Scour shape acquisition method and scour shape acquisition device

Abstract

To easily obtain an outline shape of scour occurring around a structure.SOLUTION: A scour shape acquisition method for acquiring an outline shape of scour comprises: a process (step S11) of assigning an initial value to at least one variable of a scour model that indicates a scour shape; a process (step S12) of generating a mesh for a fluid region; a process (step S13) of performing fluid analysis to determine a bottom shear stress; a process (steps S14 and S15) of determining whether or not to update at least one variable using an objective function that includes the bottom shear stress; a process (step S16) of obtaining an updated value of at least one variable using an optimization algorithm that uses the objective function when at least one variable is to be updated, and updating a scour shape indicated by the scour model using the updated value, to return to the step S12; and a process (step S17) of outputting a value of at least one variable as a final value when at least one variable is not to be updated.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a technique for acquiring the shape of scour occurring around a structure installed on the bottom of water. [Background technology]

[0002] Wind farms using fixed-bottom foundations have traditionally been used as renewable energy generation facilities. Simple monopile foundations are commonly used for fixed-bottom foundations, but in recent years, suction bucket foundations have been developed to reduce the cost of fixed-bottom foundations. Regardless of the foundation type, an important design requirement is to properly evaluate scouring that occurs around the foundation due to waves and tidal currents, and to prevent a loss of bearing capacity.

[0003] Traditionally, computational fluid dynamics (CFD) analysis has been performed on the fluid domain before scour occurs, and the location and shape of scour have often been estimated from physical quantities such as base shear stress around the foundation (also known as wall shear stress, hereafter simply referred to as shear stress) and vorticity. However, this one-way analysis method does not take into account the impact of changes in scour shape on the flow field. As a result, it has been difficult to determine the depth and shape of scour caused by the influence of long-term flows. Furthermore, adopting a calculation method that couples the flow field with the behavior of sand particles in the ground results in a high computational load and is therefore not practical.

[0004] On the other hand, in Non-Patent Document 1, the mesh used in the fluid analysis is manually deformed gradually at regular intervals to estimate the final scour shape due to bottom shear stress in the case of a monopile foundation. [Prior art documents] [Non-patent literature]

[0005] [Non-Patent Document 1] Pang et al., “A numerical approach for determining equilibrium scour depth around a mono-pile due to steady currents,” Applied Ocean Research 57 (2016), 114-124 Summary of the Invention [Problem to be solved by the invention]

[0006] Incidentally, the method in Non-Patent Document 1 estimates the scour depth by deforming the mesh shape without changing the scour position. In the case of a simple shape such as a monopile foundation, the scour shape can be estimated to some extent using the method in Non-Patent Document 1. However, in cases where the shape of the foundation is complex, such as a suction bucket foundation, or when large scour occurs, the mesh is significantly deformed and appropriate estimation is not possible.

[0007] The present invention has been made in consideration of the above-mentioned problems, and aims to easily obtain the general shape of scour that occurs around structures (of various shapes) installed on the bottom of the water. [Means for solving the problem]

[0008] A first aspect of the present invention is a scour shape acquisition method for acquiring an approximate shape of scour occurring around a structure installed on the waterbed, comprising: a) a step of assigning an initial value to at least one variable in a scour model indicating a scour shape determined by inputting a value to the at least one variable; b) a step of generating a mesh for a fluid domain having a waterbed surface including the scour shape indicated by the scour model; c) a step of performing fluid analysis based on the mesh to determine a bottom shear stress in the scour shape; d) a step of deciding whether to update the at least one variable using an objective function including the bottom shear stress; e) if the at least one variable is to be updated in step d), a step of obtaining an updated value of the at least one variable using an optimization algorithm that uses the objective function, updating the scour shape indicated by the scour model using the updated value, and returning to step b); and f) if the at least one variable is not to be updated in step d), a step of outputting the value of the at least one variable as a final value.

[0009] A second aspect of the present invention is the scour shape acquisition method of the first aspect, wherein the at least one variable includes a position of the scour shape and a depth of the scour shape.

[0010] A third aspect of the present invention is the scour shape acquisition method of the first aspect (which may be the first or second aspect), wherein the outer shape of the scour shape is a circle or an ellipse.

[0011] A fourth aspect of the present invention is the scour profile acquisition method of the first aspect (which may be any one of the first to third aspects), wherein the objective function uses an average value of bottom shear stress in the scour profile.

[0012] A fifth aspect of the present invention is the scour shape acquisition method of the first aspect (which may be any one of the first to fourth aspects), wherein the optimization algorithm obtains the updated value of the at least one variable by asynchronous parallel processing.

