Method for adaptive variable spacing arrangement of side-type intake vortex suppression beam

Abstract

The application constructs a side-type water inlet vortex beam self-adaptive variable interval arrangement method, specifically: obtaining multi-condition water inlet front flow field data, including the time sequence velocity field of the vortex occurrence area under the typical operation condition; extracting vortex characteristics based on the flow field data, calculating the spatial distribution index for representing the contribution degree of each spatial position in the occurrence area to the vortex circulation; envelope processing the spatial distribution index under multiple conditions to obtain the comprehensive envelope distribution for design; determining the initial installation position and interval distribution sequence of the vortex beam within the effective range based on the comprehensive envelope distribution; constructing the beam-containing flow field based on the initial interval distribution sequence and performing numerical simulation, feeding back and updating the interval according to the evaluation result of the beam-containing flow field, terminating the iteration when the convergence criterion is met, and outputting the vortex beam self-adaptive variable interval arrangement scheme. The application solves the problems of missing traditional design mechanism and flow field feedback divergence, and realizes the physical self-adaptive arrangement of the vortex prevention component.

Description

Technical Field

[0001] This invention relates to the field of hydraulics and structural design technology for water intakes in water conservancy and hydropower projects, and in particular to an adaptive variable spacing arrangement method for vortex-eliminating beams in side-type water intakes. Background Technology

[0002] Side-mounted inlets are susceptible to flow shear, boundary layer separation, and water level fluctuations during operation, leading to the generation of intermittent suction vortices in the water area at the inlet's leading edge. These suction vortices not only cause vibration and cavitation erosion in pumps or turbines but also reduce flow capacity and operating efficiency. Properly arranging vortex-suppressing beams can disrupt the vortex rotation structure and weaken the tangential and radial velocities of the flow; determining the spatial spacing of the vortex-suppressing beams can also reduce vortex circulation, ensuring the hydraulic vortex-prevention safety of the inlet.

[0003] Currently, in the design of anti-vortex structures for side-mounted inlets, the arrangement of vortex-suppressing beams relies on physical model tests or empirical formulas to ensure equal spacing, or on qualitative non-uniform arrangement guidelines based on the probability density of vortex occurrence. When obtaining design basis, transient flow field observation methods are used to statistically analyze the frequency of vortex events at various spatial locations, thereby constructing a two-dimensional spatial probability field. In the stage of determining component parameters, designers often employ a parameter scanning method with controlled variables. Based on the beamless flow field diagnostic results, several equally spaced or qualitatively non-uniform arrangement schemes are assumed, and physical installation or numerical geometric reconstruction is performed for each scheme. This allows for comparison of the inlet cross-sectional velocity distribution or surface vortex morphology under each scheme, enabling the selection of the relatively superior structural parameters.

[0004] Existing technologies suffer from problems in practical applications, such as the lack of physical mechanisms for design indicators, the break in flow field feedback mechanisms, and the tendency for the optimization process to diverge, making it difficult to achieve suitable arrangement of vortex-reducing beams. Summary of the Invention

[0005] Purpose of the invention: In view of the above-mentioned problems in the prior art, this application provides an adaptive variable spacing arrangement method for the side inlet anti-vortex beam.

[0006] Technical solution: On the one hand, a method for adaptive variable spacing arrangement of vortex-eliminating beams for side-type water inlets, including:

[0007] Acquire flow field data in front of the inlet under multiple operating conditions. The flow field data includes time-series velocity fields of the vortex generation zone under multiple typical operating conditions.

[0008] Based on the flow field data, vorticity characteristics are extracted, and spatial distribution indicators are calculated to characterize the contribution of each spatial location in the vortex generation zone to the vortex circulation.

[0009] Envelope processing is performed on spatial distribution indices under multiple typical operating conditions to obtain a comprehensive envelope distribution for the design of vortex-reducing beams.

[0010] Based on the comprehensive envelope distribution, the initial installation position and spacing distribution sequence of the vortex-eliminating beam are determined within the preset range of action of the vortex-eliminating beam.

[0011] The flow field containing beams is constructed based on the initial spacing distribution sequence and numerical simulation is performed. The spacing is updated based on the evaluation results of the flow field containing beams. The iteration terminates when the preset convergence criterion is met, and the adaptive variable spacing arrangement scheme of the vortex-eliminating beams is output.

[0012] Beneficial effects: This invention solves the problems of missing design mechanisms and flow field feedback divergence in traditional design, and realizes the physical adaptive arrangement of anti-vortex components. Attached Figure Description

[0013] Figure 1 A flowchart illustrating an adaptive variable spacing arrangement method for a side-type inlet vortex-eliminating beam, provided in an embodiment of this application.

