Construction control method of anti-scour device for seabed data center

Abstract

The invention provides a construction control method of an anti-scour device for a seabed data center, and belongs to the technical field of construction control, and the method comprises the steps: 1, carrying out the data collection and preprocessing of the anti-scour device for the seabed data center; 2, constructing a prediction model for the anti-scour device used for the seabed data center; 3, executing a predictive adjustment algorithm according to the predicted ocean current flow velocity and the cumulative scouring depth; and step 4, determining the thickness of an anti-scour plate and the density of bionic aquatic plants of the anti-scour device based on the maximum flow velocity in the ocean current flow velocities output by the prediction model. By constructing the multi-dimensional data fusion model and the prediction adjustment algorithm, predictive control over the attitude of the anti-scouring device is achieved, meanwhile, parameter design in the construction process of the seabed data center pile foundation anti-scouring device is optimized, the adaptability of the anti-scouring device to the long-term and dynamic marine environment is improved, and the anti-scouring effect is improved. And long-term safety and stability of the seabed data center pile foundation are guaranteed.

Description

Technical Field

[0001] This invention belongs to the field of construction control technology, and specifically relates to a construction control method for an anti-scour device for a submarine data center. Background Technology

[0002] Submarine data centers, leveraging the natural cooling effect of seawater, possess significant advantages in resource utilization efficiency and cost control. However, their pile foundations face the severe challenge of long-term ocean current scouring. Existing anti-scouring devices for submarine data center pile foundations, such as the one mentioned in patent publication number "CN119163078A," while mitigating immediate scouring to some extent by using water flow sensors and topographic scanners to collect data in real time and dynamically adjusting the device's tilt and rotation angles through a control system, have significant technical shortcomings:

[0003] The existing control system can only make passive, responsive adjustments based on real-time data, and cannot integrate long-term information such as historical scour data and seasonal ocean current patterns to make predictive adjustments. In the actual marine environment, ocean current velocity and direction exhibit significant seasonal fluctuations. For example, temperate seas often experience low-velocity monsoon currents in summer, while high-velocity cold wave currents accompany them in winter. At the same time, seabed scour and deposition processes have cumulative characteristics. The formation of historical scour pits can change the local flow field distribution, leading to an increased risk of subsequent scour. Due to the lack of predictive capabilities, the existing control methods cannot adjust the device's attitude in advance to cope with upcoming extreme ocean currents or cumulative scour and deposition risks. Adjustments are often made only after the scour has intensified, at which point the pile foundation has already been subjected to additional scour loads. Over the long term, this can easily lead to problems such as decreased pile foundation support capacity and structural fatigue damage, seriously threatening the safe and stable operation of the subsea data center.

[0004] Furthermore, in the construction and control process of existing scour protection devices, the device parameters (such as the tilt angle of the scour protection plate and the density of biomimetic aquatic plants) are mostly based on empirical design, without establishing a quantitative correlation model with long-term marine environmental parameters. This results in insufficient adaptability of the device in different sea areas and seasons, further reducing the stability and sustainability of the scour protection effect. Therefore, developing a construction and control method for scour protection devices that can integrate multi-dimensional data and achieve predictive adjustment has become a key technical requirement for ensuring the long-term safe operation of seabed data centers. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a construction control method for scour protection devices used in subsea data centers. This method overcomes the limitations of existing methods, which only passively respond to real-time data and lack predictive adjustment capabilities. By constructing a multi-dimensional data fusion model and predictive adjustment algorithms, this method achieves predictive control of the scour protection device's attitude. Simultaneously, it optimizes parameter design during the construction of the scour protection device for subsea data center pile foundations, enhancing the device's adaptability to long-term, dynamic marine environments and ensuring the long-term safety and stability of the subsea data center pile foundations.

[0006] The present invention employs the following technical solution.

[0007] A method for controlling the construction of an anti-scour device for a submarine data center includes:

[0008] Step 1: Data acquisition and preprocessing of the anti-scour device used for the seabed data center;

[0009] Step 2: Build a predictive model for scour protection devices used in subsea data centers;

[0010] Step 3: Execute a predictive regulation algorithm based on the predicted ocean current velocity and cumulative scour depth;

[0011] Step 4: Based on the maximum current velocity output by the prediction model, determine the thickness of the anti-scour plate and the density of the bionic aquatic plants in the anti-scour device.

[0012] Furthermore, step 1 specifically includes:

[0013] Step 1-1: Collect three types of core data from the anti-scour device used for the seabed data center;

[0014] Step 1-2: Preprocess the collected data.

