Circulating water tank earthquake simulation test system and test method based on electromagnetic driving and artificial intelligence simulation

Abstract

The invention relates to a circulating water tank earthquake simulation test system and test method based on electromagnetic driving and artificial intelligence simulation, a circulating water tank main body, a wave making system and a three-axis electromagnetic seismic station are integrated for the first time, X / Y / Z three-axis vibration can be realized, and each axis is independently controlled. A similar criterion library is constructed by combining a contact sensor and non-contact optical measurement, working condition errors of an actual environment and an experimental environment can be reduced as much as possible according to different experimental purposes, complex working conditions that a structure bears wave, water flow and earthquake effects at the same time in a real marine environment are simulated, and an experimental result has higher reference value. One-key process control is achieved through the central control unit, waveform generation and data processing are conducted through the AI technology, and the experiment efficiency and the scientific research depth are greatly improved.

Description

Technical Field

[0001] This invention relates to a circulating water tank seismic simulation test system and test method based on electromagnetic drive and artificial intelligence simulation, belonging to the field of experimental technology of water conservancy engineering, coastal engineering, geotechnical engineering and earthquake engineering. Background Technology

[0002] In the field of earthquake simulation testing, especially in test scenarios involving underwater environments or multiple coupled environments, earthquake simulation tanks and shaking tables are core equipment for conducting related research, and their performance directly determines the reliability and accuracy of the test results. Currently, traditional earthquake simulation tanks or shaking tables generally adopt hydraulic or electric servo drive technologies, such as a hydraulically driven shaking table system for tank experiments disclosed in patent CN110206012A, an electric servo-driven underwater earthquake simulation shaking table device disclosed in CN112146838A, and a large-scale simulation test system for simulating the combined effects of earthquakes, waves, and ocean currents disclosed in patent CN104020007B. These patent solutions represent the current mainstream technological direction in this field.

[0003] However, existing traditional drive schemes and the experimental systems built upon them all have many inherent defects, making it difficult to meet the ever-increasing demands for accurate simulation. Regarding the drive method itself, while hydraulic drive schemes offer the advantage of providing enormous thrust, they have insurmountable shortcomings, such as slow response speed (typically exceeding 10ms), inability to accurately reproduce the rich high-frequency components (frequency >20Hz) in seismic waves; low control accuracy, with waveform distortion exceeding 5%, making it difficult to meet the accuracy requirements of complex seismic wave simulations; high energy consumption; and the risk of oil leakage, which can easily pollute the experimental environment. The hydraulic drive schemes used in existing technologies CN110206012A and patent CN104020007B both suffer from the aforementioned problems. Although patent CN104020007B has achieved the integrated simulation function of earthquake, wave and ocean current combined action, the earthquake simulation device of the system still achieves vibration by switching the horizontal actuator and the vertical actuator through a three-stage servo valve. In essence, it is still limited by the inherent defects of the hydraulic system. The response speed is difficult to break through the millisecond level, the control accuracy is constrained by factors such as oil compressibility and pipeline loss, and the waveform reproduction accuracy is limited.

[0004] Compared to hydraulic drives, electric servo drive solutions offer improved control precision, such as the electric servo drive structure used in patent CN112146838A. However, they still fall short in achieving multi-axis (three-axis and above) coupled motion and high-frequency response (frequency > 50Hz). Furthermore, the presence of mechanical transmission backlash and inertial delay results in insufficient waveform reproduction fidelity. In particular, when simulating high-frequency, multi-directional composite earthquakes, phase errors and waveform distortion are quite noticeable, failing to meet the requirements for high-precision multi-directional earthquake simulation.

[0005] On the other hand, in addition to the inherent defects of the driving method, traditional earthquake simulation test devices and systems also have many shortcomings in supporting technologies. The observation methods rely heavily on contact sensors, which have limited deployment points and interfere with the model, making it difficult to obtain continuous deformation and flow data across the entire field, thus affecting the integrity and accuracy of the test data. In terms of test design, there is a lack of scientific and supporting methods. When conducting multi-parameter collaborative test simulations involving earthquakes, water flow, waves, and sediment, there is a lack of effective basis for selecting similarity criteria, resulting in serious distortion of some key parameters (such as earthquake input spectrum, sediment transport rate, and structural dynamic response) during the test, making it difficult to accurately reproduce the actual field conditions. The termination conditions of traditional scour tests usually rely on a single direct observation, which cannot respond to the dynamic characteristics of the scour process, easily leading to deviations in the timing of the test termination and affecting the scientific validity of the test results.

[0006] In summary, existing traditional earthquake simulation flumes and shaking tables have significant shortcomings in several aspects, including driving performance, observation methods, experimental design, data processing, and system intelligence. These limitations make it difficult to meet the demands of high-precision earthquake simulation tests under complex conditions, hindering the development of earthquake simulation technology and the in-depth advancement of research in related fields. Therefore, developing an earthquake simulation device and system that overcomes these shortcomings, possesses high-frequency response, high-precision control, multi-directional coupling simulation capabilities, and is equipped with comprehensive observation, experimental design, and data processing technologies has become an urgent technical problem to be solved in this field. Summary of the Invention

[0007] This invention provides a circulating water tank seismic simulation test system and method based on electromagnetic drive and artificial intelligence simulation. It can reproduce complex three-dimensional seismic motion with high precision on the basis of simulating a stable marine environment, and can observe the dynamic coupling response process of water-soil-structure in the whole field by combining contact sensors and non-contact optical measurements.

