjustment strategy;

[0070] The steps for implementing dynamic parameter adjustment and intelligent decision-making are as follows: First, the wave frequency / wave steepness parameters output by the CNN and the vorticity / turbulent kinetic energy indicators output by the ADCP model are received in real time and integrated into a multidimensional state vector. Based on the preset test objectives (such as maintaining a specific wave spectrum and stimulating vortex shedding), a dynamic adjustment rule library is constructed:

[0071] 1) Threshold trigger mechanism (e.g. automatic reduction of wave generator frequency when wave steepness exceeds a limit);

[0072] 2) PID feedback control (dynamically adjusting the flow rate and pump power based on the deviation between the measured vortex value and the target value);

[0073] 3) Reinforcement learning strategy (training the agent to select the optimal action based on the state vector, such as adjusting the deflector angle). Safety boundary constraints (such as maximum flow rate limits) are also embedded. The decision engine executes the following steps per frame (≥30 fps): analyzing the real-time state → matching the rule base → calculating the adjustment amount → generating execution instructions (such as PWM signals) → driving the actuators (wave generators, pumps, valves, etc.). Through online verification modules (such as retesting features after adjustments), the closed-loop optimization strategy achieves autonomous control through the "perception-analysis-decision-execution" process.

[0074] (4) Develop dedicated control software to achieve three-dimensional visual monitoring, including dynamic display of wave height field cloud map, velocity vector field and device motion trajectory:

[0075] (5) The intelligent decision-making system accesses the wave height sensor array and ADCP data in real time through the edge computing unit embedded in the drive control module, and uses a dual-channel transmission mechanism to reduce latency and ensure real-time response. The system realizes wave height time series prediction and abnormal alarm based on the LSTM model, identifies the non-steady-state vortex structure in the velocity field through the autoencoder, and uses a CNN with void convolution to extract the spatiotemporal characteristics of waves. By dynamically adjusting the model parameters and fusing the outputs of multiple models (wave height prediction, vortex identification, velocity monitoring, etc.), the system can output intelligent decision-making instructions in real time (such as equipment attitude adjustment or emergency plan triggering), and finally realize adaptive control of the ocean simulation environment, effectively improving the stability and safety of marine equipment in complex fluid environments.

[0076] Intelligent decision-making implementation code:

[0077]

[0078]

[0079]

[0080]

[0081]

[0082]

[0083] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the large-scale test pool wave and current testing method as described above is implemented.

[0084] The large-scale test pool wave and current testing device and method provided by the present invention systematically solve the above-mentioned shortcomings through an automated mobile platform and high-precision closed-loop control technology. The specific beneficial effects are as follows:

[0085] 1. High-precision positioning and automatic leveling: Traditional methods rely on manual positioning using a visual scale and adjusting the instrument's posture, resulting in horizontal positioning errors greater than 5cm and tilt errors greater than 3°. This new method utilizes a positioning module and a multi-axis tilt sensor to achieve improved accuracy.

[0086] 2. Automated movement and rapid deployment: Traditional processes require manual handling and repeated calibration, and a single instrument deployment takes up to 30 minutes. This invention uses remote control to quickly move to the measurement area and quickly complete deployment, supporting rapid switching experiments in multiple working conditions.

[0087] 3. Dynamic continuous measurement capability: Traditional fixed mounts only support static single-point measurement. Dynamic conditions require manual disassembly and reassembly of the instrument, resulting in data gaps. The present invention uses a mobile platform that can continuously navigate along the wave propagation direction or a preset trajectory, enabling dynamic, full-process collection of velocity fields and wave parameters.

[0088] 4. Safety and simplified operation: Traditional manual handling poses risks of equipment slipping and personnel wading into water. This invention uses a fully automated remote control solution, allowing operators to deploy and move instruments without entering the pool. Combined with anti-collision sensors and an emergency braking mechanism, this completely eliminates safety hazards and lowers the barrier to entry for experimental operation.

[0089] 5. Lack of intelligent methods: Traditional data processing cannot guarantee the real-time and correctness of data. It is very likely that the measured data will be erroneous due to problems with the sensor or other reasons, which will waste time. However, through machine learning and other methods, real-time intelligent decision-making can be made, invalid data collection can be reduced, experimental efficiency can be improved, and real-time feedback capabilities can be provided for the study of transient characteristics of dynamic evolution processes.

[0090] In summary, the present invention replaces manual vision with high-precision positioning, replaces manual handling with automated movement, and replaces static measurement with dynamic tracking. It not only solves the core contradictions of traditional technologies such as "low precision, poor efficiency, and limited scenarios", but also upgrades wave and current detection from a crude mode relying on experience to an intelligent and standardized measurement system. It makes real-time intelligent decisions through machine learning, reduces invalid data collection, improves experimental efficiency, and provides real-time feedback capabilities for the study of transient characteristics of dynamic evolution processes. It provides highly reliable and intelligent experimental tools for fields such as marine engineering, water conservancy and hydropower, and promotes the technological leap from static analysis to dynamic modeling in water flow dynamics research.