[0014] Figure 2 This is a flowchart illustrating an example of calculating a spatial distribution index used to characterize the contribution of each spatial location within the vortex generation zone to the vortex circulation, provided as an embodiment of this application.

[0015] Figure 3 This is another flowchart provided as an embodiment of the present application for calculating a spatial distribution index used to characterize the contribution of each spatial location within the vortex generation zone to the vortex circulation.

[0016] Figure 4 A flowchart for obtaining the comprehensive envelope distribution for vortex-reducing beam design, provided for embodiments of this application.

[0017] Figure 5 This is a flowchart illustrating how to determine the initial installation position and spacing distribution sequence of a vortex-eliminating beam within its effective range, as provided in an embodiment of this application. Detailed Implementation

[0018] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0019] It should be noted that the terms "first," "second," etc., in the specification and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a predetermined order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of the invention described herein can be implemented in sequences other than those illustrated or described herein. Furthermore, the terms "including" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0020] To address the aforementioned issues, the applicant conducted in-depth searches and analyses, and discovered:

[0021] The probability density of vortex occurrence only reflects whether a vortex occurs. It cannot quantify the physical contribution of high-frequency weak vortices and low-frequency strong vortices to the total circulation. Furthermore, directly calculating the absolute vorticity will introduce interference from reverse shear vortices, leading to the incorrect allocation of space defense resources.

[0022] In addition, most existing layout schemes are open-loop one-time designs. When using the point-by-point mapping method to determine the spacing, the deflection effect caused by the vortex-eliminating beam as a rigid water-blocking body is not considered. This leads to the high blockage ratio in the dense area causing the water flow to abruptly move to the sparse area, generating secondary vortices in the flow field.

[0023] Furthermore, when attempting to introduce feedback correction, the algorithm relies on the rate of change of the spacing coordinates as a purely numerical convergence criterion, neglecting the spatial distribution of the vortex elimination physical effect. Moreover, the extraction of vorticity in the beam flow field is easily contaminated by the Karman vortex street of the component wake, causing the spacing arrangement algorithm to fall into a nonlinear positive feedback divergence loop that constantly requires densification.

[0024] To solve these problems, combined with Figures 1 to 5 The present invention will be specifically described through the following embodiments.

[0025] This invention provides an optional implementation of a side-mounted inlet vortex-eliminating beam adaptive variable spacing arrangement method. Accordingly, it includes:

[0026] Step 101: Obtain flow field data in front of the inlet under multiple operating conditions. The flow field data includes the time series velocity field of the vortex generation zone under multiple typical operating conditions.

[0027] In this embodiment, the flow field data in front of the inlet under multiple operating conditions serves as the global initial input for all subsequent calculations; it can be obtained by calling a physical model and using particle image velocimetry technology, or by numerical simulation using computational fluid dynamics software.

[0028] Typical operating conditions are defined by different combinations of inlet flow Froude number and relative submersion depth.

[0029] The time-series velocity field is represented internally by a vector array on a discrete spatial grid. By acquiring instantaneous velocities at multiple time steps, the unsteady evolution of intermittent inhalation vortices can be recorded, providing a data foundation for subsequent extraction of time-domain fluctuation characteristics.

[0030] In some alternative implementations, when acquiring flow field data, spatial interpolation and low-pass filtering can be performed on the original collected grid velocity field in advance to filter out high-frequency background noise introduced by the testing instrument, making subsequent spatial derivative calculations more stable and reducing calculation deviations.

[0031] Step 102: Extract vorticity features based on flow field data and calculate spatial distribution indicators to characterize the contribution of each spatial location in the vortex generation zone to vortex circulation.

[0032] Alternatively, flow field data can be used to extract vorticity characteristics and calculate spatial distribution indices, which characterize the degree of contribution of each spatial location within the vortex generation zone to the vortex circulation.

[0033] Specifically, extracting vorticity features involves using a spatial difference algorithm to process the time-series velocity field and calculate the normal vorticity of the corresponding grid nodes. The spatial distribution index corresponds to a scalar field characterizing the contribution of local locations to the total vortex intensity. Using this index as a driving parameter for reverse design avoids the problems associated with relying solely on the probability of vortex occurrence as an evaluation basis, establishing a quantitative physical connection from the underlying hydrodynamic mechanisms to the arrangement of engineering structures.

[0034] In some embodiments, the spatial distribution index not only includes the static contribution component in the sense of time average, but also incorporates the fluctuation component characterizing the transient burst intensity of the flow field, which can more comprehensively measure the potential threat posed by intermittent vortices to engineering safety.