[0015] Furthermore, in step 1-1, the three types of core data include:

[0016] Real-time monitoring data includes information obtained from water flow sensors connected to the control system regarding the tidal currents in the sea area where the subsea data center pile foundation scour protection device is located. Real-time flow rate at any given moment Furthermore, the data is transmitted to the control system, where it is acquired by the terrain scanner connected to the control system. Real-time flushing depth And the data is transmitted to the control system;

[0017] Historical data, including the maximum seasonal ocean current velocity in the sea area where the subsea data center pile foundation scour protection device is located, over the past 5 to 10 years and recorded in the control system. And the seasonal average scour depth of the sea area where the anti-scour device for the pile foundation of the submarine data center is located over the past 5 to 10 years. ;

[0018] Environmental parameters, including the particle size of marine sediment in the sea area where the anti-scouring device for the subsea data center pile foundation is located, are entered into the control system. Seawater density The depth of the seawater where the pile foundation is located .

[0019] Furthermore, in step 1-1, the terrain scanner acquires... Real-time flushing depth The methods specifically include:

[0020] S1: Establish a baseline terrain model;

[0021] S2: Perform real-time terrain data acquisition and matching, and determine high-risk and low-risk sea areas;

[0022] S3: Calculate the real-time scour depth.

[0023] Furthermore, S1 specifically includes:

[0024] Before installing the anti-scouring device, or within 1-2 weeks after installation when the seabed is in a stable state, benchmark topographic data acquisition and modeling should be completed. Benchmark topographic data acquisition and modeling includes data acquisition, elevation calibration, and benchmark modeling to obtain relative elevation data. ;

[0025] Furthermore, S2 specifically includes:

[0026] During the operation of the anti-erosion device, a terrain scanner is used to collect real-time terrain data at a set frequency:

[0027] Real-time scanning: Activate the fixed-deployment terrain scanner and collect the current seabed relative elevation data using the same scanning density and elevation calibration method as the baseline modeling. ;

[0028] Spatial matching: using the GPS positioning module built into the terrain scanner and the center coordinates of the pile foundation ( , This involves spatially aligning real-time terrain data with a baseline terrain model to ensure that scan points at the same physical location have consistent coordinates in both the real-time terrain data and the baseline terrain model.

[0029] Furthermore, S3 specifically includes:

[0030] Based on relative elevation data Relative elevation data 2 The real-time scour depth is calculated using the elevation difference method, and the specific formula is as follows:

[0031] For any spatial coordinates ( , The scan points are used to flush depth features in real time. The calculation formula is:

[0032] = - ;

[0033] like >0: This indicates that the seabed elevation at this scan point is lower than the reference level, indicating erosion. The value is the scour depth;

[0034] like ≤0: This indicates that the seabed elevation at this scanning point is higher than the reference surface, and the scour depth is recorded as 0;

[0035] Next The average value of the scour depth obtained at each time step is taken as .

[0036] Furthermore, in S2, the method for determining high-risk and low-risk sea areas is as follows:

[0037] First, four core indicators affecting seabed erosion were selected. The weights of each indicator were determined using the Analytic Hierarchy Process (AHP). The four core indicators are as follows: , , and , , , and The weights of the indicators are 0.4, 0.25, 0.2 and 0.15, respectively;

[0038] The total risk score is calculated based on four core indicators and their weights. The calculation formula is as follows:

[0039] ;

[0040] in , , and They are , , and The normalized value;

[0041] if Therefore, the sea area where the anti-scouring device for the pile foundation of the submarine data center is located is determined to be a low-risk sea area.

[0042] if Therefore, the area where the anti-scouring device for the pile foundation of the submarine data center is located is determined to be a high-risk sea area.

[0043] Furthermore, steps 1-2 specifically include:

[0044] The moving average method is used to remove high-frequency noise from real-time monitoring data, and linear interpolation is used to fill in missing values ​​in historical data.

[0045] Furthermore, step 2 specifically includes:

[0046] Step 2-1: Construct a seasonal ocean current prediction model;

[0047] Step 2-2: Construct a cumulative scour depth prediction model.

[0048] Furthermore, step 2-1 specifically includes:

[0049] Based on historical seasonal ocean current data of the sea area where the subsea data center pile foundation scour protection device is located, an improved time-series ARIMA model is used to predict the ocean current velocity for the next 1-7 days. The improved time series ARIMA model formula is as follows:

[0050] ;

[0051] Where k is the prediction step size, 1≤k≤168 hours. For the set constant term, The set autoregressive coefficients, The set moving average coefficient, Here, p and q are the values ​​determined by the AIC information criteria, representing the initial error term. exist Real-time flow rate at any given moment.

[0052] Furthermore, step 2-2 specifically includes:

[0053] Predicted cumulative scour depth over the next 1-3 months The calculation formula is as follows:

[0054] ;

[0055] Where m is the predicted number of months, 1≤m≤3, and g is the gravitational acceleration. H1 is the set drag coefficient, H1 is the height of the anti-scour plate of the anti-scour device for the pile foundation of the submarine data center, and Δt is the time step.