[0008] The technical solution adopted by this invention to solve its technical problem is: A circulating water tank seismic simulation test system based on electromagnetic drive and artificial intelligence simulation includes a circulating water tank body, which includes an upstream water inlet section, a test section and a downstream water storage section connected in sequence. The test section serves as the observation and operation area, and a stainless steel sandbox for testing is embedded at its bottom. A triaxial electromagnetic seismic simulation subsystem is installed at the bottom of the test section. It includes a shaking table, an electromagnetic shaking table support column, and an electromagnetic drive device. A stainless steel sand box is placed on the shaking table, which is supported by the electromagnetic shaking table support column. At the same time, the electromagnetic shaking table support column provides a position reference for the electromagnetic drive device. That is, the electromagnetic shaking table support column forms a space for installing the electromagnetic drive device below the test section. The electromagnetic drive device provides translational vibration force in the X, Y, and Z axis directions to the stainless steel sand box. A wave-generating and rectification system is installed at the beginning of the upstream intake section to simulate waves and set the incoming flow conditions. A downstream water circulation and treatment system, including a water pump, is installed at the downstream water storage section. The system pumps the water that has been collected in the downstream water storage section after the experiment back to the upstream water intake section, forming a water circulation channel. Non-contact optical measurement equipment is installed at the opening of the main body of the circulating water tank. Several contact sensors are also installed in the circulating water tank earthquake simulation test system. The triaxial electromagnetic earthquake simulation subsystem, wave generation and rectification system, downstream water circulation and treatment system, non-contact optical measurement equipment and contact sensors are all connected to the intelligent control and data processing center to realize distributed management and control of the entire circulating water tank earthquake simulation test system. Furthermore, the wave-generating and rectification system of the upstream intake section includes a wave generator and a rectification device, with the wave generator located close to the test section; the wave generator adopts a pusher type, rocker type, or piston type according to the wave sequence of the sea state to be simulated, and the rectification device adopts a multi-layer honeycomb plate, a damping mesh, or a guide plate. The downstream water circulation and treatment system of the downstream water storage section includes a water tank, a sedimentation tank, a water pump, a sand pump, and a water injection pipeline. A sedimentation tank is set at the bottom of the water tank on the side closest to the test section. The water pump is installed on a pipeline connected to the water tank and this pipeline is connected to the upstream water inlet section. The sand pump is installed on a pipeline connected to the sedimentation tank. One end of the water injection pipeline is connected to an external water source, and the other end is connected to the water tank. Furthermore, a vibrating support plate is horizontally installed on the top of the vibration table, and the surface of the vibrating support plate supports the stainless steel sand box. The electromagnetic drive device includes a horizontal magnetic drive end, a horizontal magnetic receiver end, a vertical magnetic drive end, and a vertical magnetic receiver end. Several vertical magnetic drive ends are vertically installed on the base of the electromagnetic vibration table support column located below the vibration plate, and vertical magnetic receiving ends are installed on the bottom surface of the vibration plate opposite to the vertical magnetic drive ends; several horizontal magnetic drive ends are vertically installed on the lateral mounting surface of the electromagnetic vibration table support column relative to the vibration plate, and horizontal magnetic receiving ends are installed on the side wall of the vibration plate opposite to the horizontal magnetic drive ends. Furthermore, several centroid adjustment blocks are set at the four corners of the vibration bearing plate to finely adjust the centroid position of the circulating water tank seismic simulation test system. Elastic roller guides are installed at the vibration table and the vibration plate support to ensure that the vibration table moves along the preset degrees of freedom; The drive structure of the vibration table and the space formed by the electromagnetic vibration table support column for installing the electromagnetic drive device are covered with high conductivity or high magnetic permeability material to suppress the outward radiation of strong electromagnetic fields. Furthermore, the top opening of the stainless steel sandbox is flush with the bottom plate of the upstream water inlet section and the downstream water storage section, and a sandbox drainage outlet is opened at the bottom of the stainless steel sandbox; a flexible sealing rubber skirt is provided at the connection between the bottom of the test section and the bottom of the stainless steel sandbox. Furthermore, the contact sensors include an accelerometer, an optical displacement sensor, a magnetic field sensor, an ADV flow meter, a turbidity meter, a laser ranging matrix, a capacitive wave height meter, and a pore water pressure gauge array. The acceleration sensor is installed on the bottom surface of the vibration plate and is used to measure the acceleration response of the vibration table and the stainless steel sandbox in three directions. Several optical displacement sensors are installed on the base of the electromagnetic vibration table support column and the bottom surface of the vibration plate to measure the displacement of the vibration table in three directions. The magnetic field sensor is placed close to the electromagnetic drive device to monitor the magnetic field strength around the electromagnetic drive device. The ADV flow meter, turbidity meter, and capacitive wave height meter extend into the main body of the circulating water tank. The ADV flow meter measures the three-dimensional flow velocity distribution at different locations in the water tank and sedimentation tank in real time. The turbidity meter monitors the real-time changes in sediment concentration in the main body of the circulating water tank. The capacitive wave height meter measures water surface fluctuations. The laser ranging matrix includes several laser displacement sensors, which are arranged in a regular grid pattern and mounted on the opening end of the main body of the circulating water tank; The pore water pressure gauge array consists of several pore water pressure gauges, all of which are buried in a stainless steel sand box to measure the distribution and changes of pore water pressure in sand. The non-contact optical measurement equipment includes a high-speed camera and a high-definition video recorder, both of which are mounted at the opening of the main body of the monitoring circulating water tank. The high-speed camera is used to capture the subtle processes of sand movement, water turbulence, or high-speed deformation of the structural model, while the high-definition video recorder is used to record the dynamics of the entire test section. Furthermore, the intelligent control and data processing center includes a central control unit, an intelligent waveform generation module, and an AI image processing unit; The central control unit uniformly schedules the start-up, shutdown, parameter setting, and operation synchronization of the triaxial electromagnetic seismic simulation subsystem, wave generation and rectification system, downstream water circulation and treatment system, non-contact optical measurement equipment, and contact sensors. The intelligent waveform generation module embeds machine learning algorithms to learn and simulate the characteristics of real seismic wave spectra and generate multi-directional driving signals that meet experimental requirements. The AI ​​image processing unit operates on high-speed photography and video recording data in real time, extracting the full-field displacement, strain field, and flow velocity field distribution. The test method for the circulating water tank seismic simulation test system based on electromagnetic drive and artificial intelligence simulation includes the following steps: Step S1, test preparation: Lay sand samples in a stainless steel sand box according to the test design requirements, and inject test water into the water tank through the water injection pipe until the water level reaches the preset height. Step S2: Initialize the system and check the connection status and normal operation of the main body of the circulating water tank, the triaxial electromagnetic seismic simulation subsystem, the wave generation and rectification system, the downstream water circulation and treatment system, the non-contact optical measurement equipment, and the contact sensors. Saturation treatment was performed through the sand box drainage outlet, and the sand sample was allowed to stand until it reached a stable state. Calibrate accelerometers, pore water pressure gauge arrays, ADV flow meters, laser ranging matrices, high-speed cameras, and high-definition video recorders; The intelligent control and data processing center sets the test stop conditions; Step S3: Select the dominant similarity criterion based on the experimental research object, and set the initial test parameters of the circulating water flume seismic simulation test system according to the selected dominant similarity criterion; Step S4: Based on the dominant similarity criterion determined in step S3, the actual environmental parameters are converted to the experimental environmental parameters; after the experimental parameter conversion is completed, wave-current-seismic coupling simulation is performed. In step S5, the central control unit triggers the triaxial electromagnetic seismic simulation subsystem, wave generation and rectification system, downstream water circulation and treatment system, non-contact optical measurement equipment and contact sensors to acquire data simultaneously. Step S6: The central control unit analyzes the sensor data in real time and uses AI algorithm to perform seismic vibration denoising on the original measurement values. Based on the denoised data, it determines whether the preset test termination condition is met. If the test termination condition is met, proceed to step S7; otherwise, continue the denoising process. Specifically, data denoising involves combining the raw flow velocity signal measured by the ADV flow meter with the high-precision seismic station vibration data output in step S5, performing frequency domain filtering to remove periodic disturbances caused by earthquakes and obtain a purer water flow velocity value. Filter out the pressure fluctuation component directly caused by seismic vibration in the data collected by the pore water pressure gauge array; Step S7: The AI ​​image processing unit processes and intelligently analyzes the data determined in step S6. Specifically, The AI ​​image processing unit is used to perform particle image velocimetry and digital image correlation analysis on the acquired image data to extract the overall flow velocity, displacement and strain field. Multi-physics coupling analysis of sediment transport, soil liquefaction, and structural response was conducted by combining sensor data. Based on deep learning algorithms, the evolution of the bed surface is reconstructed in three dimensions and intelligently identified to generate real-time three-dimensional landforms and display flow field characteristics. Step S8: End the test, shut down the triaxial electromagnetic seismic simulation subsystem, wave generation and rectification system, downstream water circulation and treatment system, non-contact optical measurement equipment and contact sensors, and start the sand pump to remove the silt accumulated in the sedimentation tank. Furthermore, in step S3, based on the experimental research object, the dominant similarity criterion is selected and the initial experimental parameters are set, specifically: If the research object is the seismic response of a structure, priority should be given to ensuring the geometric similarity and seismic dynamic similarity of the structure, and the structural model size and seismic wave input spectrum should be determined first. If the research object is sediment erosion, priority should be given to ensuring that the initiation of sediment particles is similar to that of water flow. First, the sediment particle size and water flow velocity in the model should be determined. If the research object is seabed liquefaction, priority should be given to ensuring the similarity of soil dynamics and pore water pressure response, and the seismic vibration intensity and saturated sand parameters should be determined first. After determining the dominant similarity criterion and initial experimental parameters, we continue to evaluate the impact of other secondary similarity criteria on the experimental results, select the similarity criterion with the least impact on experimental distortion for approximation, and finally determine all experimental parameters. Furthermore, step S4, which performs wave-current-seismic coupling simulation, includes: Step S41: If the research object is the seismic response of a structure, the actual seismic spectrum is converted into the experimental seismic spectrum firstly based on the seismic dynamic similarity criterion; then, based on the geometric similarity and water flow similarity criteria, and considering the influence of the seismic dynamic similarity criterion on the hydrodynamic similarity, the actual wave parameters and water flow parameters are converted into experimental values. If the research object is sediment erosion, the actual flow parameters and wave parameters are converted into experimental values ​​first based on the similarity criteria of sediment initiation and water flow motion. Then, based on the seismic dynamic similarity criterion, and taking into account the influence of the similarity criteria of sediment initiation and water flow motion, the actual seismic spectrum is converted into the experimental seismic spectrum. If the research object is seabed liquefaction, the seismic vibration intensity and saturated sand parameters are determined first based on the soil dynamic similarity and pore water pressure response similarity criteria. Then, the hydrodynamic parameter conversion is completed based on the sediment initiation similarity and water flow movement similarity criteria, while also considering the influence of the soil dynamic similarity and pore water pressure response similarity criteria. Step S42: Start the water pump and wave generator, set the converted wave parameters and water flow parameters through the central control unit, and use the rectifier to homogenize the incoming flow. Monitor and adjust the wave parameters and water flow parameters in real time until a stable hydraulic environment that meets the test requirements is formed. Step S43: The central control unit continues to set the converted test seismic spectrum and convert it into a three-axis drive signal; Step S44: Start the triaxial electromagnetic seismic simulation subsystem and apply horizontal and vertical vibrations through the horizontal magnetic drive end and the vertical magnetic drive end, respectively; the vibrating plate reproduces the seismic waveform along the preset degrees of freedom under the guidance of the elastic roller guide.