[0035] Step 103: Perform envelope processing on the spatial distribution indicators under multiple typical operating conditions to obtain the comprehensive envelope distribution for the design of the vortex-reducing beam;

[0036] Because the intake faces frequent rises and falls in water level and changes in discharge flow during actual operation, the flow field characteristics under a single operating condition cannot represent the overall risk during the project's operation. Therefore, envelope processing is performed on multiple typical operating conditions to extract boundary states that can cover various hydraulic conditions.

[0037] The integrated envelope distribution is represented internally by the computer as an integrated two-dimensional grid field or a one-dimensional discrete numerical array. As a unified design benchmark, it enables the spacing arrangement scheme calculated subsequently to have vortex-reducing protection capabilities across the entire operating range.

[0038] Furthermore, when performing envelope processing, it is necessary not only to filter out the extreme values ​​of each spatial location under multiple operating conditions, but also to introduce the actual occurrence probability of each operating condition in the reservoir scheduling plan as a weight parameter to avoid excessive redundancy in the overall layout scheme due to low probability operating conditions.

[0039] Step 104: Based on the comprehensive envelope distribution, determine the initial installation position and spacing distribution sequence of the vortex-eliminating beam within the preset range of action of the vortex-eliminating beam;

[0040] In this process, the initial installation position defines the center coordinates of each vortex-suppressing beam in geometric space, and the spacing distribution sequence is composed of the physical differences between adjacent installation positions. The spacing arrangement is guided by a comprehensive envelope distribution, following the physical mapping principle of matching small spacing to high contribution density regions and large spacing to low contribution density regions. This step completes the spatial adaptive allocation of discrete physical components to the continuous flow field, and the output initial installation position and spacing distribution sequence provide an initial guess solution for subsequent closed-loop correction.

[0041] Based on this, the calculated local spacing must meet the predetermined upper and lower limits, which are determined by the construction feasibility restrictions and the minimum flow area requirement of the structure.

[0042] Step 105: Construct the beam-containing flow field based on the initial spacing distribution sequence and perform numerical simulation. Update the spacing based on the evaluation results of the beam-containing flow field. Terminate the iteration when the preset convergence criterion is met, and output the adaptive variable spacing arrangement scheme of the vortex-eliminating beam.

[0043] Alternatively, based on the initial installation position and initial spacing distribution sequence, a beam-containing flow field is constructed and numerically simulated. The spacing in the initial spacing distribution sequence is updated based on the evaluation results of the beam-containing flow field. The iteration terminates when the preset convergence criterion is met, and the adaptive variable spacing arrangement scheme of the vortex-eliminating beam is output.

[0044] Furthermore, an adaptive feedback correction mechanism is introduced. Since the initial arrangement is based on the derivation of the beamless flow field, the addition of solid beams will cause the water flow to deflect towards the sparser regions with larger spacing, inducing new vorticity accumulation locally. Therefore, the flow field with beams can be reconstructed, its actual vortex-reducing capacity can be evaluated, and the current spacing distribution sequence can be dynamically adjusted according to the evaluation deviation.

[0045] The iterative calculation terminates when the differences between the arrangement schemes in multiple consecutive iterations are less than the preset convergence criterion. An adaptive variable-spacing arrangement scheme for the anti-vortex beams is output to suppress logical contradictions between the design basis and the design results, achieving self-consistency between the fluid dynamics state and the arrangement of the physical structure.

[0046] On the other hand, the specific implementation process of constructing and calculating the spatial distribution index is described, covering the absolute value superposition algorithm, the optimized directed same-sign vortex screening algorithm, and the time-domain fluctuation substitution reconstruction algorithm based on steady-state turbulent kinetic energy. Among these, the spatial distribution index is used to characterize the contribution of each spatial location to the vortex circulation. Further, the following steps can be adopted:

[0047] Step 201: Calculate the instantaneous vorticity field at each time step based on the time-series velocity field in the flow field data;

[0048] In this step, the collected discrete velocity field is processed using a spatial difference algorithm to calculate the normal vorticity of the grid nodes. Since flow field data is usually stored in the form of a two-dimensional vector array in a computer, the transverse and longitudinal velocities of each grid node are extracted, and the local velocity gradient is calculated based on the spatial distance between adjacent nodes to obtain the instantaneous vorticity field matrix characterizing the rotation intensity of the fluid micro-particles.

[0049] Step 202: Statistically analyze the instantaneous vorticity field in the time dimension to obtain the time-averaged vorticity field and the standard deviation field of vorticity fluctuation.