[0091] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, or of course, by hardware. Based on this understanding, the essence of the above technical solution or the part that contributes to the existing technology can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or certain parts of the embodiments.

[0092] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The above examples are only intended to help understand the method and core concept of the present invention. At the same time, those skilled in the art will find that the specific implementation methods and application scopes may vary based on the concept of the present invention. In summary, the contents of this specification should not be construed as limiting the present invention.

Claims

1. A large-scale test pool wave and current testing device, characterized in that: include: C-shaped track, fixedly connected to the bottom of the bridge beam through a connecting plate; A transverse frame connected to the C-shaped track and equipped with a transverse linear module and an electromagnetic brake; The longitudinal frame is connected to the lower side of the transverse frame and is equipped with a longitudinal linear module and a built-in drive control module. The longitudinal linear module controls the longitudinal displacement. A vertical frame connected to the bottom of the longitudinal frame and having a built-in battery pack; A vertical lifting mechanism connected to the bottom of the vertical frame; An instrument mounting frame connected to the end of the vertical lifting mechanism and used for mounting measuring instruments, including ADCP and wave height sensor; The retractable reels are arranged on both sides of the C-shaped track and are used for retracting and releasing the cables.

2. A large-scale test pool wave and current testing method, applied to the large-scale test pool wave and current testing device according to claim 1, characterized in that: The following steps are involved: The C-type guide rail forms the basis for lateral movement, which is used for lateral movement. The lateral movement is driven by the reel, which retracts and releases the power cable at the same time. After it is in place, it is locked by the electromagnetic brake located on the lateral frame. After reaching the test area, the lateral movement control during the test is realized by the lateral linear module, the longitudinal linear module controls the longitudinal displacement, and the vertical lifting mechanism controls the lifting and lowering of the instrument mounting frame, ultimately realizing the measurement of wave velocity and wave height in the test area, and outputting intelligent decision-making instructions through the drive control module to control the equipment.

3. A large-scale test pool wave and current testing method according to claim 2, characterized in that: The drive control module outputs intelligent decision-making instructions to control the equipment, specifically including: The drive control module has an embedded edge computing unit to achieve real-time data access between the wave height sensor array and ADCP, and uses a dual-channel transmission mechanism to reduce latency and ensure real-time response. A wave height time series prediction model is established based on the LSTM model to achieve wave height time series prediction and abnormal alarm. The ADCP velocity field is reconstructed through an autoencoder, and the spatial distribution differences of the reconstruction error are calculated to identify the non-steady-state structure of the turbulent vortex, and a velocity dynamic monitoring model is established. A parameter dynamic adjustment strategy is designed based on CNN. By dynamically adjusting model parameters and fusing multi-model outputs, intelligent decision-making instructions are output in real time, ultimately achieving adaptive control of the ocean simulation environment.

4. A large-scale test pool wave and current testing method according to claim 3, characterized in that: The wave height time series prediction model is established based on the LSTM model to realize wave height time series prediction and abnormal alarm, specifically including: The current wave height data is collected and preprocessed in real time, and the wave height data of the Tth second is maintained as the model input; the model outputs the standardized wave height prediction sequence of the τth second, and the physical quantity prediction value is obtained by denormalization. After obtaining the actual measurement value Y at the τth second, calculate the absolute residual When the residual error at any forecast time point exceeds the preset threshold, an alarm is triggered immediately.

5. A large-scale test pool wave and current testing method according to claim 4, characterized in that: The dynamic adjustment strategy based on CNN design parameters specifically includes: A CNN model is constructed based on the dynamic characteristics of waves and water flow. The wave height sensor array data is converted into a space-time matrix and input into a CNN model with void convolution to extract the characteristics of narrowband waves and focused waves, and output the parameters of main frequency and wave steepness in real time. The vertical flow velocity profile data of ADCP is encoded into a pseudo-color image, and the attention mechanism is used to enhance the CNN model's ability to capture vortex boundaries and quantify the indicators of vorticity and turbulent kinetic energy. Based on the real-time analysis results, a dynamic parameter adjustment strategy is designed.

6. A large-scale test pool wave and current testing method according to claim 5, characterized in that: The method of dynamically adjusting model parameters and fusing multiple model outputs to output intelligent decision instructions in real time specifically includes: First, the wave main frequency / wave steepness parameters output by the CNN model and the vorticity / turbulent kinetic energy indicators output by the ADCP are received in real time and integrated into a multi-dimensional state vector. Based on the preset test objectives, a dynamic adjustment rule library is constructed, including: threshold trigger mechanism, PID feedback control, reinforcement learning strategy, and embedded safety boundary constraints. The adjustment amount is calculated according to the rule library, and intelligent decision-making instructions are generated to drive the equipment execution.

7. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, a large-scale test pool wave and current testing method according to any one of claims 2 to 6 is implemented.

Images