[0009] By employing the above technical solutions, the present invention has the following beneficial effects compared to the prior art: 1. The present invention provides a circulating water tank seismic simulation test system based on electromagnetic drive and artificial intelligence simulation. It adopts three-axis electromagnetic drive technology and accurately realizes complex vibrations of three axes (X, Y, and Z) or even more degrees of freedom through multiple independently controlled electromagnetic drive units. The response speed reaches the millisecond level or even the sub-millisecond level. It can reproduce the high-frequency components in the earthquake without delay. The interference between axes is small. Multi-axis coordinated control achieves higher precision and purer vibration output. 2. The present invention provides a circulating water tank seismic simulation test system based on electromagnetic drive and artificial intelligence simulation, which integrates contact sensor array and non-contact optical observation technology to achieve all-round data acquisition from "point" to "field". It avoids the defects of limited deployment points and interference model of traditional contact measurement. At the same time, it also supports the data processing method to filter out noise, which greatly improves the monitoring accuracy of data under strong vibration environment and effectively reduces measurement error. 3. The test method of the circulating water tank seismic simulation test system based on electromagnetic drive and artificial intelligence simulation provided by the present invention is the first to propose an intelligent selection mechanism of similarity criteria based on research objectives. For different test objects (structural response, sediment erosion, seabed liquefaction), the core similarity criteria are given priority to meet, and the conflicts of secondary criteria are coordinated. This solves the problem of parameter distortion caused by the blind selection of similarity criteria in traditional tests, and significantly improves the simulation accuracy and scientific nature of the test. 4. The test method of the circulating water tank seismic simulation test system based on electromagnetic drive and artificial intelligence simulation provided by the present invention integrates functions such as wave generation, current generation, triaxial seismic simulation, water circulation treatment, and multi-source data fusion analysis. It can simulate various working conditions such as structural response, seabed liquefaction, and sediment scouring under the coupled environment of "wave-current-earthquake". The entire process is automatically controlled by the central control unit without human intervention, which improves the efficiency of test decision-making. Attached Figure Description

[0010] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0011] Figure 1 This is a detailed diagram of the test section in the circulating water tank seismic simulation test system based on electromagnetic drive and artificial intelligence simulation provided by the present invention; Figure 2 This is an overall side view of the circulating water tank seismic simulation test system based on electromagnetic drive and artificial intelligence simulation provided by the present invention; Figure 3 This is a top view of the test section in the circulating water tank seismic simulation test system based on electromagnetic drive and artificial intelligence simulation provided by the present invention; Figure 4 This is a top view of the test section in the circulating water tank earthquake simulation test system based on electromagnetic drive and artificial intelligence simulation provided by the present invention; Figure 5 This is a three-dimensional view of the test section in the circulating water tank seismic simulation test system based on electromagnetic drive and artificial intelligence simulation provided by the present invention; Figure 6 This is a schematic diagram of the functions of the intelligent control and data processing center in the circulating water tank earthquake simulation test system based on electromagnetic drive and artificial intelligence simulation provided by the present invention.

[0012] In the diagram: 1 is a turbidity meter; 2 and 4 are both high-definition video recorders; 3 is a high-speed camera; 5 is an ADV flow meter; 6 is a capacitive wave height meter; 7 is a stainless steel sandbox; 8 is a pore water pressure gauge array; 9 is a sandbox drain outlet; 10 is a flexible sealing rubber skirt; 11 is a centroid adjustment block; 12 is an accelerometer; 13 is an optical displacement sensor; 14 is a vibration support plate; 15 is an elastic roller guide rail; 16 is a horizontal magnetic drive end; 17 is a horizontal magnetic receiver end; 18 is a magnetic field sensor; 19 is a vertical magnetic drive end; 20 is a vertical magnetic receiver end; 21 is an emergency stop limit protection cable; 22 is an electromagnetic vibration table support column; 23 is an electromagnetic shielding area; 24 is the main body of the circulating water tank; 25 is a rectifier; 26 is a wave generator; 27 is a water injection pipe; 28 is a water pool; 29 is a sedimentation tank; 30 is a sand pump; 31 is a water pump; and 32 is a laser ranging matrix. Detailed Implementation

[0013] The present invention will now be described in further detail with reference to the accompanying drawings. In the description of this application, it should be understood that the terms "left side," "right side," "upper part," "lower part," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. "First," "second," etc., do not indicate the importance of the components, and therefore should not be construed as a limitation of the present invention. The specific dimensions used in this embodiment are only for illustrating the technical solution and do not limit the scope of protection of the present invention.

[0014] In the field of earthquake simulation testing involving underwater environments or multiple coupled environments, traditional hydraulic drive schemes, while offering the advantage of providing enormous thrust, suffer from slow response speed, low control accuracy, high energy consumption, and the risk of oil leakage. Even when hydraulic drive is replaced with electric servo drive, its performance in coupled motion and high-frequency response remains insufficient, especially when simulating high-frequency, multi-directional composite ground motions, where phase errors and waveform distortions are quite pronounced, failing to meet the requirements of high-precision multi-directional earthquake simulation. Furthermore, traditional earthquake simulation test devices and systems lack scientifically sound support in terms of observation methods and test design, affecting the scientific validity of the test results.

[0015] To address the aforementioned issues, this application proposes a circulating water tank seismic simulation test system and method based on electromagnetic drive and artificial intelligence simulation.

[0016] First, the provided circulating water tank seismic simulation test system based on electromagnetic drive and artificial intelligence simulation is described. Figures 1-5 This demonstrates the system's structure from various angles. Compared to existing technologies, this application achieves the following breakthroughs in the system: First, it uses electromagnetic drive instead of hydraulic or electric servo drive, resulting in exceptionally high drive response and accuracy, especially with the adoption of independent three-axis control to achieve interference-free coupling. Second, it employs a combination of contact and non-contact optical measurement to achieve continuous, interference-free data acquisition across the entire field.

[0017] Specifically, the circulating water tank seismic simulation test system based on electromagnetic drive and artificial intelligence simulation includes a circulating water tank body 24. As the core structure and basic carrier of the entire test system, the circulating water tank body is manufactured using corrosion-resistant high-strength stainless steel through professional welding processes, and has an overall rectangular closed-loop structure. The bottom of the circulating water tank body is precision-machined to ensure that the overall flatness error is controlled within ±1mm, thus ensuring the uniformity of water flow boundary conditions and the reliability of experimental results. The circulating water tank body includes an upstream inlet section, a test section, and a downstream storage section connected sequentially. These three parts are connected by a precision hydraulic connection, forming a complete and closed water flow circulation channel. The test section, as the core observation and operation area, has observation windows made of high-strength, high-transparency composite tempered glass on both sides. A stainless steel sandbox 7 for testing is embedded at its bottom, which not only facilitates non-contact optical observation and image acquisition of internal water flow, sediment transport, and structural response by researchers, but also effectively withstands long-term water pressure and possible external impacts. The stainless steel sandbox mentioned here is used to hold experimental sand or similar foundation materials. Its corrosion resistance ensures stable performance even after long-term exposure to salt water or chemical solutions. The top opening of the sandbox is flush with the bottom plate of the test section to ensure that the water flow boundary conditions are consistent with the actual seabed or foundation scenario. The flexible sealing rubber skirt 10 connects the top of the sandbox to the bottom plate of the water tank. It is made of highly elastic and aging-resistant rubber material, which allows the vibration table to move in multiple degrees of freedom while effectively preventing water and sediment from leaking out from the joints, maintaining the airtightness of the internal environment of the test section.