[0050] Accordingly, for the instantaneous vorticity values ​​recorded at all time steps, the arithmetic mean over the time dimension is calculated at each fixed spatial grid node to filter out high-frequency random fluctuations and obtain the time-averaged vortex field reflecting the macroscopic vortex structure. Simultaneously, the standard deviation of the instantaneous vortex relative to this time average is calculated, forming a vortex fluctuation standard deviation field. Extracting the vortex fluctuation standard deviation field can quantify the sudden intensity of intermittent intake vortices in front of the inlet, facilitating the coverage of such transient physical risks in subsequent design boundaries.

[0051] Step 203: The absolute value of the time-averaged vorticity field is superimposed with the vorticity fluctuation standard deviation field multiplied by a preset risk coefficient to obtain a two-dimensional circulation contribution density field.

[0052] Based on Stokes' theorem, total circulation equals the surface integral of vorticity at each point within the vortex region, and local vorticity characterizes the surface integral density of circulation at the corresponding spatial location. To quantify this density, the absolute value of the time-averaged vorticity field is directly extracted to eliminate the cancellation effect of positive and negative signs of vorticity during integration, and then added to a term representing the intensity of time-varying fluctuations. A preset risk coefficient can be set between 1.0 and 2.0. When the risk coefficient is 0, it is equivalent to considering only the static average vorticity; when the risk coefficient is greater than 0, intermittent fluctuations are given corresponding weights, forming a two-dimensional circulation contribution density field encompassing sudden risks.

[0053] Step 204: Integrate and project the two-dimensional circulation contribution density field along the preset length direction of the anti-vortex beam to obtain the one-dimensional circulation contribution density as a spatial distribution index.

[0054] In this design, the vortex-eliminating beams are typically arranged along a predetermined length, with variations in their spacing occurring only in the spatial dimension. Therefore, the cumulative trapezoidal numerical integration rule is used to continuously project the density field on the two-dimensional spatial plane along the length of the vortex-eliminating beams. Through this projection operation, the two-dimensional spatial physical information is reduced and compressed into a one-dimensional design variable consistent with the beam arrangement dimension. The output discrete one-dimensional array serves as the reference data for subsequent spacing mapping.

[0055] Based on this, an optional implementation method based on directional same-sign vorticity screening is provided to solve the technical problem that the reverse shear vortices generated by boundary layer separation in the flow field of a side inlet may interfere with the judgment of the main vortex.

[0056] Optionally, the absolute value of the global maximum vorticity of the time-averaged vorticity field can be extracted;

[0057] In the computer processing logic, all discrete grid nodes of the time-averaged vorticity field are traversed, and a numerical comparison search is performed to extract the maximum extreme value of the absolute vorticity. This global maximum absolute vorticity value calibrates the peak benchmark of the flow field rotation intensity under the current operating conditions.

[0058] Optionally, the absolute value of the global maximum vorticity is multiplied by a preset percentage threshold to obtain the background vorticity threshold;

[0059] The preset percentage threshold can be set between 0.05 and 0.10. Multiplying the global extreme value by this proportional coefficient yields the dynamic background vorticity threshold. This threshold serves as a numerical benchmark for distinguishing between the main vortex structure and weak background turbulence, helping to overcome the problem of fixed absolute values ​​failing under varying operating conditions.

[0060] Optionally, the spatial region in the time-averaged vortex field where the absolute value of vortex is greater than the background vortex threshold is defined as the vortex generation region.

[0061] By traversing all nodes in the flow field, the grid coordinates that satisfy the inequality criteria are combined to form a spatial mask set. The connected region corresponding to this set on the spatial plane is the vortex generation region. Through an adaptive threshold segmentation mechanism, the algorithm can automatically delineate the physical boundaries of the water area requiring vortex suppression design.

[0062] Optionally, the time-averaged vorticity field is calculated based on the flow field data, and the surface integral of the time-averaged vorticity field is performed in the vortex generation zone to extract the main rotation direction indicator of the vortex generation zone.

[0063] In other words, the instantaneous vorticity field at each time step is calculated based on the flow field data, and the instantaneous vorticity field at each time step is averaged over time to obtain the time-averaged vorticity field. The surface integral of the time-averaged vorticity field is then performed within the vortex generation region to extract the main rotation direction indicator of the vortex in the vortex generation region.

[0064] Within the defined vortex generation region, the time-averaged vorticity field is spatially discretized and cumulatively integrated in two dimensions. Based on the algebraic sign of the integration result, a marker for the dominant rotation direction of the vortex is assigned. For example, if the surface integral result is positive, the extracted dominant rotation direction marker is +1, corresponding to the vortex rotating counterclockwise; if the surface integral result is negative, the marker is -1, corresponding to clockwise rotation. It should be understood that the extraction of this marker is automatically determined based on the objective properties of the flow field.