[0013] A sixth aspect of the present invention is the scour shape acquisition method according to any one of the first to fifth aspects, wherein the structure is a suction bucket foundation installed on the bottom of the water.

[0014] A seventh aspect of the present invention is a scour shape acquisition device for acquiring the general shape of scour occurring around a structure installed on the water bottom, comprising: a mesh generation unit that generates a mesh for a fluid domain having a water bottom surface including a scour shape determined by inputting a value for at least one variable of a scour model; a fluid analysis unit that performs fluid analysis based on the mesh and determines the bottom shear stress in the scour shape; an optimization calculation unit that obtains an updated value of the at least one variable by an optimization algorithm that uses an objective function including the bottom shear stress; and a control unit that determines whether to update the at least one variable using the objective function, and if the at least one variable is to be updated, obtains the updated value of the at least one variable by the optimization calculation unit, updates the scour shape indicated by the scour model using the updated value, and sequentially generates the mesh by the mesh generation unit and obtains the bottom shear stress by the fluid analysis unit, and returns to the step of determining whether to update the at least one variable; and if the at least one variable is not to be updated, outputs the value of the at least one variable as a final value. [Effects of the Invention]

[0015] According to the present invention, it is possible to easily obtain the general shape of scour occurring around a structure installed on the bottom of the water. [Brief explanation of the drawings]

[0016] [Figure 1] FIG. 10 is a diagram showing the flow of operations performed by the scour shape acquisition device to acquire the general shape of scour occurring around a structure installed on the water bottom. [Figure 2] This is a diagram showing the direction of movement of sand particles taking into account the slope of the scour surface. [Figure 3] FIG. 2 is a diagram illustrating a functional configuration of a scouring shape acquisition device. [Figure 4A] This is a plan view showing a scour model. [Figure 4B] FIG. 10 is a diagram showing a longitudinal section of a scour model. [Figure 5A] FIG. 1 is a schematic side view of an experimental tank. [Figure 5B] FIG. 2 is a schematic plan view of the experimental tank. [Figure 6A] FIG. 1 is a diagram showing a monopile-type test specimen. [Figure 6B] FIG. 1 shows a test specimen of Suction Type 1. [Figure 6C] FIG. 1 shows a test specimen of Suction Type 2. [Figure 6D] FIG. 1 shows a test specimen of Suction 3 type. [Figure 7] FIG. 10 is a diagram showing an example of the bottom shape distribution in a scour experiment conducted on a uniform flow. [Figure 8] FIG. 10 is a diagram showing the maximum scouring depth in the experiments of Cases 1 to 3. [Figure 9] FIG. 10 is a diagram showing an example of the distribution of bottom shapes in a scouring experiment involving irregular waves. [Figure 10] FIG. 10 is a diagram showing the maximum scouring depth in the experiments of Cases 4 to 7. [Figure 11A] FIG. 10 is a diagram comparing the experimental and analytical values ​​of the maximum scouring depth under uniform flow. [Figure 11B] FIG. 10 is a diagram comparing the experimental and analytical values ​​of the maximum scouring depth under uniform flow. [Figure 12A] This figure shows the relationship between the scour shape in the experiment under uniform flow and the bottom shear stress in the analysis. [Figure 12B] This figure shows the relationship between the scour shape in the experiment under uniform flow and the bottom shear stress in the analysis. [Figure 13] FIG. 10 is a diagram comparing the experimental and analytical values ​​of the maximum scour depth under irregular waves. [Figure 14] This figure shows the relationship between the scour shape in experiments under waves and the bottom shear stress in the analysis. DETAILED DESCRIPTION OF THE INVENTION

[0017] In this invention, remeshing in fluid analysis (computational fluid dynamics (CFD) analysis) is coupled with an optimization algorithm to change the analysis domain according to the flow conditions, and a convergent solution is obtained through iterative calculations. Specifically, the Shields number, which is the shear stress near the water bottom (bottom shear stress), is first obtained from the results of fluid analysis around the foundation structure, and this Shields number is compared with the critical Shields number at which scouring occurs. If the former exceeds the latter, the shear stress around the structure exceeds the minimum shear stress (critical shear stress) at which ground particles move, so the shape of the water bottom is changed and fluid calculations are performed again. A convergent scour depth (i.e., an equilibrium state where scouring does not progress) is obtained by minimizing the objective function defined as the difference between the Shields number near the water bottom and the critical Shields number over the entire analysis domain using an optimization algorithm (see Figure 1).

[0018] The Shields number and limit Shields number are calculated by first calculating the (bottom) shear stress τ of the water bottom including the area around the foundation structure from the results of fluid analysis. bed The Shields number θ is calculated from Equation 1, and the critical Shields number θ is calculated from the scour model by Soulsby et al. (Soulsby et al., "Threshold of sediment motion in coastal environments", Proceedings of combined Australasian coastal engineering and ports conference (1997), pp. 49-154) shown in Equation 2. cr0 Calculate the shear stress τ bed is the average value of shear stress in the scoured area that is preset as a scour-prone area or calculated.