[0018] A triaxial electromagnetic seismic simulation subsystem is installed on an independent foundation at the bottom of the test section. This system serves as the power core and excitation source of this application. The foundation uses large-mass concrete blocks and is isolated from the surrounding structures through vibration isolation trenches, effectively preventing vibration energy leakage and interference from external environmental vibrations, ensuring the purity and reliability of the excitation during the experiment. The triaxial electromagnetic seismic simulation subsystem is then firmly connected to the bottom plate of the water tank through a high-rigidity support structure, enabling precise simulation of seismic waves of different directions and intensities during the experiment. This provides a realistic and controllable physical environment for studying the dynamic response of the seabed and structures under combined wave-earthquake action. Specifically, the triaxial electromagnetic seismic simulation subsystem includes a shaking table, an electromagnetic shaking table support column 22, and an electromagnetic drive device. A stainless steel sandbox is placed on the shaking table, which is supported by the electromagnetic shaking table support column. The electromagnetic shaking table support column, as a key load-bearing and positioning structure, is made of high-strength cast iron through precision casting and machining, possessing extremely high rigidity and stability. Preferably, a vibration support plate 14 is horizontally installed on the top of the shaking table, with the surface of the vibration support plate supporting the stainless steel sandbox. The electromagnetic vibration table support column not only supports the upper vibration plate, the stainless steel sand box, and the entire weight of the sand and water inside, but also provides a position reference for the electromagnetic drive device. That is, the electromagnetic vibration table support column forms a space for installing the electromagnetic drive device below the test section, ensuring that even under strong unbalanced excitation, the entire vibration platform can still maintain a precise motion trajectory and dynamic balance. The electromagnetic drive device provides translational vibration force in the X, Y, and Z axes to the stainless steel sand box.

[0019] The vibration support plate, used to mount the stainless steel sandbox, is a precision-machined cast iron platform with a flatness error of less than 0.1 mm / m. Elastic roller guides 15 are installed at the connection between the vibration table and the vibration support plate. This specially designed guiding mechanism can withstand huge vertical loads while strictly constraining the vibration table to move only in the X, Y, and Z translational degrees of freedom, effectively suppressing any undesired rotation or lateral swaying and ensuring accurate reproduction of the input seismic waveform. Several center-of-gravity adjustment blocks 11 are installed at the four corners of the vibration support plate. These stainless steel counterweights can move along the longitudinal and transverse guides at the bottom of the vibration support plate via threaded connections, precisely adjusting the system's center of gravity to coincide with the geometric center, reducing off-center vibration. To further enhance safety, emergency stop limit protection cables 21 are installed at the ends of the elastic roller guides on the vibration table (i.e., the extreme positions of the vibration table's movement). When the vibration amplitude exceeds the preset safety range or the system becomes uncontrollable, mechanical locking or damping braking is immediately triggered, rapidly limiting the vibration table's displacement, effectively preventing equipment structural damage and avoiding safety accidents during the experiment.

[0020] The electromagnetic drive device includes a horizontal magnetic drive end 16, a horizontal magnetic receiving end 17, a vertical magnetic drive end 19, and a vertical magnetic receiving end 20. Several vertical magnetic drive ends are vertically installed on the base of the electromagnetic vibration table support column below the vibration plate, and vertical magnetic receiving ends are installed on the bottom surface of the vibration plate opposite to the vertical magnetic drive ends. Vertical (Z-axis) vibration is generated by the aforementioned magnetic coupling unit. Several horizontal magnetic drive ends are vertically installed on the lateral mounting surface of the electromagnetic vibration table support column relative to the vibration plate, and horizontal magnetic receiving ends are installed on the side wall of the vibration plate opposite to the horizontal magnetic drive ends. Preferably, in this application, one set is provided in both the X-axis and Y-axis directions to jointly achieve horizontal (X and Y-axis) vibration excitation.

[0021] It is important to note that the electromagnetic drive system generates an extremely strong time-varying magnetic field during operation. To prevent these strong magnetic fields from causing electromagnetic interference to the high-sensitivity measuring equipment (such as the high-speed camera 3, laser displacement sensor, ADV current meter 5, etc.) arranged around the test section, leading to distorted measurement data or equipment damage, this system has a specially designed electromagnetic shielding area 23. This area includes the drive structure of the vibration table and the space formed by the electromagnetic vibration table support column for installing the electromagnetic drive device. It is encased in a material with high conductivity or high magnetic permeability; preferably, a 1mm thick pure copper plate is welded into a complete enclosed enclosure, completely enclosing all the magnetic drive end, magnetic receiver end, and the upper moving parts of the support column, creating a clean electromagnetic environment for the experimental area and ensuring the accuracy and reliability of all sensor measurement data.

[0022] Regarding the upstream intake section, a wave-generating and rectification system is installed at its starting position to simulate waves and set incoming flow conditions. The wave-generating and rectification system includes a wave generator 26 and a rectification device 25. The wave generator is positioned close to the test section, and the rectification device is positioned from... Figure 1 From a mid-range perspective, it is located approximately 1.5 meters to the right (rear) of the wave generator. The wave generator employs a pusher-plate, rocker-plate, or piston-type mechanism based on the wave sequence of the simulated sea state. Commands are sent from the central control unit to generate waves with specified wave height, period, and pattern. The rectifier device uses multi-layered honeycomb panels, damping mesh, or guide vanes to effectively break up large-scale eddies, making the water flow more uniform and stable, and providing a high-quality inflow for the test section.

[0023] The downstream storage section is mainly used to collect the water flow after the experiment, initially separate the sediment particles carried in the water, and stabilize the system water level and regulate the flow rate. A downstream water circulation and treatment system is installed in the downstream storage section, including a water pump 31, which pumps the water flow collected in the downstream storage section back to the upstream inlet section, forming a water circulation channel. The downstream water circulation and treatment system also includes a water tank 28, a sedimentation tank 29, water pumps, sand pumps 30, and a water injection pipe 27. The water tank is used to store the circulating water. A sedimentation tank is set at the bottom of the water tank near the test section, with a conical bottom design to facilitate sediment collection. The water pumps are installed on a pipe connected to the water tank, and this pipe is connected to the upstream inlet section, which mainly receives the water flow from the water pumps and initially regulates the water flow pattern. The sand pumps are installed on a pipe connected to the sedimentation tank and automatically remove sand periodically. One end of the water injection pipe is connected to an external water source, and the other end is connected to the water tank, automatically compensating for evaporation and leakage losses.

[0024] The above is the first breakthrough of this application. The second breakthrough of this application is that a non-contact optical measurement device is installed at the opening of the main body of the circulating water tank, and several contact sensors are also installed in the circulating water tank seismic simulation test system. The triaxial electromagnetic seismic simulation subsystem, wave generation and rectification system, downstream water circulation and treatment system, non-contact optical measurement device and contact sensors are simultaneously connected to the intelligent control and data processing center to realize distributed management and control of the entire circulating water tank seismic simulation test system.

[0025] The contact sensors include an accelerometer 12, an optical displacement sensor 13, a magnetic field sensor 18, an ADV flow meter, a turbidity meter 1, a laser ranging matrix 32, a capacitive wave height meter 6, and a pore water pressure gauge array 8. The accelerometer is installed on the bottom surface of the vibration plate and is used to measure the acceleration response of the vibration table and the stainless steel sandbox in three directions. Several optical displacement sensors are installed on the base of the electromagnetic vibration table support column and the bottom surface of the vibration plate to measure the displacement of the vibration table in three directions. A magnetic field sensor is positioned near the electromagnetic drive unit to monitor the magnetic field strength around it. An ADV flowmeter, turbidity meter, and capacitive wave height meter extend from the opening of the circulating water tank. The ADV flowmeter measures the three-dimensional velocity distribution at different locations within the water tank and sedimentation tank in real time. The turbidity meter monitors the real-time changes in sediment concentration within the circulating water tank, and the capacitive wave height meter measures water surface fluctuations. A laser ranging matrix includes several laser displacement sensors arranged in a regular grid array at the opening of the circulating water tank, capable of measuring the distance between any point and the initial bed surface, thus displaying the instantaneous scour depth at any point in real time. A pore water pressure gauge array includes several pore water pressure gauges, all embedded in a stainless steel sandbox, used to measure the pore water pressure distribution in the sand and its changes over time, reflecting the dynamic response of the soil under liquefaction, seepage, and seismic action.