[0065] Furthermore, based on this (the vortex main rotation direction indicator), the instantaneous vorticity fields corresponding to the flow field data are filtered for the same sign, and the vorticity components that are opposite to the vortex main rotation direction indicator are filtered out. The remaining vorticity components are converted into non-negative vorticity intensities with the same sign.

[0066] Alternatively, based on the vortex main rotation direction indicator, the instantaneous vorticity field at each time step is screened for the same sign, the vorticity components that are opposite to the vortex main rotation direction indicator are filtered out, and the retained vorticity components are converted into non-negative vorticity intensities with the same sign.

[0067] In this step, the instantaneous vorticity field is subjected to node-by-node logical operations using the vortex's main rotation direction indicator, as detailed below:

[0068] When the product of the original instantaneous vorticity and the sign of the main rotation direction of the vortex is greater than or equal to zero, the vorticity is determined to be a vorticity in the same direction, and the product of the two is taken as the retained vorticity intensity of the same sign.

[0069] When the product is less than zero, the vorticity is determined to be reverse vorticity, and the intensity of the corresponding vorticity of the same sign is forcibly set to zero.

[0070] By using the same sign filtering process, it is helpful to filter out the shear vorticity that has a counteracting effect on the formation of the main vortex, and uniformly retain the non-negative mathematical properties of vorticity, so as to ensure the self-consistency of the theoretical model for subsequent circulation calculation.

[0071] Furthermore, the intensity of the same vorticity at each time step is statistically analyzed to obtain the time-averaged vorticity field and the standard deviation field of the vorticity fluctuation.

[0072] Accordingly, after completing the same-sign vortex intensity extraction at each time step, the arithmetic mean and time fluctuation standard deviation are calculated for the retained same-sign vortex intensity sequences in the time dimension. The resulting same-sign vortex fluctuation standard deviation field only reflects the sudden fluctuations of the vortex components, which can suppress the random interference caused by the reverse background turbulence and make the statistical characteristics measure the temporal hazard of the effective vortex intensity.

[0073] Furthermore, the time-averaged vorticity field with the same sign is superimposed with the standard deviation field of vorticity fluctuation multiplied by a preset risk coefficient to obtain the directed circulation contribution density as a spatial distribution index. This spatial distribution index can be calculated using the following formula:

[0074] ρ _Γ =ω _mean_co +α×σ _co ;

[0075] Where, ρ _Γ For the density of directed circulation, ω _mean_co The time-averaged vorticity field with the same sign, α is the preset risk coefficient, and σ _co This is the standard deviation field of vorticity fluctuations with the same sign.

[0076] Since the components possess non-negativity after screening and flipping, the superposition calculation formula is always greater than or equal to zero. Compared to the method for calculating the absolute value of total vorticity, the index based on the extraction of directional vorticity of the same sign locks in the flow field region in the predetermined rotation direction that requires component protection, thus suppressing the resource waste and redundancy of excessively arranging vortex-suppressing beams in the reverse vorticity region.

[0077] In some implementations, when the vortex-eliminating beams are arranged in a single layer, the directed circulation contribution density can be further integrated and projected along the length of the vortex-eliminating beams to reduce the two-dimensional spatial distribution index to a one-dimensional spatial distribution index. This facilitates subsequent envelope processing and spacing calculation based on one-dimensional variables. This projection operation is the same as the integral projection method and is a conventional process that can be determined according to the dimensional requirements of the component layout.

[0078] Considering the computational cost of transient time series simulations, a dimensionality reduction method based on steady-state flow fields is provided, as detailed below:

[0079] Accordingly, when obtaining the standard deviation field of vorticity fluctuations of the same sign, the turbulent kinetic energy and integral scale corresponding to the flow field data are extracted;

[0080] In this method, the Reynolds-averaged Navier-Stokes equations are used for steady-state solutions, directly extracting the associated turbulent physical quantities from the computational matrix. Turbulent kinetic energy characterizes the energy magnitude of turbulent fluctuations in the flow field, while the integral scale characterizes the macroscopic geometric size of turbulent vortices. These two discrete field data directly constitute the input parameters for wave reconstruction.

[0081] Furthermore, based on the isotropic turbulence assumption, the ratio of the square root of the turbulent kinetic energy to the integral scale is calculated, and the ratio is multiplied by a preset empirical constant to reconstruct the standard deviation field of the same sign vorticity fluctuation.

[0082] Specifically, the equivalent reconstruction can be performed using the following formula:

[0083] σ _co_eq =C_k ×(sqrt(k) / L _t );

[0084] Where, σ _co_eq For the equivalent reconstructed standard deviation field of the same sign vorticity fluctuation, C _k Where L is a preset empirical constant, such as 0.4-0.6, k is the extracted turbulent kinetic energy, and L is the kinetic energy. _t The integral scale is used for extraction.