[0056] Furthermore, step 3 specifically includes:

[0057] Step 3-1: Based on predicted ocean current velocity With cumulative scour depth Construct the objective function for attitude adjustment of the anti-scouring device. :

[0058] ;

[0059] in, For the MIN function, 1. The set weighting coefficients, 1+ =1, The optimal oncoming flow velocity for the anti-impact plate is set. The set safe cumulative scour depth, This represents the highest historical ocean current velocity in the sea area where the anti-scour device for the subsea data center's pile foundation is located. The maximum allowable scour depth for the pile foundation;

[0060] Step 3-2: Solve the objective function for attitude adjustment of the anti-scour device using the gradient descent method to obtain the optimal tilt angle. With optimal rotation angle The formula for calculating gradient descent is:

[0061] ;

[0062] in and Each is the current The tilt and rotation angles of the anti-erosion device are monitored at all times. These angles are acquired and transmitted to the control system via a tilt sensor and a rotary potentiometer connected to the control system, respectively. The set learning rate;

[0063] Step 3-3: The control system drives the hydraulic rod and ball bearing to adjust the tilt angle of the anti-erosion device. The rotation angle with the anti-scouring device is .

[0064] Furthermore, step 4 specifically includes:

[0065] During the segmented prefabrication stage of the anti-scouring device, the ocean current velocity output based on the prediction model was used. Maximum flow rate in Determine the thickness h of the anti-erosion plate of the anti-erosion device and the density of the bionic aquatic plants. :

[0066] ;

[0067] in, Let σ be the MAX function, where L is the span of the impact shield and σ is the tensile strength of the impact shield material. K is the set minimum effective flow rate threshold; K is the set conversion factor.

[0068] The beneficial effects of the present invention are as follows, compared with the prior art:

[0069] Predictive adjustment effect: Compared with the existing passive response control, this method can cope with high-velocity ocean currents 48 hours in advance, reducing the scour load on the pile foundation of the submarine data center by 35%-45% and the cumulative scour depth by 25%-30%, effectively avoiding the passive situation of "adjusting after scour intensifies".

[0070] Improved Adaptability: The thickness of the anti-scour plate and the density of biomimetic aquatic plants, optimized based on seasonal forecast data, have reduced the fluctuation range of the anti-scour efficiency of the subsea data center pile foundation scour protection device from ±20% to ±8% in different seasons, significantly improving its adaptability to the dynamic marine environment.

[0071] Long-term stability assurance: By integrating historical and real-time data, the predictive model can provide early warning of cumulative scour risks up to 3 months in advance. Combined with parameter optimization, the design service life of the subsea data center pile foundation is extended from 20 years to 25-30 years, reducing the operation and maintenance costs of the subsea data center.

[0072] Optimized response efficiency: The predictive adjustment algorithm of the control system has a response time of ≤30 seconds, which is more than 60% faster than the existing control system (1-2 minutes), ensuring rapid attitude adjustment in the event of sudden changes in ocean currents and further enhancing the reliability of protection. Attached Figure Description

[0073] Figure 1 This is a flowchart of the construction control method for the anti-scouring device used in the submarine data center in this invention. Detailed Implementation

[0074] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of this invention. The embodiments described in this application are merely some embodiments of this invention, and not all embodiments. Based on the spirit of this invention, other embodiments obtained by those skilled in the art without creative effort are all within the protection scope of this invention.

[0075] like Figure 1 As shown, a construction control method for an anti-scour device for a submarine data center includes:

[0076] Step 1: Data acquisition and preprocessing for the scour protection device used in the submarine data center; the scour protection device used in the submarine data center is the scour protection device for the pile foundation of the submarine data center.

[0077] In a preferred but non-limiting embodiment of the present invention, step 1 specifically includes:

[0078] Step 1-1: Collect three types of core data from the anti-scour device used for the seabed data center;

[0079] In a preferred but non-limiting embodiment of the present invention, in step 1-1, the three types of core data include:

[0080] Real-time monitoring data includes data obtained from water flow sensors connected to the control system of the subsea data center pile foundation scour protection device, which acquires data on the tidal current in the sea area where the device is located. Real-time flow rate at any given moment Furthermore, the data is transmitted to the control system, where it is acquired by the terrain scanner connected to the control system. Real-time flushing depth And the data is transmitted to the control system;

[0081] Historical data, including the maximum seasonal ocean current velocity in the sea area where the subsea data center pile foundation scour protection device is located, over the past 5 to 10 years and recorded in the control system. And the seasonal average scour depth of the sea area where the anti-scour device for the pile foundation of the submarine data center is located over the past 5 to 10 years. ;

[0082] Environmental parameters, including the particle size of marine sediment in the sea area where the anti-scouring device for the subsea data center pile foundation is located, are entered into the control system. Seawater density The depth of the seawater where the pile foundation is located .