[0026] The non-contact optical measurement equipment includes a high-speed camera and a high-definition video recorder (2, 4), both of which are mounted at the opening of the main body of the monitoring circulating water tank. The high-speed camera is used to capture the subtle processes of sand movement, water turbulence, or high-speed deformation of the structural model. The high-definition video recorder is used to record the dynamics of the entire test section, record the experimental process from different angles, and realize high-precision real-time acquisition of all sensor and image data through the data acquisition and synchronization unit.

[0027] Finally, regarding the circulating water tank seismic simulation test system based on electromagnetic drive and artificial intelligence simulation provided in this application, an intelligent control and data processing center has also been designed. Figure 6 This is a schematic diagram of its functional modules. It mainly includes a central control unit, an intelligent waveform generation module, and an AI image processing unit. The central control unit uniformly schedules the start-up, shutdown, parameter setting, and operation synchronization of the three-axis electromagnetic seismic simulation subsystem, wave generation and rectification system, downstream water circulation and treatment system, non-contact optical measurement equipment, and contact sensors. As the core of intelligent decision-making and scheduling for the entire experimental system, it possesses highly automated process control and intelligent criterion response capabilities. It can intelligently set various start-up and shutdown criteria, including equilibrium conditions, based on user-preset experimental objectives and physical process characteristics. For example, in scour tests, the equilibrium condition for automatic termination of the test can be customized as "bed topography change rate is below a threshold," "key point scour depth time history is stable," or "sediment transport flux reaches dynamic equilibrium." The intelligent waveform generation module embeds machine learning algorithms to learn and simulate the characteristics of real seismic wave spectra, generating multi-directional driving signals that meet experimental requirements. The AI ​​image processing unit is equipped with a high-performance GPU computing card, relying on a parallel acceleration architecture to process massive amounts of image data collected by high-speed cameras and high-definition video recorders in real time. It employs advanced particle image velocimetry (PIV) and digital image correlation (DIC) algorithms to accurately capture and track subtle features of sand particle motion, water turbulence, and structural deformation during the experiment, thereby reconstructing the dynamic distribution of the full-field three-dimensional displacement, strain, and velocity fields with high spatiotemporal resolution. Based on the above physical field data, the unit further introduces deep learning algorithms, combining convolutional neural networks (CNN) and generative adversarial networks (GAN) to intelligently identify and reconstruct the evolution of the bed surface morphology in three dimensions.

[0028] The intelligent control and data processing center also features a remote monitoring and control platform for information interaction, integrating an IoT-based remote monitoring and control platform. Authorized users can remotely monitor the experimental status in real time from mobile terminals (such as smartphone apps) or desktop clients via secure network connections. The platform provides dynamic visualization of core experimental parameters (such as flow rate, pore water pressure, and key point displacement), phased feedback on the experimental progress, and intelligent prediction of the remaining experimental time based on real-time data and preset criteria. Users can remotely intervene in the experimental process based on monitoring data and system alerts, achieving distributed management and control of the entire system.

[0029] The aforementioned system description provided in this application is an integrated system encompassing wave generation, current generation, triaxial seismic simulation, water circulation treatment, and multi-source data fusion. Its initial purpose was to comprehensively simulate various working conditions such as structural response, seabed liquefaction, and sediment erosion under a wave-current-earthquake coupled environment. However, a comprehensive system design alone is insufficient; therefore, this application further provides matching experimental methods. These methods also incorporate several breakthroughs, such as intelligent matching of dominant criteria to coordinate conflicts between secondary criteria, making the experimental design more scientific. A "noise template" is constructed based on seismic station vibration data, and AI-powered intelligent denoising significantly improves the signal-to-noise ratio, effectively avoiding the large measurement errors caused by the lack of effective denoising methods in existing technologies. Another advantage is the fully automated control process, multi-dimensional dynamic termination conditions, and response to dynamic characteristics such as erosion / liquefaction, truly achieving experimental automation.

[0030] The test method for the circulating water tank seismic simulation test system based on electromagnetic drive and artificial intelligence simulation includes the following steps: Step S1, test preparation: Lay sand samples in a stainless steel sandbox according to the test design requirements, and inject test water into the water tank through the water injection pipe until the water level reaches the preset height.

[0031] Step S2: Initialize the system and check the connection status and normal operation of the main body of the circulating water tank, the triaxial electromagnetic seismic simulation subsystem, the wave generation and rectification system, the downstream water circulation and treatment system, the non-contact optical measurement equipment, and the contact sensors; apply lubricating oil to the mechanical connections of the triaxial electromagnetic seismic simulation subsystem to ensure smoother vibration simulation. Saturate the sand sample through the sandbox drainage outlet 9 until it reaches a stable state; calibrate the accelerometer, pore water pressure gauge array, ADV flow meter, laser ranging matrix, high-speed camera, and high-definition video recorder; the intelligent control and data processing center sets the test stop conditions; at the start of the experiment, all monitoring devices immediately begin collecting data and transmitting it in real time to the central control unit, which determines whether the test stop conditions are met.

[0032] Step S3 is one of the core innovations of the entire experimental method. Based on the user's preset experimental research object, such as "structural seismic response", "seabed liquefaction", "sediment erosion", etc., the system intelligently recommends and prioritizes meeting 1-2 dominant similarity criteria that are most relevant to it, and determines the initial experimental parameters accordingly.

[0033] Specifically, this method prioritizes the following: if the research object is the seismic response of a structure, then prioritize ensuring geometric and seismic dynamic similarity of the structure, first determining the structural model dimensions and seismic wave input spectrum; if the research object is sediment erosion, then prioritize ensuring similarity initiation of sediment particles and similarity in water flow, first determining the model sediment particle size and water flow velocity; if the research object is seabed liquefaction, then prioritize ensuring similarity in soil dynamics and pore water pressure response, first determining the seismic vibration intensity and saturated sand parameters. After determining the dominant similarity criterion and initial experimental parameters, the impact of other secondary similarity criteria on the experimental results is evaluated, and the similarity criterion with the least impact on experimental distortion is selected for approximation, ultimately determining all experimental parameters. This method effectively avoids the systematic distortion caused by conflicting similarity criteria in traditional experiments, significantly improving the simulation accuracy and scientific rigor of the experiment.

[0034] Preferably, a specific implementation process is provided regarding how to intelligently select similarity criteria and set parameters based on research objectives.

[0035] First, when the research object is "structural seismic response", such as assessing the dynamic response of structures like offshore wind turbine foundations and offshore platforms under seismic action, the system intelligently recommends and prioritizes structural geometric similarity and seismic dynamic similarity (Cauchy similarity criterion) as the dominant similarity criteria.

[0036] Regarding the determination of the dominant similarity criterion and the derivation of parameters, (1) Determination of geometric similarity constant: First, based on the experimental site and model manufacturing conditions, determine the geometric similarity constant S. L :S L =L m / L p , where L m L represents the feature size of the model structure. p These are the feature dimensions of the prototype structure.

[0037] (2) Application of Cauchy similarity criterion: Cauchy number C a =ρV 2 / E represents the ratio of inertial force to elastic force, requiring the model to have the same Cauchy number as the prototype, i.e., C. a,m =C a,p Where ρ is density, V is characteristic velocity, and E is elastic modulus. From C a,m =C a,p Derivation: ρ m V m2 / E m =ρ p V p 2 / E p Introducing geometric similarity S L Similar to time S t (V=L / t, therefore S) V =S L / S t Define the density similarity constant S. ρ =ρ m / ρ p similarity constant of elastic modulus S E =E m / E p Substituting into the above equation, we get: S ρ (S L / S t ) 2 / S E =1. After simplification, the expression for the time similarity constant St is obtained: .