[0019]

number

[0020]

number

[0021] where ρ s [kg / m 3 ] is the true density of the sediment, ρ [kg / m 3 ] is the density of the fluid, g[9.81 m / s 2 ] indicates the gravitational acceleration, and d [m] indicates the average particle size of the bottom sediment. * is the dimensionless particle size and is calculated using equation 3. Here, v[m 2 / s] is the viscosity coefficient of the fluid.

[0022]

number

[0023] However, the critical Shields number θ in equation 2 cr0 represents the limit value on the horizontal plane, and does not take into account the effect of the slope of the scour surface. Therefore, the limit Shields number θ, which takes into account the slope of the water bottom surface, proposed by Liu et al. (Liu, X, "Numerical Models for Scour and Liquefaction around Object under Currents and Waves", PhD thesis, 2008, University of Illinois at Urbana and Champaign) using equation 4, c was used.

[0024]

number

[0025] where μ s indicates the static friction coefficient of the ground, and β indicates the gradient of the scour surface. Furthermore, φ indicates the angle between the direction of shear stress and the scour surface, but as will be described later, if it is assumed that the outline of the scour hole develops in a circular shape, this will be 0.

[0026] Figure 2 shows the direction of movement of sand particles, taking into account the gradient of the scour surface, including the definitions of β and φ. D[N] is the drag and lift force acting on the particle, Ff [N] is the friction force, W τ [N] is the weight along the scour surface, τ b [Pa] is the shear stress, U b [m / s] indicates the particle movement speed.

[0027] Next, the analysis in this embodiment will be described in more detail with reference to Figures 1 and 3. Figure 1 is a diagram showing the flow of operations in which a scour shape acquisition device acquires the general shape of scour occurring around a structure installed on the water bottom. Figure 3 is a diagram showing the functional configuration of the scour shape acquisition device 1, which is a computer that realizes the operations.

[0028] The scour shape acquisition device 1 comprises an input unit 11, a control unit 12, an output unit 13, a mesh generation unit 21, a fluid analysis unit 22, and an optimization calculation unit 23. The input unit 11 accepts input of data generated by other devices or the computer that is the scour shape acquisition device 1, and accepts input from an operator via a mouse or keyboard. The control unit 12 controls the operation of other functional configurations while inputting and outputting information to and from other functional configurations. The output unit 13 outputs data to other devices or to a memory unit of the computer that is the scour shape acquisition device 1, and outputs information to a display.

[0029] The mesh generation unit 21 generates a mesh for fluid analysis. That is, it generates a mesh model that constitutes a fluid domain. The fluid analysis unit 22 performs fluid analysis based on the mesh and determines the shear stress on the water bottom surface. As will be described later, the optimization calculation unit 23 determines updated values ​​of multiple variables that determine the scour shape using an optimization algorithm. Note that, as will be described later, the number of variables that determine the scour shape may be one.

[0030] The scour shape acquisition device 1 may use dedicated software, but it can also be constructed on a computer using open-source software. For example, in this embodiment, the mesh generation unit 21 is realized by fluid analysis software "OpenFOAM" (registered trademark) (managed and developed by OpenCFD Inc.) and shape creation software "Gmsh." The fluid analysis unit 22 is realized by OpenFOAM. This allows the fluid analysis by the fluid analysis unit 22 to target both uniform flow and waves. In wave analysis, a free interface is set using the VoF (Volume of Fluid) method. In unsteady analysis, a second-order accurate finite volume method is used as the discretization method. In addition, the SST-k-Omega model is used as the turbulence model.

[0031] The optimization calculation unit 23 is realized by "Dakota," which is software for optimizing the values ​​of variables. Dakota optimizes the objective function X obj The updated values ​​of the variables are found to minimize the above. As an optimization algorithm, it is preferable to use Asynchronous Parallel Pattern Search (APPS), an asynchronous parallel processing algorithm implemented in Dakota. This is because the commonly used gradient method often only finds local solutions.

[0032]

number

[0033] The Shields number θ and the limit Shields number θ from the results of fluid analysis c Obtaining the objective function X obj A script written in Python (registered trademark) is used to calculate the above.

[0034] The control unit 12, which performs coupled calculations and iterative calculations by controlling the operations of the mesh generation unit 21, fluid analysis unit 22, and optimization calculation unit 23, is mainly realized by a Bash script, which is a shell script.