[0083] In a preferred but non-limiting embodiment of the present invention, in step 1-1, the terrain scanner acquires... Real-time flushing depth The methods specifically include:

[0084] I. Methodology and Principles

[0085] The core principle of topographic scanners for obtaining real-time scour depth is to acquire three-dimensional topographic data of the seabed around the pile foundation using high-precision distance measurement technologies (such as multibeam echo sounding, side-scan sonar, or laser sounding). Combined with benchmark topographic calibration and dynamic difference calculation, the elevation change of the seabed surface relative to the initial stable state is determined, and then the real-time scour depth near the pile foundation (i.e., the vertical depth of the depression area formed by the seabed after scour) relative to the benchmark surface is extracted.

[0086] This method requires two key prerequisites: first, a "benchmark topographic model" must be established in advance to represent the initial stability of the seabed after the installation of the pile foundation, serving as a reference for subsequent scour depth calculation; second, "dynamic topographic data" of the current seabed must be obtained through periodic or real-time scanning, and the scope of the scour pit and its depth can be accurately located and calculated through spatial matching and elevation difference analysis of the two.

[0087] II. Selection and Deployment of Core Equipment

[0088] (a) Equipment selection

[0089] Considering the unique characteristics of the seabed environment (low visibility, high pressure, and sediment interference), high-precision terrain scanning equipment adapted to underwater working conditions should be prioritized to form a terrain scanner.

[0090] A topographic scanner consisting of a multibeam echo sounder and an underwater laser sounder is deployed. The multibeam echo sounder is responsible for acquiring topographic data over a large area of ​​10-50m around the pile foundation, while the underwater laser sounder focuses on the near-field area of ​​0.5-5m at the bottom of the pile foundation (where the seabed meets the pile), compensating for the measurement blind spots of the multibeam echo sounder in the near-field area and ensuring the integrity of the scour depth data.

[0091] (ii) Equipment deployment location

[0092] Fixed deployment: On the support beam (or support frame of the anti-scouring device) of the subsea data center pile foundation, install 3-4 topographic scanners (such as a topographic scanner formed by 2 multibeam echo sounders and 2 laser sounders) evenly in the circumference. The probe of the topographic scanner should face the seabed surface of the sea area where the subsea data center pile foundation is located, and the vertical distance between the scanner and the seabed should be controlled at 1-3m (to avoid silt and sand obstruction, and at the same time ensure measurement accuracy).

[0093] Dynamic deployment: Equip transport ships or underwater robots (ROVs) with portable side-scan sonar to conduct dynamic scanning of a 50-100m radius around the pile foundation every 1-2 weeks to verify the accuracy of data from fixed equipment and monitor the trend of seabed erosion and siltation over a large area.

[0094] III. Specific Implementation Steps

[0095] S1: Establish a baseline terrain model (before installation / initial stabilization);

[0096] In a preferred but non-limiting embodiment of the present invention, S1 specifically includes:

[0097] Before installing the anti-scouring device, or within 1-2 weeks after installation when the seabed is in a stable state (at which point the initial disturbance of the seabed by the ocean currents has subsided), complete the baseline topographic data acquisition and modeling. Baseline topographic data acquisition and modeling includes data acquisition, elevation calibration, and baseline modeling to obtain relative elevation data. The details are as follows:

[0098] Data acquisition: The seabed within a range of 0-50m around the pile foundation was fully scanned using a multibeam echo sounder and laser depth sounder of a fixed-deployment topographic scanner. The scanning density was set as follows: one data point every 0.1m within 5m of the pile foundation and one data point every 0.5m in the range of 5-50m.

[0099] Elevation calibration: The elevation of the lower surface of the pile foundation support beam is used as the reference (denoted as ). ), and the elevation data of all scanned points ( , which indicates the first Elevation data of each scan point is converted into relative elevation data relative to the lower surface of the support beam. , Indicates the first The relative elevation of each scanning point (1) ensures consistency in subsequent measurements;

[0100] Benchmark modeling: Using ArcGIS or Surfer software, the preprocessed relative elevation data is fitted into a "benchmark topographic model of the seabed around the pile foundation" (raster model, resolution 0.1-0.5m), and the center coordinates of the pile foundation are marked. , This serves as a reference benchmark for subsequent scour depth calculations.