[0085] Based on the isotropic turbulence mechanism in fluid mechanics, the turbulent fluctuation velocity is proportional to the square root of the turbulent kinetic energy, while the vorticity dimension is expressed as velocity divided by length. Therefore, by using this algebraic ratio relationship, combined with empirical constant corrections, the vorticity fluctuation amplitude in the time domain can be reconstructed and restored using steady-state characteristic parameters without performing expensive transient time-series simulations, thus reducing the computational overhead in the automated layout process.

[0086] The above method extracts the main rotation direction indicator of vortices and performs directional same-sign vorticity screening to construct a directional circulation contribution density index. This method solves the problems that traditional probability density cannot quantify the physical contribution and that reverse shear vortex interference leads to the misallocation of space defense resources, thus achieving a connection between design indicators and the actual hydrodynamic vortex reduction mechanism.

[0087] On the other hand, an exemplary scheme describing the fusion algorithm for the comprehensive envelope distribution of multiple operating conditions covers the mixed envelope calculation process of pure maximum value and probability-weighted expected value, normalization examples, and supplementary stripping and verification strategies for extreme operating conditions. Specifically:

[0088] Step 301: Extract the pure maximum value of the spatial distribution index of each spatial location under multiple typical operating conditions;

[0089] In this step, for multiple preset typical operating conditions, the system traverses every discrete grid node in the three-dimensional fluid computational domain or the two-dimensional projection plane. At any given grid node, the spatial distribution index corresponding to that node under all operating conditions is extracted, and the distribution index with the largest value is selected as the pure maximum value through numerical comparison. Since the pure maximum value is extracted without considering the frequency of occurrence of the corresponding operating condition during the actual operation of the project, it constitutes a conservative boundary covering all unfavorable hydraulic conditions. Using this pure maximum value as one of the reference components ensures that the determined vortex-eliminating beam spacing distribution sequence still has defensive strength when encountering severe hydraulic operating conditions.

[0090] Step 302: Obtain the occurrence probability of each pre-configured typical operating condition, and calculate the probability-weighted expected value of the spatial distribution index of each spatial location under multiple typical operating conditions.

[0091] Specifically, the occurrence probability of each pre-configured typical operating condition can reflect the time proportion of each water level and flow rate combination in the long-term operation scheduling plan of the intake. After obtaining the occurrence probability, at the same discrete grid node, the spatial distribution index corresponding to each operating condition is multiplied by its corresponding occurrence probability, and the product results of all operating conditions are arithmetically summed.

[0092] Based on this, the probability-weighted expected value represents the average circulation contribution level under long-term operation of the project. By introducing the expected value component, the absolute dominance of low-frequency but high-value abnormal operating conditions on the overall layout design is suppressed, reflecting the trade-off made in the allocation of structural resources for economic efficiency.

[0093] Step 303: Using a preset conservatism coefficient, the pure maximum value and the probability-weighted expected value are linearly weighted and fused to obtain the comprehensive envelope distribution.

[0094] Accordingly, a formula can be used to process the two extracted components to balance engineering safety and economy, specifically:

[0095] ρ _env =η×ρ _max +(1-η)×ρ _exp ;

[0096] Where, ρ _env For the comprehensive envelope distribution, η is the preset conservatism coefficient, and ρ _max For the pure maximum value, ρ _exp This is the probability-weighted expected value.

[0097] The preset conservatism coefficient can be set from 0.6 to 0.8; the larger the coefficient, the more conservative the design. When the coefficient is set to 1, the calculation result is equivalent to the maximum value of each working condition density, and the design is dominated by unfavorable working conditions. When the coefficient is set to 0, the calculation result degenerates into a probability-weighted expected value dominated by more common working conditions. By adjusting this parameter, a transition between conservative and economical approaches can be achieved in the layout design.

[0098] Based on this, the normalization logic derivation process is given.

[0099] Furthermore, select any spatial grid node, obtain the normalized spatial distribution index of the node under multiple typical operating conditions, and determine the occurrence probability of each operating condition.

[0100] The value of the conservatism coefficient is preset, and the extreme value index is selected from each normalized spatial distribution index as a pure extreme value reference quantity.

[0101] By combining the normalized spatial distribution index of each working condition with the corresponding probability of occurrence, the probability-weighted expected value is obtained.

[0102] Substituting the conservatism coefficient, the pure extreme value reference, and the probability-weighted expected value into the calculation formula, the comprehensive envelope distribution result is obtained by solving the formula.

[0103] The comprehensive envelope distribution result can simultaneously take into account the peak characteristics of extreme conditions and the distribution characteristics of high-probability normal conditions.