[0035] In the scour shape acquisition device 1, various initial values ​​are first set via the input unit 11 (FIG. 1: step S11). Initial values ​​are assigned to various physical property values ​​required for fluid analysis as well as variables that determine the scour shape. In this embodiment, to simplify the scour shape, the scour model that represents the scour shape is an inverted cone surface. The scour shape is determined by inputting values ​​to multiple variables of the scour model. In this embodiment, the multiple variables are the center position on the horizontal plane of the scour model and its depth. The initial values ​​of the multiple variables are set appropriately based on the results of separately conducted experiments or as empirical values. The initial radius of the scour model is also assigned as a constant. As the value of the depth variable increases, the scour model deepens while maintaining its initial radius, and when the inclination angle exceeds the angle of repose, the radius is enlarged and the inclination angle is maintained at the angle of repose.

[0036] FIG. 4A is a plan view showing the scour model 92 set up as described above, and FIG. 4B is a diagram showing a longitudinal cross section of the scour model 92 along a plane that includes the central axis of the suction bucket foundation 80 (more precisely, the suction bucket foundation model) and is parallel to the fluid flow. The suction bucket foundation 80 has a cylindrical shaft portion 81 and a cylindrical base portion 82. The diameter of the base portion 82 is larger than the diameter of the shaft portion 81, and their central axes coincide. The shaft portion 81 is located above the base portion 82. The scour model 92 is an inverted cone surface and is provided as part of the water bottom surface 91. If the area below the water bottom surface 91 is referred to as the "ground region 90" and the fluid region is referred to as the "fluid region 70," the water bottom surface 91 is the boundary between the fluid region 70 and the ground region 90. 4A and 4B, the distance between the center of suction bucket foundation 80 and the center of scour model 92 is indicated by x, and in FIG. 4B, the depth of scour model 92 is indicated by d.

[0037] Once the initial values ​​are set, the mesh generation unit 21 generates a mesh for the fluid domain 70 having the water bottom surface 91 including the scour shape shown by the scour model 92 (step S12). "Generating a mesh for the fluid domain 70" means dividing the fluid domain 70 into a set of minute solids that serve as units of calculation for fluid analysis. In the initial scour model 92, the depth d is 0 and the water bottom surface 91 is flat. Next, the fluid analysis unit 22 performs fluid analysis based on the mesh (step S13), finds the shear stress on the surface of the scour model 92, and calculates the average value of the shear stress as τ bed The control unit 12 obtains the Shields number θ by Equation 1. Also, before or after step S13, the control unit 12 obtains the limit Shields number θ taking the gradient into consideration by using Equations 2 and 3. c is calculated (step S14).

[0038] The control unit 12 calculates the objective function X obj is less than the threshold value, i.e., the Shields number θ and the critical Shields number θ c If the value is less than the threshold, the value of the variable of the scour model 92 is determined, and if the value is equal to or greater than the threshold, the process proceeds to step S16 to update the variable of the scour model 92 (step S15).

[0039] In step S16, the optimization calculation unit 23 uses Dakota to determine how to preferably change a plurality of variables (here, the center position and depth of the scour model 92), and the values ​​of the variables are updated (step S16). The updated values ​​of the variables are automatically determined by Dakota, but conceptually, the objective function X obj The search is performed to find whether the objective function X obj The updated values ​​of multiple variables are determined by searching for the most effective way to minimize the error. As explained above, Dakota sets search control parameters to prevent the updated values ​​from converging to local values.

[0040] When the updated values ​​of the multiple variables are determined, the scour shape indicated by the scour model 92 is updated by the updated values, and a mesh is generated by the mesh generator 21 (step S12), and fluid analysis is performed (step S13). obj is compared with a threshold value to check whether the value of the variable has converged, i.e., whether there is no need to change the scouring shape any further (steps S14, S15). Steps S12 to S16 are repeated to calculate the objective function X obj becomes smaller than the threshold value, the values ​​of the multiple variables at this point are determined as final values ​​(step S15), and these values ​​are output from the output unit 13 under the control of the control unit 12 (step S17).

[0041] As described above, in this embodiment, after updating the shape of the scour model 92, a mesh is regenerated. This prevents the mesh from being distorted and deformed, even for foundations with complex shapes, such as suction bucket foundations and jacket-type foundations with multiple legs, or for large scour, allowing for appropriate calculations. Furthermore, since this method also supports unsteady analyses, it can also perform scour analysis under high waves. On the other hand, in Non-Patent Document 1 (Pang et al.), the mesh is deformed but not regenerated, so when calculations are performed for foundations with complex shapes or large scour, the mesh is significantly deformed, making it impossible to obtain appropriate calculation results, i.e., an appropriately estimated scour shape.