[0101] S2: Perform real-time terrain data acquisition and matching, and determine high-risk and low-risk sea areas;

[0102] In a preferred but non-limiting embodiment of the present invention, S2 specifically includes:

[0103] During the operation of the anti-scouring device, real-time topographic data is collected using a terrain scanner at a set frequency (which can be adjusted according to the scouring risk of the sea area: once every 2 hours for high-risk sea areas and once every 6 hours for low-risk sea areas):

[0104] Real-time scanning: Activate the fixed-deployment terrain scanner and collect the current seabed relative elevation data using the same scanning density and elevation calibration method as the baseline modeling. ( Indicates the first Relative elevation of each scanning point (II)

[0105] Spatial matching: The GPS positioning module (accuracy ±0.5m) of the terrain scanner is used to match the center coordinates of the pile foundation. , The real-time terrain data is spatially aligned with the baseline terrain model to ensure that the coordinates of the scan points at the same physical location are consistent in the real-time terrain data and the baseline terrain model (error ≤ 0.1m).

[0106] S3: Calculate the real-time scour depth.

[0107] In a preferred but non-limiting embodiment of the present invention, S3 specifically includes:

[0108] Based on relative elevation data Relative elevation data 2 The real-time scour depth is calculated using the elevation difference method. The specific formula and logic are as follows:

[0109] Single-point scour depth calculation:

[0110] For any spatial coordinates ( , The scan points are used to flush depth features in real time. The calculation formula is:

[0111] = - ;

[0112] like >0: This indicates that the seabed elevation at this scan point is lower than the reference level, indicating erosion. The value is the scour depth;

[0113] like ≤0: This indicates that the seabed elevation at this scanning point is higher than the reference surface (possibly due to siltation), and the scour depth is recorded as 0;

[0114] Next The average value of the scour depth obtained at each time step is taken as .

[0115] Using the methods described above, topographic scanners can achieve high-precision, real-time monitoring of the scour depth around the pile foundations of seabed data centers. Specific application effects are as follows:

[0116] Data accuracy: Single-point scour depth measurement error ≤ 0.05m, regional average scour depth error ≤ 0.03m, meeting the engineering accuracy requirements for pile foundation scour risk assessment;

[0117] Response speed: The total time from data acquisition and calculation to transmission to the control system is ≤30 seconds, which can support real-time attitude adjustment of the anti-scouring device or provide high-frequency data input for predictive adjustment;

[0118] Stability: Under conditions of sediment concentration ≤50mg / L and ocean current velocity ≤3m / s, the equipment has a continuous operation failure rate of ≤1% and can output scour depth data stably for a long time, providing reliable data support for the safety protection of the pile foundation of the seabed data center.

[0119] In a preferred but non-limiting embodiment of the present invention, in S2, the method for determining high-risk and low-risk sea areas is as follows:

[0120] First, four core indicators affecting seabed erosion were selected. The weights of each indicator were determined using the Analytic Hierarchy Process (AHP). The four core indicators are as follows: , , and , , , and The weights of the indicators are 0.4, 0.25, 0.2 and 0.15, respectively;

[0121] The total risk score is calculated based on four core indicators and their weights. The calculation formula is as follows:

[0122] ;

[0123] in , , and They are , , and The normalized value;

[0124] if Therefore, the sea area where the anti-scouring device for the pile foundation of the submarine data center is located is determined to be a low-risk sea area.

[0125] if Therefore, the area where the anti-scouring device for the pile foundation of the submarine data center is located is determined to be a high-risk sea area.

[0126] Step 1-2: Preprocess the collected data.

[0127] In a preferred but non-limiting embodiment of the present invention, steps 1-2 specifically include:

[0128] High-frequency noise in real-time monitoring data is removed using a moving average method, and missing values ​​in historical data are filled in using linear interpolation to ensure data integrity. The real-time monitoring data and historical data mentioned in the following steps are their respective preprocessed data.

[0129] Step 2: Build a predictive model for scour protection devices used in subsea data centers;

[0130] In a preferred but non-limiting embodiment of the present invention, step 2 specifically includes:

[0131] Step 2-1: Construct a seasonal ocean current prediction model;

[0132] In a preferred but non-limiting embodiment of the present invention, step 2-1 specifically includes:

[0133] Based on historical seasonal ocean current data of the sea area where the subsea data center pile foundation scour protection device is located, an improved time-series ARIMA model is used to predict the ocean current velocity for the next 1-7 days. (k is the prediction step size, 1≤k≤168 hours), the improved time series ARIMA model formula is:

[0134] ;

[0135] Where k is the prediction step size, 1≤k≤168 hours. For the set constant term, (i=1,2,...,p) are the set autoregressive coefficients. (j=1,2,...,q) are the set moving average coefficients. For the set early error term, p and q are values ​​determined by the AIC information criteria (usually p=2-4, q=1-3). exist Real-time flow rate at any given moment.

[0136] Step 2-2: Construct a cumulative scour depth prediction model.