[0038] (3) Derivation of acceleration similarity constant: Therefore, the acceleration similarity constant for, The above Substituting the expression, we get: .

[0039] (4) Seismic wave input spectrum conversion: Based on the above similarity relationship, the prototype earthquake acceleration time history is converted. Converted to experimental parameters, i.e., acceleration time history , That is, the acceleration amplitude of the prototype seismic wave is calculated according to S... a Zoom in, press S on the timeline t Compress it.

[0040] Regarding compromises on secondary similarity criteria: After determining the parameters of the dominant criterion mentioned above, the system further evaluates other criteria. For example, the Freud similarity criterion (gravitational similarity) typically requires... However, this is consistent with the Cauchy criterion derived from... Inconsistencies are often observed. Given that this experiment focuses on seismic dynamic response, the system determines the Froude criterion as a secondary criterion, allowing it to be disregarded. The system also indicates the potential distortion of gravity effects caused by this criterion in the test parameter settings, prompting users to pay attention to its impact on the results.

[0041] Second, when the research object is "sediment erosion", such as studying the erosion mechanism of the seabed around the structure under the combined action of earthquake and wave current, the system intelligently recommends and prioritizes the similarity criteria of sediment particle initiation similarity (Shields similarity criterion) and water flow movement similarity (Froude similarity criterion) as the dominant similarity criteria.

[0042] Regarding the determination of the dominant similarity criterion and the derivation of its parameters, (1) Application of the Froude similarity criterion: Froude number The ratio of inertial force to gravity is represented by the Froude number, which requires the model to have the same Froude number as the prototype. .Depend on And the gravitational acceleration similarity constant The similarity constant of the flow velocity was derived. and time similarity constant : .

[0043] (2) Application of Shields similarity criterion: Shields parameter Characterizing the initiation conditions of sediment particles, among which For bed shear stress, For the density of sediment, The median particle size is required. The model must have the same Shields parameters as the prototype, i.e. = .

[0044] For the transition zone from the turbulent smooth region to the rough region, the bed surface shear stress With flow rate It is proportional to the square of, that is .generation = have to: Assuming the model and prototype use the same fluid (water), then ,and =1. Substitute After sorting, the similarity constant of sediment particle size was obtained. : If the model and the prototype use the same mud and sand material, then The above formula simplifies to: .

[0045] Compromise regarding the secondary similarity criterion: The system then assesses seismic dynamic similarity (Cauchy criterion). If the Cauchy criterion is strictly satisfied, it requires... This is consistent with the principles established by Froude. Conflict. Given that this experiment focuses on scour under hydrodynamic action, the system prioritizes maintaining the Froude similarity and approximates the temporal similarity of ground motions, i.e., adopts... The seismic waves are time-scaled, and the impact of this approximation on the simulation of high-frequency components of the seismic waves is clearly stated in the report.

[0046] Third, when the research object is "seabed liquefaction", such as studying the liquefaction process and extent of saturated sandy soil foundation under seismic action, the system intelligently recommends and prioritizes soil dynamic similarity and pore water pressure response similarity as the dominant similarity criteria.

[0047] Regarding the determination of the dominant similarity criterion and parameter derivation, (1) Soil dynamic similarity: To achieve similarity in soil dynamic properties, the normal consolidation state and dynamic stress-strain relationship of the model soil and the prototype soil must be similar. This is usually achieved by controlling the relative density of the soil. This is achieved through vibration intensity. The relative density is similar. This means that the model soil and the prototype soil must have the same relative density.

[0048] (2) Similarity of pore water pressure response: In order to achieve similarity of pore water pressure growth, the excess pore water pressure generated by seismic vibration in the model and the prototype must be similar. With initial effective cover stress The ratio changes in the same way.

[0049] Pore ​​water pressure similarity constant : Similarly, if the soil material is the same, ,but (3) Determination of input ground motion: In order to make the soil elements of the model reach the same dynamic shear stress level as the prototype, the seismic acceleration of the model base is input. Scaling should be performed according to the following relationship: If the model and the prototype soil material are the same ( ), then: The time similarity constant is usually determined based on the dynamic properties of the soil (such as shear wave velocity similarity). If the soils are identical and have the same density, then... .

[0050] Compromise regarding secondary similarity criteria: Under this research objective, similarity of water flow motion (Froude criterion) may become a secondary criterion. While ensuring similarity between soil dynamics and pore pressure response, the system will recommend that users conduct experiments at the lowest possible flow rate to minimize the interference of water flow dynamics on the pore pressure field and soil structural stability, and will focus on examining liquefaction phenomena dominated by seismic vibrations during the analysis.

[0051] Step S4: Based on the dominant similarity criterion determined in step S3, the actual environmental parameters are converted to the experimental environmental parameters; after the experimental parameter conversion is completed, wave-current-seismic coupling simulation is performed. The specific steps are as follows: Step S41: If the research object is the seismic response of a structure, the actual seismic spectrum is converted into the experimental seismic spectrum firstly based on the seismic dynamic similarity criterion; then, based on the geometric similarity and water flow similarity criteria, and considering the influence of the seismic dynamic similarity criterion on the hydrodynamic similarity, the actual wave parameters and water flow parameters are converted into experimental values. If the research object is sediment erosion, the actual flow parameters and wave parameters are converted into experimental values ​​first based on the similarity criteria of sediment initiation and water flow motion. Then, based on the seismic dynamic similarity criterion, and taking into account the influence of the similarity criteria of sediment initiation and water flow motion, the actual seismic spectrum is converted into the experimental seismic spectrum. If the research object is seabed liquefaction, the seismic vibration intensity and saturated sand parameters are determined first based on the soil dynamic similarity and pore water pressure response similarity criteria. Then, the hydrodynamic parameter conversion is completed based on the sediment initiation similarity and water flow movement similarity criteria, while also considering the influence of the soil dynamic similarity and pore water pressure response similarity criteria. Step S42: Start the water pump and wave generator, set the converted wave parameters and water flow parameters through the central control unit, and use the rectifier to homogenize the incoming flow. Monitor and adjust the wave parameters and water flow parameters in real time until a stable hydraulic environment that meets the test requirements is formed. In step S43, the central control unit continues to set the converted test seismic spectrum and convert it into a three-axis drive signal.

[0052] Step S44: Start the triaxial electromagnetic seismic simulation subsystem and apply horizontal and vertical vibrations through the horizontal magnetic drive end and the vertical magnetic drive end, respectively; the vibrating plate reproduces the seismic waveform along the preset degrees of freedom under the guidance of the elastic roller guide.

[0053] In step S5, the central control unit triggers the triaxial electromagnetic seismic simulation subsystem, wave generation and rectification system, downstream water circulation and treatment system, non-contact optical measurement equipment, and contact sensors to simultaneously acquire data. The acquired data includes multi-dimensional data such as acceleration, displacement, pore water pressure, flow velocity, turbidity, wave height, and bed topography. High-speed photography and high-definition video data are recorded simultaneously for subsequent full-field motion analysis and morphological reconstruction.

[0054] In step S6, the central control unit analyzes the sensor data in real time and uses AI algorithms to perform seismic vibration denoising on the original measured values. Based on the denoised data, it determines whether the preset test termination conditions are met. If the test termination conditions are met, the process proceeds to step S7; otherwise, the denoising process continues. Specifically, the data denoising includes frequency domain filtering of the original flow velocity signal measured by the ADV flow meter, combined with the high-precision seismic station vibration data output in step S5, to remove periodic disturbances caused by the earthquake and obtain a purer water flow velocity value; and filtering out the pressure fluctuation component directly caused by seismic vibration in the data collected by the pore water pressure gauge array.

[0055] Preferably, the conditions for determining whether the test should be terminated provided in this application include: the rate of change of the bed surface (calculated by the laser ranging matrix 32) being less than 0.1 mm / min for 5 consecutive minutes, indicating that the scouring is stabilizing; the pore water pressure ratio after noise reduction (i.e., the ratio of pore pressure to overlying effective stress) continuously exceeding 0.95, indicating that sand bed liquefaction has occurred; the scouring depth at key points reaches a preset threshold, or its rate of change tends to stabilize; and the sediment transport flux (combined with turbidity meter 1 and image analysis) reaches dynamic equilibrium.