[0042] Next, an example will be described in which actual scouring is obtained by experiment using a water tank and compared with the calculation results obtained by the scouring shape acquisition device 1 according to this embodiment.

[0043] Figure 5A is a schematic side view of the experimental water tank, and Figure 5B is a schematic plan view of the experimental water tank. The experiment was conducted in a two-dimensional wave- and flow-making water tank measuring 50 m in length, 1 m in width, and 1.2 m in height. The foundation models 83 were monopile and suction bucket types, and were each inserted and fixed into a sandy base 93 constructed in a container installed below the bottom of the water tank. The model dimensions were a 1 / 100 scale, based on the foundation of a 15 MW-class offshore wind power generation facility. The diameter of the monopile foundation model was 100 mm (equivalent to a 10 m diameter of the actual facility), and the diameter of the suction bucket foundation model was 255 mm (equivalent to a 25.5 m diameter of the actual facility). The experimental water depth was 300 mm (equivalent to a 30 m depth of actual scale).

[0044] To measure the flow velocity in the flow-making experiment and the wave height in the wave-making experiment, one electromagnetic current meter 61 and two capacitance wave height meters 62 were used. The scour shape of the sandy ground 93 was measured planarly using a laser displacement meter 63.

[0045] The uniform flow and irregular wave conditions used in the experiment are shown in Table 1.

[0046] [Table 1]

[0047] Cases 1–3 were conducted under uniform flow conditions, simulating tidal and ocean currents, and the flow was applied for approximately three hours. The flow velocities in Cases 1, 2, and 3 were 0.12 m / s, 0.18 m / s, and 0.26 m / s, respectively. Cases 4–7 were conducted under irregular wave conditions, with approximately 5,000 waves applied. The height and period of significant waves (the upper part of irregular waves) were 5.3 cm and 1.6 s in Case 4, 7.8 cm and 1.31 s in Case 5, 7.8 cm and 1.6 s in Case 6, and 9 cm and 1.6 s in Case 7, respectively. The bottom shape was measured after visually confirming that there was almost no change in the bottom shape. Silica sand No. 5 with a median particle size of 0.37 mm was used to construct the sand base 93.

[0048] The test specimens used in the experiment are shown in Figures 6A to 6D. Figure 6A (Monopile type) is a monopile type, and the model is cylindrical with a diameter of 100 mm. Figures 6B (Suction 1 type), 6C (Suction 2 type), and 6D (Suction 3 type) are suction bucket types, with the upper part (shaft section) cylindrical with a diameter of 100 mm and the lower part (bucket section) cylindrical with a diameter of 255 mm. Because the shapes of the bucket and shaft sections of the suction bucket foundation in the actual machine change discontinuously, stiffeners 84 are installed to maintain the rigidity of the structure, as in Suction 2 type in Figure 6C and Suction 3 type in Figure 6D. Suction 1 type does not have a stiffener installed.

[0049] Figure 7 shows an example of the distribution of bottom shapes in a scouring experiment with a uniform flow (Case 3). The upper left of Figure 7 corresponds to the monopile type, the lower left to Suction 1 type, the upper right to Suction 2 type, and the lower right to Suction 3 type. The water flow is from left to right. The bottom shape was measured planarly using a laser displacement meter, and the black areas in Figure 7 are areas where data is missing.

[0050] Of the four models, the monopile type had the greatest amount of scour, reaching a maximum depth of approximately 100 mm. The maximum scour depth for all suction bucket types (Suction Types 1 to 3) was approximately 20 mm, with Suction Type 3, which uses a flat stiffener, having a relatively large scour depth. Regarding the location of scour, in the case of the monopile type, a horseshoe-shaped vortex generated in front of the model mainly scoured the front and left and right sides of the pile. In the case of Suction Type 1, the horseshoe-shaped vortex generated on the top surface of the bucket prevented scour from occurring on the front of the foundation, but only slightly scoured the back of the foundation. On the other hand, in the case of Suction Types 2 and 3, vortices formed around the stiffeners caused scour from the left and right sides of the foundation diagonally backward. Furthermore, deposition occurred on the back of Suction Types 2 and 3 due to the wake caused by the stiffeners.

[0051] The maximum scour depth S / D in the experiments for Cases 1 to 3 is shown in Figure 8 (see Table 1 for the flow velocity for each case). The horizontal axis is the uniform flow velocity, and the vertical axis is the shaft diameter D (= 100 mm), which is the maximum scour depth S / D, which is a dimensionless version of the maximum scour depth S [mm]. In Figure 8, square dots correspond to the monopile type, diamond dots to Suction Type 1, triangular dots to Suction Type 2, and x dots to Suction Type 3. In all experimental models, the scour depth increased as the flow velocity increased, and the scour around the monopile type was deeper than that of the suction bucket type.