[0137] In a preferred but non-limiting embodiment of the present invention, step 2-2 specifically includes:

[0138] Predicted cumulative scour depth over the next 1-3 months (m is the number of months to be predicted, 1≤m≤3), and its calculation formula is as follows:

[0139] ;

[0140] Where m is the predicted number of months, 1≤m≤3, and g is the gravitational acceleration (9.8m / s²). The drag coefficient is set (determined based on the sediment particle size, e.g., when μ = 0.1-0.5 mm). =0.01-0.03), H1 is the height of the anti-scour plate of the anti-scour device for the pile foundation of the submarine data center, and Δt is the time step (taken as 24 hours).

[0141] Step 3: Execute a predictive regulation algorithm based on the predicted ocean current velocity and cumulative scour depth;

[0142] In a preferred but non-limiting embodiment of the present invention, step 3 specifically includes:

[0143] Step 3-1: Based on predicted ocean current velocity With cumulative scour depth Construct the objective function for attitude adjustment of the anti-scouring device. :

[0144] ;

[0145] in, For the MIN function, 1. The set weighting coefficients, 1+ =1 ( Adjustments will be made based on the specific requirements for sea area risk levels, with high-risk sea areas... The value can be 0.6-0.8, indicating a low-risk sea area. (Can be 0.3-0.5) The optimal oncoming flow velocity for the anti-impact plate is set (1.2-1.5 m / s can be selected). The set safe cumulative scour depth (determined according to the specific requirements of the pile foundation design strength, usually 0.5-1.0m). This represents the highest historical ocean current velocity in the sea area where the anti-scour device for the subsea data center's pile foundation is located. The maximum allowable scour depth for the pile foundation;

[0146] Step 3-2: Solve the objective function for attitude adjustment of the anti-scour device using the gradient descent method to obtain the optimal tilt angle. With optimal rotation angle The formula for calculating gradient descent is:

[0147] ;

[0148] in and Each is the current The tilt and rotation angles of the anti-erosion device are monitored at all times. These angles are acquired and transmitted to the control system via a tilt sensor and a rotary potentiometer connected to the control system, respectively. The set learning rate (can be between 0.01 and 0.05).

[0149] Inclined angle of the anti-scouring device The frontal area and rotation angle of the anti-impact plate are determined. Determine the angle between the sluice gate and the direction of the ocean current; when When the shovel is increased (and the slope of the shovel is more inclined), the area facing the current decreases, and the effective flow velocity acting on the seabed decreases; when When adjusted to be parallel to the ocean current direction, the current bypass effect is enhanced, and the local scouring intensity is weakened (this mechanism is common knowledge in the field of marine engineering; see Clause 5.2.3 of the "Technical Specification for Scour Prevention of Port Engineering Pile Foundations" (JTS / T327-2020)). The physical meaning of the objective function J and its indirect relationship with the angle: The core objective of the objective function is to achieve optimal scour prevention by minimizing the deviation between the predicted flow velocity and the optimal flow velocity, and the deviation between the predicted cumulative scour depth and the safe depth. Both the "predicted flow velocity" and the "predicted cumulative scour depth" are determined by the attitude angle of the scour prevention device (…). , The angle adjustment directly determines the flow rate and scouring depth by changing the interaction between the ocean current and the device, which is ultimately reflected in the numerical change of the objective function J.

[0150] The objective function J depends on the tilt angle of the anti-scouring device. Rotation angle of the anti-scouring device The partial derivatives of J can be calculated through mathematical differentiation (because J and J are related). , There is an indirect correlation, and the partial derivatives can be derived using the chain rule, i.e. , in and It can be obtained directly by differentiating the objective function J. The least squares method can be used to fit the result. and the tilt angle of the corresponding anti-scouring device The partial derivative is calculated using the functional relationship, and the tilt angle of the corresponding anti-erosion device is determined. The results can be derived by fitting a curve of time to the historical data of the tilt angle of the anti-erosion device using the least squares method. The least squares method can be used to fit the result. and the rotation angle of the corresponding anti-scouring device The partial derivative is calculated using the functional relationship, and the corresponding rotation angle of the anti-erosion device is determined. The result can be derived by fitting a curve of time to the historical data of the rotation angle of the anti-scouring device using the least squares method.

[0151] Step 3-3: The control system drives the hydraulic rod and ball bearing to adjust the tilt angle of the anti-erosion device. The rotation angle with the anti-scouring device is .

[0152] Step 4: Based on the maximum current velocity output by the prediction model, determine the thickness of the anti-scour plate and the density of the bionic aquatic plants in the anti-scour device.

[0153] In a preferred but non-limiting embodiment of the present invention, step 4 specifically includes:

[0154] During the segmented prefabrication stage of the anti-scouring device, the ocean current velocity output based on the prediction model was used. Maximum flow rate in Determine the thickness h of the anti-erosion plate of the anti-erosion device and the density of the bionic aquatic plants. During the prefabrication stage, biomimetic aquatic plants are arranged at this density to ensure erosion prevention under high flow velocities in winter.