[0056] Step S7: The AI ​​image processing unit processes and intelligently analyzes the data determined in step S6. Specifically, the AI ​​image processing unit performs particle image velocimetry and digital image correlation analysis on the collected image data to extract the overall flow velocity, displacement, and strain field. Combined with sensor data, multi-physics field coupling analysis of sediment transport, soil liquefaction, and structural response is performed. Based on deep learning algorithms, the bed surface evolution is reconstructed in three dimensions and intelligently identified to generate real-time three-dimensional landforms and display flow field characteristics.

[0057] Preferably, AI algorithms are used, combined with actual vibration data output from high-precision seismic stations as a "noise template," to intelligently identify and filter periodic interference signals caused by seismic vibrations to surrounding sensing equipment (such as ADV flow meters, pore water pressure gauges, and optical measurement equipment) during the experiment. This significantly improves the signal-to-noise ratio and accuracy of multi-dimensional measurement data under strong vibration environments. Finally, based on deep learning algorithms, the three-dimensional dynamic reconstruction of the scour pit is automatically completed, generating a full-field digital model that integrates topographic, flow field, and mechanical characteristics in real time. The central control unit transmits the model and data to the mobile device in real time, intelligently analyzing the remaining experimental time, allowing experimenters to promptly grasp experimental information and improve experimental decision-making efficiency.

[0058] Step S8: End the test, shut down the triaxial electromagnetic seismic simulation subsystem, wave generation and rectification system, downstream water circulation and treatment system, non-contact optical measurement equipment and contact sensors, start the sand pump to remove the silt accumulated in the sedimentation tank, and ensure long-term stable operation of the system.

[0059] In summary, the circulating water tank seismic simulation test system and method based on electromagnetic drive and artificial intelligence simulation provided in this application are the first to integrate the circulating water tank body, wave generation system, and three-axis electromagnetic seismic station, enabling X / Y / Z three-axis vibration with independent control of each axis. This improves the physical simulation of the "wave-current-earthquake" coupled environment, filling a gap in experimental equipment in this research field. By combining contact sensors (acceleration, displacement, flow velocity, liquid level) and non-contact optical measurements (high-speed photography, high-definition video recording), a similarity criterion library is constructed. This minimizes the error between the actual and experimental environments according to different experimental purposes, improving the model simulation accuracy and simulating the complex conditions of structures simultaneously subjected to waves, currents, and seismic forces in a real marine environment, making the experimental results more valuable. One-click process control is achieved through a central control unit, and AI technology is used for waveform generation and data processing, greatly improving experimental efficiency and research depth.

[0060] Those skilled in the art will understand that, unless otherwise defined, all terms used herein (including technical and scientific terms) have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains. It should also be understood that terms such as those defined in general dictionaries should be understood to have the same meaning as in the context of the prior art, and should not be interpreted in an idealized or overly formal sense unless defined as herein.

[0061] The meaning of "and / or" as used in this application includes situations where each exists alone or both exist simultaneously.

[0062] The term "connection" as used in this application can mean a direct connection between components or an indirect connection between components through other components.

[0063] Based on the above-described preferred embodiments of the present invention, and through the foregoing description, those skilled in the art can make various changes and modifications without departing from the inventive concept. The technical scope of this invention is not limited to the contents of the specification, but must be determined according to the scope of the claims.

Claims

1. A circulating water tank seismic simulation test system based on electromagnetic drive and artificial intelligence simulation, comprising a circulating water tank body, characterized in that: The main body of the circulating water tank includes an upstream water inlet section, a test section and a downstream water storage section connected in sequence. The test section serves as an observation and operation area, and a stainless steel sandbox for testing is embedded at its bottom. A triaxial electromagnetic seismic simulation subsystem is installed at the bottom of the test section. It includes a shaking table, an electromagnetic shaking table support column, and an electromagnetic drive device. A stainless steel sand box is placed on the shaking table, which is supported by the electromagnetic shaking table support column. At the same time, the electromagnetic shaking table support column provides a position reference for the electromagnetic drive device. That is, the electromagnetic shaking table support column forms a space for installing the electromagnetic drive device below the test section. The electromagnetic drive device provides translational vibration force in the X, Y, and Z axis directions to the stainless steel sand box. A wave-generating and rectification system is installed at the beginning of the upstream intake section to simulate waves and set the incoming flow conditions. A downstream water circulation and treatment system, including a water pump, is installed at the downstream water storage section. The system pumps the water that has been collected in the downstream water storage section after the experiment back to the upstream water intake section, forming a water circulation channel. Non-contact optical measurement equipment is installed at the opening of the main body of the circulating water tank. Several contact sensors are also installed inside the circulating water tank seismic simulation test system. The triaxial electromagnetic seismic simulation subsystem, wave generation and rectification system, downstream water circulation and treatment system, non-contact optical measurement equipment, and contact sensors are all connected to the intelligent control and data processing center to realize distributed management and control of the entire circulating water tank seismic simulation test system.

2. The circulating water tank seismic simulation test system based on electromagnetic drive and artificial intelligence simulation according to claim 1, characterized in that: The wave-generating and rectification system of the upstream intake section includes a wave generator and a rectification device. The wave generator is located close to the test section. The wave generator adopts a pusher type, rocker type, or piston type according to the wave sequence of the sea state to be simulated. The rectification device adopts a multi-layer honeycomb plate, a damping net, or a guide plate. The downstream water circulation and treatment system of the downstream water storage section includes a water tank, a sedimentation tank, a water pump, a sand pump, and a water injection pipeline. A sedimentation tank is set at the bottom of the water tank on the side closest to the test section. The water pump is installed on a pipeline connected to the water tank and this pipeline is connected to the upstream water inlet section. The sand pump is installed on a pipeline connected to the sedimentation tank. One end of the water injection pipeline is connected to an external water source, and the other end is connected to the water tank.

3. The circulating water tank seismic simulation test system based on electromagnetic drive and artificial intelligence simulation according to claim 1, characterized in that: A vibrating support plate is horizontally installed on the top of the vibration table, and the surface of the vibrating support plate supports the stainless steel sand box. The electromagnetic drive device includes a horizontal magnetic drive end, a horizontal magnetic receiver end, a vertical magnetic drive end, and a vertical magnetic receiver end. Several vertical magnetic drive ends are vertically installed on the base of the electromagnetic vibration table support column located below the vibration plate, and vertical magnetic receiver ends are installed on the bottom surface of the vibration plate opposite to the vertical magnetic drive ends; several horizontal magnetic drive ends are vertically installed on the lateral mounting surface of the electromagnetic vibration table support column relative to the vibration plate, and horizontal magnetic receiver ends are installed on the side wall of the vibration plate opposite to the horizontal magnetic drive ends.

4. The circulating water tank seismic simulation test system based on electromagnetic drive and artificial intelligence simulation according to claim 3, characterized in that: Several centroid adjustment blocks are set at the four corners of the vibration bearing plate to finely adjust the centroid position of the circulating water tank seismic simulation test system. Elastic roller guides are installed at the vibration table and the vibration plate support to ensure that the vibration table moves along the preset degrees of freedom; The drive structure of the vibration table and the space formed by the electromagnetic vibration table support column for installing the electromagnetic drive device are covered with high conductivity or high magnetic permeability material to suppress the outward radiation of strong electromagnetic fields.

5. The circulating water tank seismic simulation test system based on electromagnetic drive and artificial intelligence simulation according to claim 1, characterized in that: The top opening of the stainless steel sandbox is flush with the bottom plate of the upstream water inlet section and the downstream water storage section, and a sandbox drainage outlet is opened at the bottom of the stainless steel sandbox; a flexible sealing rubber skirt is set at the connection between the bottom of the test section and the bottom of the stainless steel sandbox.