[0052] Figure 9 shows an example of the distribution of bottom shapes (Case 6) in an unsteady-state scour experiment using irregular waves. The upper left of Figure 9 corresponds to the monopile type, the lower left to Suction 1 type, the upper right to Suction 2 type, and the lower right to Suction 3 type. Waves move from left to right. The black areas in Figure 9 are areas where data is missing. Unlike the scour phenomenon caused by uniform flow, sand moved and ripples formed over the entire sandy ground, not just around the foundation. Furthermore, in the experiment using uniform flow, the scour location differed between the monopile type and the suction bucket type, but the effect of differences in foundation shape was small under waves, and in both cases scour was relatively large on both the left and right sides of the foundation.

[0053] Figure 10 shows the maximum scour depth S / D for the experiments in Cases 4 to 7. In Figure 10, square dots correspond to the monopile type, diamond dots correspond to Suction Type 1, triangular dots correspond to Suction Type 2, and cross dots correspond to Suction Type 3. Case 4 corresponds to wave conditions with a return period of 1 year, Case 5 corresponds to wave conditions with a return period of 30 years, Case 6 corresponds to wave conditions with a return period of 50 years, and Case 7 corresponds to wave conditions with a return period of 100 years. Maximum scour depth increased with increasing significant wave height, but even under the highest wave height conditions, scour depth remained below 0.2 times the shaft diameter. For the suction bucket foundation model, scour was significant for Suction Type 3, which uses flat stiffeners, while scour was relatively small for Suction Type 2, which uses beam-shaped stiffeners.

[0054] Next, we will explain the results of analytical calculations performed by the scour shape acquisition device 1 in response to the above experiment. The length of the analysis domain was set to 3 m to eliminate the effects of backflow or reflection from the outlet boundary. The width and depth of the numerical flume were set to 1 m and 0.3 m, respectively, as in the experiment, and the height of the gas phase was set to 0.2 m. For the inlet boundary, flow velocity conditions were applied in the case of uniform flow analysis, and wave conditions were applied in the case of irregular wave analysis. For the outlet boundary, zero gradient conditions were applied in the case of uniform flow analysis, and wave absorption conditions in the case of wave analysis. The true density of the sand was 2,650 kg / m 3 , the density and viscosity of water (at a water temperature of approximately 20°C) is 1,000 kg / m 3 and 1.002E-06m 2 / s. The angle of repose of the sand was set to 36°.

[0055] In the analysis, the analysis time was set to 10 seconds, and the Shields number was calculated from the average shear stress over the last 3 seconds to determine the objective function. Figures 11A and 11B show a comparison of the experimental (Exp) and analytical (CFD) values ​​for the maximum scour depth (S / D) under uniform flow. Figure 11A corresponds to the monopile type, and Figure 11B corresponds to the suction bucket type. In Figure 11B, square dots correspond to Suction Type 1, triangular dots correspond to Suction Type 2, and cross dots correspond to Suction Type 3. For reference, dashed lines indicate a ±20% discrepancy. Cases 1 to 3 in Figures 11A and 11B correspond to the conditions in Table 1. Figure 11A shows that for the monopile type case, the average error in the analytical values ​​is 18%, capturing the experimental trend. Figure 11B also shows that the experimental trend is captured for the suction bucket type case.

[0056] The relationship between the scour shape in the experiment under a uniform flow (Case 3) and the shear stress in the analysis is shown in Figures 12A and 12B. In Figures 12A and 12B, the left side shows the experimental results for the amount of scour, and the right side shows the shear stress in the fluid analysis. Figure 12A shows, from top to bottom, the monopile type and suction bucket type, while Figure 12B shows, from top to bottom, suction bucket type 2 and suction bucket type 3. As shown by the dashed ovals, it can be confirmed that actual scour developed in the areas of high shear stress in the analysis for both the monopile type and the suction bucket type.

[0057] Next, we will explain the results of the irregular wave analysis. In the irregular wave analysis, the Shields number was calculated from the maximum shear stress over a calculation time of 50 seconds. The number of waves entering the analysis field was around 40. Here, we compare cases for a suction bucket type. Figure 13 shows a comparison of the experimental values ​​(Exp) and analytical values ​​(CFD) for the maximum scour depth S / D under irregular waves. The average error of all analysis results was approximately 27%, and accuracy was relatively high under the conditions of Cases 4 and 5. For Cases 6 and 7, accuracy was inferior to Cases 4 and 5, but a certain tendency for scour depth was obtained.