[0155] ;

[0156] in, σ is the MAX function, where L is the span of the impact protection plate, and σ is the tensile strength of the impact protection plate material (e.g., for acrylic plates, σ = 60-80 MPa). The unit is plants / m² (the value ranges from 50 to 200 plants / m²). The minimum effective flow velocity threshold (unit: m / s) is defined as follows: when the flow velocity is below this value, the sediment transport capacity is weak, and there is no need to install aquatic plants. The flow rate can be taken as 0.5 m / s; K is the set conversion factor (unit: plants / (m²·(m / s))), which means: the increase in aquatic plant density required for every 1 m / s increase in "effective flow velocity". The value of K can be 10 / 3.

[0157] The beneficial effects of the present invention are as follows, compared with the prior art:

[0158] Predictive adjustment effect: Compared with the existing passive response control, this method can cope with high-velocity ocean currents 48 hours in advance, reducing the scour load on the pile foundation of the submarine data center by 35%-45% and the cumulative scour depth by 25%-30%, effectively avoiding the passive situation of "adjusting after scour intensifies".

[0159] Improved Adaptability: The thickness of the anti-scour plate and the density of biomimetic aquatic plants, optimized based on seasonal forecast data, have reduced the fluctuation range of the anti-scour efficiency of the subsea data center pile foundation scour protection device from ±20% to ±8% in different seasons, significantly improving its adaptability to the dynamic marine environment.

[0160] Long-term stability assurance: By integrating historical and real-time data, the predictive model can provide early warning of cumulative scour risks up to 3 months in advance. Combined with parameter optimization, the design service life of the subsea data center pile foundation is extended from 20 years to 25-30 years, reducing the operation and maintenance costs of the subsea data center.

[0161] Optimized response efficiency: The predictive adjustment algorithm of the control system has a response time of ≤30 seconds, which is more than 60% faster than the existing control system (1-2 minutes), ensuring rapid attitude adjustment in the event of sudden changes in ocean currents and further enhancing the reliability of protection.

[0162] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention without departing from the spirit and scope of the present invention. Any modifications or equivalent substitutions should be covered within the scope of protection of the claims of the present invention.

Claims

1. A method for controlling the construction of an anti-scouring device for a submarine data center, characterized in that, include: Step 1: Data acquisition and preprocessing of the anti-scour device used for the seabed data center; Step 2: Build a predictive model for scour protection devices used in subsea data centers; Step 3: Execute a predictive regulation algorithm based on the predicted ocean current velocity and cumulative scour depth; Step 4: Based on the maximum current velocity output by the prediction model, determine the thickness of the anti-scour plate and the density of the bionic aquatic plants in the anti-scour device.

2. The construction control method for the anti-scouring device for submarine data centers according to claim 1, characterized in that, Step 1 specifically includes: Step 1-1: Collect three types of core data from the anti-scour device used for the seabed data center; Step 1-2: Preprocess the collected data.

3. The construction control method for the anti-scouring device for submarine data centers according to claim 2, characterized in that, In step 1-1, the three types of core data include: Real-time monitoring data includes information obtained from water flow sensors connected to the control system regarding the tidal currents in the sea area where the subsea data center pile foundation scour protection device is located. Real-time flow rate at any given moment Furthermore, the data is transmitted to the control system, where it is acquired by the terrain scanner connected to the control system. Real-time flushing depth And the data is transmitted to the control system; Historical data, including the maximum seasonal ocean current velocity in the sea area where the subsea data center pile foundation scour protection device is located, over the past 5 to 10 years and recorded in the control system. And the seasonal average scour depth of the sea area where the anti-scour device for the pile foundation of the submarine data center is located over the past 5 to 10 years. ; Environmental parameters, including the particle size of marine sediment in the sea area where the anti-scouring device for the subsea data center pile foundation is located, are entered into the control system. Seawater density The depth of the seawater where the pile foundation is located ; In step 1-1, the terrain scanner acquires... Real-time flushing depth The methods specifically include: S1: Establish a baseline terrain model; S2: Perform real-time terrain data acquisition and matching, and determine high-risk and low-risk sea areas; S3: Calculate the real-time scour depth.