[0104] In some implementations, when the preset conservatism coefficient is too small, the economic expectation value term may cause the impact of low-probability extreme operating conditions to be excessively diluted in the fusion calculation. If the spatial distribution index corresponding to a certain predetermined operating condition exceeds the average level of the flow field by a large margin, and the safety consequences it induces have a significant impact, it may not be covered by the comprehensive envelope distribution even if its probability of occurrence is low.

[0105] In other implementations, an extreme condition stripping and verification mechanism is provided to mitigate potential safety hazards arising from this computational characteristic. Specifically, before performing multi-condition envelope processing, the maximum spatial distribution index corresponding to each condition is compared. Abnormal conditions whose values ​​exceed a preset safety threshold are extracted and marked as safety verification conditions. These marked safety verification conditions are then removed from the overall probability-weighted fusion system, and only the remaining set of regular operating conditions is used to generate a comprehensive envelope distribution and deduce the spacing arrangement scheme. After obtaining the arrangement scheme, independent anti-vortex performance verification calculations are performed on the stripped safety verification conditions to prevent low-frequency, high-risk incidents.

[0106] In some embodiments, optional implementations describing the inversion arrangement and truncation adjustment of the spacing distribution are provided. Based on the comprehensive envelope distribution, the initial installation position and spacing distribution sequence of the vortex-eliminating beam can be determined. This method also includes the cumulative circulation equal division inversion arrangement method and its corresponding geometric ratio over-limit truncation adjustment mechanism, while providing a non-uniform spacing linear mapping arrangement method. Specifically, it includes:

[0107] Step 401: Inject a preset background noise into the comprehensive envelope distribution, perform spatial integration within the range of the vortex-eliminating beam, and construct a strictly monotonically increasing cumulative circulation contribution function.

[0108] In this step, considering the possibility of local vortex-free regions with zero vorticity in the flow field, directly integrating the comprehensive envelope distribution would result in a horizontal line segment with zero derivative in these vortex-free regions, leading to a collapse where the inverse function solution has infinitely many solutions. Therefore, positive background noise is injected into the integrand to make the integration result a monotonically increasing function. The cumulative circulation contribution function can be constructed using the following formula:

[0109] F(x) = integral(ρ)_env +ε _bg )dx;

[0110] Where F(x) is the cumulative circulation contribution function at coordinate x, ρ _env Let ε be the numerical value of the comprehensive envelope distribution at the current integration position. _bg The background noise is the preset value, integral is the integral operation performed along the specified spatial variable within the range of the anti-vortex beam, and dx is the spatial coordinate differential element.

[0111] The preset background noise can be set to a small proportion of the global maximum value of the comprehensive envelope distribution, for example, 10. -6 The magnitude of this tiny noise is negligible. It will not produce an observable bias in the arrangement result, but it will ensure that the integrand is always greater than zero, making the subsequent inversion root-finding solution unique.

[0112] Step 402: Obtain the total number of pre-configured vortex-eliminating beams, and based on the total number of vortex-eliminating beams, divide the total circulation contribution of the cumulative circulation contribution function into equal parts, and calculate the target cumulative value of each vortex-eliminating beam.

[0113] The total number of pre-configured vortex-suppressing beams is predetermined by engineering safety specifications, construction conditions, or construction cost constraints. The maximum function value of the constructed cumulative circulation contribution function at the terminal interface of the vortex-suppressing beam's effective range is extracted as the total circulation contribution, and this value is divided into several equal parts, equal in number to the number of intervals between adjacent beams. Each equal part represents the equal physical vortex-suppressing responsibility that an independent interval should bear, and the target cumulative value corresponding to the installation section of each physical vortex-suppressing beam in the flow field is calculated.

[0114] Optionally, if the total number of vortex-eliminating beams required by the actual project needs to vary, the spacing distribution parameters can be adaptively adjusted by scaling them proportionally. After setting a new target total number of beams, the original spacing distribution sequence value is multiplied by the ratio of the product of the total effective range length of the vortex-eliminating beams and the new target total number of beams minus one, to satisfy the adjusted structural resource allocation constraints.

[0115] Step 403: Perform equal division inversion and root finding on the target cumulative value of each antivortex beam in the cumulative circulation contribution function to calculate the initial installation position;

[0116] Specifically, based on discrete grid points, the cumulative trapezoidal numerical integration rule is used to generate a discrete cumulative array containing coordinates and cumulative values;

[0117] For each target cumulative value, a binary search algorithm is used to locate the boundary of the discrete interval in the discrete cumulative array;

[0118] A local continuous function is constructed within the discrete interval boundary using a one-dimensional interpolation algorithm, and the initial installation position corresponding to the continuous physical coordinates is calculated in reverse.