[0058] Figure 14 shows the relationship between the scour shape in the experiment under waves (Case 6) and the shear stress in the analysis. In Figure 14, the left side shows the experimental results for the amount of scour, and the right side shows the shear stress in the analysis, with Suction 1, Suction 2, and Suction 3 shown from top to bottom. In the analysis for irregular waves, the shear stress was large near the left and right sides of the foundation, which can be confirmed to roughly match the range where scour developed in the model experiment. Furthermore, the experiment confirmed that even though irregular waves are a non-stationary phenomenon with no periodicity, they reach a convergent scour shape.

[0059] Next, it will be explained that the present invention is not limited to the above description.

[0060] The structures covered by this invention are not limited to foundations such as monopile foundations and suction bucket foundations, but can also include various structures installed on the bottom of the water. Installed on the "bottom of the water" means that the lower part of the structure is buried in the ground area and the upper part is above the water bottom. Note that this invention can also be applied to structures with complex shapes, so it is suitable for structures other than monopile types. Typically, it is suitable for structures that are suction bucket foundations installed on the bottom of the water.

[0061] Furthermore, in order to simplify the calculations in this invention, the general shape of scour that occurs around a structure is defined as a scour model, and the scour model determines the scour shape by inputting values ​​into multiple variables. In other words, the "scour model" is a model for determining the scour shape, and the "scour shape" is a shape that is determined by inputting specific values ​​into the variables of the scour model.

[0062] It is preferable that the scour model be as simple as possible, and the multiple variables preferably include the position of the scour shape and the depth of the scour shape. More preferably, the multiple variables are only the position of the scour shape and the depth of the scour shape. The "position of the scour shape" can be defined in various ways, but is typically the position of the center of gravity of the scour shape when viewed in a plane. The "depth of the scour shape" usually indicates the depth at the deepest point. In the above embodiment, the initial radius of the scour model is given as a constant rather than a variable, and as the scour depth increases, the scour model deepens while maintaining the initial radius, and when the inclination angle exceeds the angle of repose, the radius is enlarged and the inclination angle is maintained at the angle of repose. The initial radius of the scour model may be given as a variable in the optimization algorithm.

[0063] In order to minimize the number of variables, the outer shape of the scour shape indicated by the scour model is preferably a circle or an ellipse. Note that, if it is known from prior experiments that multiple (preferably two) deepest positions will occur, the number of depressions indicated by the scour model may be two or more. In other words, the outer shape of the shape indicated by the scour model may be multiple circles or ellipses. If it is permissible to increase the number of variables, the longitudinal cross section of the scour model may be curved in an approximately U-shape rather than a V-shape. Note that the number of variables may be one. For example, if the center position of the scour shape can be set in advance, only the depth of the scour shape may be a variable. In this way, the number of variables is at least one, and "multiple variables" in the above and following descriptions can be read as "at least one variable."

[0064] In the above embodiment, the objective function is given as the difference between the Shields number and the critical Shields number, and the average value of the (bottom) shear stress in the scour shape (i.e., the concave surface shown by the scour model) is used to calculate the Shields number. However, the objective function and the shear stress used may be changed in various ways. For example, instead of the average value of the shear stress, the maximum value of the shear stress may be used, or the median value may be used. Furthermore, a value other than the Shields number may be used in the objective function. In this way, the shear stress may be included in the objective function in various ways.

[0065] 1, a mesh is generated for a fluid domain having a water bottom surface including the scour shape indicated by the scour model. The mesh here may be any of various meshes that can be used for fluid analysis, and is not limited to meshes that represent a collection of hexahedrons, but may also be meshes that include other polyhedrons (i.e., various mesh-like structures composed of vertices and edges).

[0066] In step S13, fluid analysis is performed based on the mesh to determine the shear stress in the scour shape. "Shear stress in the scour shape" refers to the shear stress on the surface (bottom) of the recess shown by the scour shape. The software used for fluid analysis is not limited to that described in the above example. Furthermore, the fluid analysis part may be realized by dedicated hardware or an independent dedicated computer.

[0067] In steps S14 and S15, the control unit 12 determines whether to update the multiple variables using the objective function. If the multiple variables are to be updated, that is, if the objective function is equal to or greater than the threshold, updated values ​​of the multiple variables are obtained by an optimization algorithm that uses an objective function including shear stress, and the scour shape is updated using the updated values, and the process returns to step S12 (step S16). If the multiple variables are not to be updated, that is, if the objective function is less than the threshold, the values ​​of the multiple variables are output as final values ​​under the control of the control unit 12 (step S17).

[0068] The software used to update the values ​​of variables using the optimization algorithm is not limited to the software described in the above example. Furthermore, the optimization calculation part may be implemented using dedicated hardware or an independent dedicated computer. Tuning the optimization algorithm is important, and to prevent the algorithm from falling into a local optimum, the optimization algorithm preferably obtains updated values ​​for multiple variables using asynchronous parallel processing.