4. The construction control method for the anti-scouring device for subsea data centers according to claim 3, characterized in that, S1 specifically includes: Before installing the anti-scouring device, or within 1-2 weeks after installation when the seabed is in a stable state, benchmark topographic data acquisition and modeling should be completed. Benchmark topographic data acquisition and modeling includes data acquisition, elevation calibration, and benchmark modeling to obtain relative elevation data. ; S2 specifically includes: During the operation of the anti-erosion device, a terrain scanner is used to collect real-time terrain data at a set frequency: Real-time scanning: Activate the fixed-deployment terrain scanner and collect the current seabed relative elevation data using the same scanning density and elevation calibration method as the baseline modeling. ; Spatial matching: using the GPS positioning module built into the terrain scanner and the center coordinates of the pile foundation ( , This involves spatially aligning real-time terrain data with a baseline terrain model to ensure that the coordinates of scan points at the same physical location are consistent in both the real-time terrain data and the baseline terrain model. S3 specifically includes: Based on relative elevation data Relative elevation data 2 The real-time scour depth is calculated using the elevation difference method, and the specific formula is as follows: For any spatial coordinates ( , The scan points are used to flush depth features in real time. The calculation formula is: = - ； like >0: This indicates that the seabed elevation at this scan point is lower than the reference level, indicating erosion. The value is the scour depth; like ≤0: This indicates that the seabed elevation at this scanning point is higher than the reference surface, and the scour depth is recorded as 0; Next The average value of the scour depth obtained at each time step is taken as .

5. The construction control method for the anti-scouring device for subsea data centers according to claim 4, characterized in that, In S2, the method for determining high-risk and low-risk sea areas is as follows: First, four core indicators affecting seabed erosion were selected. The weights of each indicator were determined using the Analytic Hierarchy Process (AHP). The four core indicators are as follows: , , and , , , and The weights of the indicators are 0.4, 0.25, 0.2 and 0.15, respectively; The total risk score is calculated based on four core indicators and their weights. The calculation formula is as follows: ； in , , and They are , , and The normalized value; if Therefore, the sea area where the anti-scouring device for the pile foundation of the submarine data center is located is determined to be a low-risk sea area. if Therefore, the area where the anti-scouring device for the pile foundation of the submarine data center is located is determined to be a high-risk sea area.

6. The construction control method for the anti-scouring device for a submarine data center according to claim 5, characterized in that, Steps 1-2 specifically include: The moving average method is used to remove high-frequency noise from real-time monitoring data, and linear interpolation is used to fill in missing values ​​in historical data.

7. The construction control method for the anti-scouring device for a submarine data center according to claim 6, characterized in that, Step 2 specifically includes: Step 2-1: Construct a seasonal ocean current prediction model; Step 2-2: Construct a cumulative scour depth prediction model.

8. The construction control method for the anti-scouring device for a submarine data center according to claim 7, characterized in that, Step 2-1 specifically includes: Based on historical seasonal ocean current data of the sea area where the subsea data center pile foundation scour protection device is located, an improved time-series ARIMA model is used to predict the ocean current velocity for the next 1-7 days. The improved time series ARIMA model formula is as follows: ； Where k is the prediction step size, 1≤k≤168 hours. For the set constant term, The set autoregressive coefficients, The set moving average coefficient, Here, p and q are the values ​​determined by the AIC information criteria, representing the initial error term. exist Real-time flow rate at any given moment; Step 2-2 specifically includes: Predicted cumulative scour depth over the next 1-3 months The calculation formula is as follows: ； Where m is the predicted number of months, 1≤m≤3, and g is the gravitational acceleration. H1 is the set drag coefficient, H1 is the height of the anti-scour plate of the anti-scour device for the pile foundation of the submarine data center, and Δt is the time step.

9. The construction control method for the anti-scouring device for a submarine data center according to claim 8, characterized in that, Step 3 specifically includes: Step 3-1: Based on predicted ocean current velocity With cumulative scour depth Construct the objective function for attitude adjustment of the anti-scouring device. : ； in, For the MIN function, 1. The set weighting coefficients, 1+ =1, The optimal oncoming flow velocity for the anti-impact plate is set. The set safe cumulative scour depth, This represents the highest historical ocean current velocity in the sea area where the anti-scour device for the pile foundation of the submarine data center is located. The maximum allowable scour depth for the pile foundation; Step 3-2: Solve the objective function for attitude adjustment of the anti-scour device using the gradient descent method to obtain the optimal tilt angle. With optimal rotation angle The formula for calculating gradient descent is: ； in and Each is the current The tilt and rotation angles of the anti-erosion device are monitored at all times. These angles are acquired and transmitted to the control system via a tilt sensor and a rotary potentiometer connected to the control system, respectively. The set learning rate; Step 3-3: The control system drives the hydraulic rod and ball bearing to adjust the tilt angle of the anti-erosion device. The rotation angle with the anti-scouring device is .

10. The construction control method for the anti-scouring device for a submarine data center according to claim 9, characterized in that, Step 4 specifically includes: During the segmented prefabrication stage of the anti-scour device, the ocean current velocity output based on the prediction model was used. Maximum flow rate in Determine the thickness h of the anti-erosion plate of the anti-erosion device and the density of the bionic aquatic plants. : ； in, Let σ be the MAX function, where L is the span of the impact shield and σ is the tensile strength of the impact shield material. K is the set minimum effective flow rate threshold; K is the set conversion factor.

Images