6. The circulating water tank seismic simulation test system based on electromagnetic drive and artificial intelligence simulation according to claim 1, characterized in that: The contact sensors include an accelerometer, an optical displacement sensor, a magnetic field sensor, an ADV flow meter, a turbidity meter, a laser ranging matrix, a capacitive wave height meter, and a pore water pressure gauge array. The acceleration sensor is installed on the bottom surface of the vibration plate and is used to measure the acceleration response of the vibration table and the stainless steel sandbox in three directions. Several optical displacement sensors are installed on the base of the electromagnetic vibration table support column and the bottom surface of the vibration plate to measure the displacement of the vibration table in three directions. The magnetic field sensor is placed close to the electromagnetic drive device to monitor the magnetic field strength around the electromagnetic drive device. The ADV flow meter, turbidity meter, and capacitive wave height meter extend into the main body of the circulating water tank. The ADV flow meter measures the three-dimensional flow velocity distribution at different locations in the water tank and sedimentation tank in real time. The turbidity meter monitors the real-time changes in sediment concentration in the main body of the circulating water tank. The capacitive wave height meter measures water surface fluctuations. The laser ranging matrix includes several laser displacement sensors, which are arranged in a regular grid pattern and mounted on the opening end of the main body of the circulating water tank; The pore water pressure gauge array consists of several pore water pressure gauges, all of which are buried in a stainless steel sand box to measure the distribution and changes of pore water pressure in sand. The non-contact optical measurement equipment includes a high-speed camera and a high-definition video recorder, both of which are mounted at the opening of the main body of the monitoring circulating water tank. The high-speed camera is used to capture the subtle processes of sand movement, water turbulence, or high-speed deformation of the structural model, while the high-definition video recorder is used to record the dynamics of the entire test section.

7. The circulating water tank seismic simulation test system based on electromagnetic drive and artificial intelligence simulation according to claim 1, characterized in that: The intelligent control and data processing center includes a central control unit, an intelligent waveform generation module, and an AI image processing unit. The central control unit uniformly schedules the start-up, shutdown, parameter setting, and operation synchronization of the triaxial electromagnetic seismic simulation subsystem, wave generation and rectification system, downstream water circulation and treatment system, non-contact optical measurement equipment, and contact sensors. The intelligent waveform generation module embeds machine learning algorithms to learn and simulate the characteristics of real seismic wave spectra and generate multi-directional driving signals that meet experimental requirements. The AI ​​image processing unit processes high-speed photography and video recording data in real time, extracting the full-field displacement, strain field, and velocity field distribution.

8. The test method for the circulating water tank seismic simulation test system based on electromagnetic drive and artificial intelligence simulation according to claim 1, characterized in that: Includes the following steps: Step S1, test preparation: Lay sand samples in a stainless steel sand box according to the test design requirements, and inject test water into the water tank through the water injection pipe until the water level reaches the preset height. Step S2: Initialize the system and check the connection status and normal operation of the main body of the circulating water tank, the triaxial electromagnetic seismic simulation subsystem, the wave generation and rectification system, the downstream water circulation and treatment system, the non-contact optical measurement equipment, and the contact sensors. Saturation treatment was performed through the sand box drainage outlet, and the sand sample was allowed to stand until it reached a stable state. Calibrate accelerometers, pore water pressure gauge arrays, ADV flow meters, laser ranging matrices, high-speed cameras, and high-definition video recorders; The intelligent control and data processing center sets the test stop conditions; Step S3: Select the dominant similarity criterion based on the experimental research object, and set the initial test parameters of the circulating water flume seismic simulation test system according to the selected dominant similarity criterion; Step S4: Based on the dominant similarity criterion determined in step S3, the actual environmental parameters are converted to the experimental environmental parameters; after the experimental parameter conversion is completed, wave-current-seismic coupling simulation is performed. In step S5, the central control unit triggers the triaxial electromagnetic seismic simulation subsystem, wave generation and rectification system, downstream water circulation and treatment system, non-contact optical measurement equipment and contact sensors to acquire data simultaneously. Step S6: The central control unit analyzes the sensor data in real time and uses AI algorithm to perform seismic vibration denoising on the original measurement values. Based on the denoised data, it determines whether the preset test termination condition is met. If the test termination condition is met, proceed to step S7; otherwise, continue the denoising process. Specifically, data denoising involves combining the raw flow velocity signal measured by the ADV flow meter with the high-precision seismic station vibration data output in step S5, performing frequency domain filtering to remove periodic disturbances caused by earthquakes and obtain a purer water flow velocity value. Filter out the pressure fluctuation component directly caused by seismic vibration in the data collected by the pore water pressure gauge array; Step S7: The AI ​​image processing unit processes and intelligently analyzes the data determined in step S6. Specifically, The AI ​​image processing unit is used to perform particle image velocimetry and digital image correlation analysis on the acquired image data to extract the overall flow velocity, displacement and strain field. Multi-physics coupling analysis of sediment transport, soil liquefaction, and structural response was conducted by combining sensor data. Based on deep learning algorithms, the evolution of the bed surface is reconstructed in three dimensions and intelligently identified to generate real-time three-dimensional landforms and display flow field characteristics. Step S8: End the test, shut down the triaxial electromagnetic seismic simulation subsystem, wave generation and rectification system, downstream water circulation and treatment system, non-contact optical measurement equipment and contact sensors, and start the sand pump to remove the silt accumulated in the sedimentation tank.

9. The test method for the circulating water tank seismic simulation test system based on electromagnetic drive and artificial intelligence simulation according to claim 8, characterized in that: In step S3, based on the experimental research object, the dominant similarity criterion is selected and the initial experimental parameters are set, specifically: If the research object is the seismic response of a structure, priority should be given to ensuring the geometric similarity and seismic dynamic similarity of the structure, and the structural model size and seismic wave input spectrum should be determined first. If the research object is sediment erosion, priority should be given to ensuring that the initiation of sediment particles is similar to that of water flow. First, the sediment particle size and water flow velocity in the model should be determined. If the research object is seabed liquefaction, priority should be given to ensuring the similarity of soil dynamics and pore water pressure response, and the seismic vibration intensity and saturated sand parameters should be determined first. After determining the dominant similarity criterion and initial experimental parameters, the impact of other secondary similarity criteria on the experimental results is further evaluated. The similarity criterion with the least impact on experimental distortion is selected for approximation, and finally all experimental parameters are determined.

10. The test method for the circulating water tank seismic simulation test system based on electromagnetic drive and artificial intelligence simulation according to claim 9, characterized in that: Step S4, wave-current-seismic coupling simulation, includes: Step S41: If the research object is the seismic response of a structure, the actual seismic spectrum is converted into the experimental seismic spectrum firstly based on the seismic dynamic similarity criterion; then, based on the geometric similarity and water flow similarity criteria, and considering the influence of the seismic dynamic similarity criterion on the hydrodynamic similarity, the actual wave parameters and water flow parameters are converted into experimental values. If the research object is sediment erosion, the actual flow parameters and wave parameters are converted into experimental values ​​first based on the similarity criteria of sediment initiation and water flow motion. Then, based on the seismic dynamic similarity criterion, and taking into account the influence of the similarity criteria of sediment initiation and water flow motion, the actual seismic spectrum is converted into the experimental seismic spectrum. If the research object is seabed liquefaction, the seismic vibration intensity and saturated sand parameters are determined first based on the soil dynamic similarity and pore water pressure response similarity criteria. Then, the hydrodynamic parameter conversion is completed based on the sediment initiation similarity and water flow movement similarity criteria, while also considering the influence of the soil dynamic similarity and pore water pressure response similarity criteria. Step S42: Start the water pump and wave generator, set the converted wave parameters and water flow parameters through the central control unit, and use the rectifier to homogenize the incoming flow. Monitor and adjust the wave parameters and water flow parameters in real time until a stable hydraulic environment that meets the test requirements is formed. Step S43: The central control unit continues to set the converted test seismic spectrum and convert it into a three-axis drive signal; Step S44: Start the triaxial electromagnetic seismic simulation subsystem and apply horizontal and vertical vibrations through the horizontal magnetic drive end and the vertical magnetic drive end, respectively; the vibrating plate reproduces the seismic waveform along the preset degrees of freedom under the guidance of the elastic roller guide.

Images