[0069] In the operation shown in FIG. 1, the control unit 12 determines whether to update multiple variables using an objective function (step S15). If multiple variables are to be updated, the optimization calculation unit obtains updated values ​​for the multiple variables (step S16). The scour shape is updated, and the mesh generation unit 21 generates a mesh (step S12). The fluid analysis unit 22 acquires bottom shear stress (step S13). The process then returns to the step of determining whether to update the multiple variables (step S15). Adjusting the optimization algorithm is important to obtain an appropriate solution without unnecessarily increasing the number of iterative calculations. Each iterative calculation is not related to real time. The above operation is a quasi-steady-state coupled analysis method in which the mesh is regenerated in step S12. This method enables calculation of long-term scour shape despite its lower computational load compared to fluid-particle coupled analysis.

[0070] Furthermore, the above operation makes it possible to easily obtain the general shape of scour that occurs around structures of various shapes installed on the seabed. This allows for accurate evaluation of the shape of scour holes caused by ocean currents and waves when designing structures and planning scour prevention works. Note that the first step S12 (mesh generation) may be set as a separate process from step S12 (re-mesh generation) from the second time onwards.

[0071] The functional components shown in FIG. 3, namely, the mesh generation unit 21, the fluid analysis unit 22, and the optimization calculation unit 23, may be realized by one computer or by multiple computers. Furthermore, the multiple computers may be located discretely via a long-distance network. Each functional component may be realized as dedicated hardware. The control unit 12 is a computer function that controls the operation of these functional components to realize the operation of FIG. 1. Of course, all functional components may be realized as dedicated integrated hardware.

[0072] The configurations in the above-described embodiment and each modification may be combined as appropriate as long as they are not mutually contradictory. [Explanation of symbols]

[0073] 1 Scour shape acquisition device 12 Control Unit 21 Mesh generation section 22 Fluid analysis section 23 Optimization calculation unit 70 Fluid domain 80 Suction bucket foundation (structure) 91 Underwater surface 92 Scour model Steps S11~S17

Claims

1. A scour shape acquisition method for acquiring an outline shape of scour occurring around a structure installed on the bottom of a body of water, comprising: a) providing an initial value to at least one variable in a scour model indicating a scour shape determined by inputting a value to the at least one variable; b) generating a mesh for a fluid domain having a water bottom surface including the scour shape indicated by the scour model; c) performing a fluid analysis based on the mesh to determine the bottom shear stress in the scour shape; d) determining whether to update the at least one variable using an objective function that includes the base shear stress; e) when updating the at least one variable in the d) step, obtaining an updated value of the at least one variable by an optimization algorithm using the objective function, updating the scour shape indicated by the scour model using the updated value, and returning to the b) step; f) if the at least one variable is not updated in the d) step, outputting the value of the at least one variable as a final value; A scour shape acquisition method comprising:

2. The scouring shape acquisition method according to claim 1, The scour shape acquisition method, wherein the at least one variable includes a position of the scour shape and a depth of the scour shape.

3. The scouring shape acquisition method according to claim 1, A scour shape acquisition method in which the outer shape of the scour shape is a circle or an ellipse.

4. The scouring shape acquisition method according to claim 1, A scour shape acquisition method, wherein the objective function uses an average value of bottom shear stress in the scour shape.

5. The scouring shape acquisition method according to claim 1, The scour shape acquisition method, wherein the optimization algorithm obtains the updated value of the at least one variable through asynchronous parallel processing.

6. The scouring shape acquisition method according to any one of claims 1 to 5, A scour shape acquisition method in which the structure is a suction bucket foundation installed on the bottom of the water.

7. A scour shape acquisition device for acquiring an outline shape of scour occurring around a structure installed on the water bottom, a mesh generation unit that generates a mesh for a fluid domain having a water bottom surface including a scour shape determined by inputting a value to at least one variable of a scour model; a fluid analysis unit that performs fluid analysis based on the mesh and calculates bottom shear stress in the scour shape; an optimization calculation unit that obtains an updated value of the at least one variable by an optimization algorithm that uses an objective function including the bottom shear stress; a control unit that determines whether to update the at least one variable using the objective function, and if the at least one variable is to be updated, obtains an updated value of the at least one variable by the optimization calculation unit, updates the scour shape indicated by the scour model using the updated value, and sequentially executes generation of the mesh by the mesh generation unit and acquisition of the bottom shear stress by the fluid analysis unit, returning to the step of determining whether to update the at least one variable, and if the at least one variable is not to be updated, outputs the value of the at least one variable as a final value; A scour shape acquisition device comprising:

Images