[0119] Since the flow field physical data extracted from physical model experiments or numerical simulations are represented as a discrete set of grid points, the cumulative circulation contribution function is actually represented as a one-dimensional discrete numerical array in the computer's memory. To solve for the installation position in a continuous physical coordinate system, the system uses the cumulative trapezoidal numerical integration rule to process the discrete grid data, generating a monotonically increasing discrete cumulative array containing coordinate indices and corresponding cumulative values. Next, for each target cumulative value, a binary search algorithm is used to search within the discrete cumulative array to locate the boundary of the discrete data interval. After the location operation is completed, a local continuous function relationship is constructed within the determined interval boundary using one-dimensional linear spline interpolation or monotonically cubic Hermitian interpolation algorithm. The continuous physical coordinates of the corresponding target cumulative value are calculated through inverse mapping and output as the initial installation position.

[0120] Step 404: Based on the coordinate difference between adjacent initial installation positions, obtain the spacing distribution sequence.

[0121] Accordingly, the physical coordinates of the next vortex-eliminating beam are extracted, and the physical coordinates of the previous adjacent vortex-eliminating beam are subtracted to calculate the physical span parameter between the two adjacent solid beams. By traversing all adjacent components, a spacing distribution sequence can be formed. The sequence derived based on the principle of cumulative integral equal division has the non-uniform spatial topology property of short physical spans in high contribution density areas and long physical spans in low contribution density areas, which can avoid the local over-density problem caused by the point-by-point mapping method when encountering extreme change points in flow field density.

[0122] After obtaining the initial spacing distribution sequence, the entity layout parameters still need to meet the physical constraints determined by structural strength and flow capacity. The following steps provide an out-of-bounds truncation correction mechanism that can maintain the equally divided topological characteristics, namely:

[0123] Using the preset maximum and minimum spacing limits, the spacing of each interval in the spacing distribution sequence is identified as exceeding the limit. The intervals that exceed the limit are truncated to the corresponding limit value, and the length of the truncated margin is calculated.

[0124] The preset minimum spacing limit parameter can be determined based on the material dimensions of the vortex suppressor beam itself, hydraulic flow control standards, and construction feasibility. The preset maximum spacing limit parameter can be set based on the minimum grid density control requirements for intercepting water flow vortices. Each value in the sequence is compared with the bidirectional limits. If a local span parameter is lower than the minimum spacing limit, it is assigned the lower limit constant; if a local span parameter is higher than the maximum spacing limit, it is assigned the upper limit constant. After completing the extreme value truncation operation, the difference between the original total effective range length of the vortex suppressor beam and the sum of the spans of all currently truncated intervals is calculated to determine the truncation margin length that needs to be redistributed.

[0125] Furthermore, identify the set of untruncated free intervals and extract the current physical interval length of each free interval in the set of free intervals;

[0126] During the truncation and screening process, compliant intervals that do not trigger the mandatory boundary assignment mechanism constitute a set of free intervals. The current physical interval length parameter of each independent interval within this set is recorded, and the original parameters retained directly map the relative density distribution characteristics determined by the inversion process.

[0127] Furthermore, based on the geometric proportion of the current physical interval length of each free interval to the total length of the free interval set, the truncation margin length is proportionally allocated and superimposed onto the corresponding free interval, thus updating the spacing distribution sequence and the corresponding initial installation position.

[0128] When allocating the surplus, it is not advisable to calculate based on the cumulative circulation contribution ratio within each interval. This is because the equal division inversion operation results in equal circulation contributions for each free interval, leading to an arithmetic equivalence of the surplus parameters and disrupting the spatial rationality of the non-uniform arrangement. Therefore, a geometric scaling calculation rule based on the current physical interval length parameter is adopted. The following formula is used for adjustment calculation:

[0129] s _adj =s _n +(s _n / S _sum )×ΔL;

[0130] Among them, s _adj For the updated corresponding free interval spacing, s _n S is the length of the current physical interval to be extracted. _sum ΔL is the sum of the lengths of all intervals within the set of free intervals, and ΔL is the truncation margin length generated by the preceding operation.

[0131] Alternatively, local linear physical laws can be used to generate the spacing distribution sequence. This method includes the following processing steps.

[0132] Optionally, the maximum and minimum distribution values ​​of the comprehensive envelope distribution within the action range of the vortex-eliminating beam can be extracted;

[0133] Traverse all discrete node data within the bounding box of the vortex-eliminating beam's effective range, search and extract the corresponding maximum and minimum values, and establish the data normalization reference interval required for local numerical mapping.

[0134] Optionally, based on the pre-defined physical mapping law that the circulation constant of the water flow surface has a linear relationship with the relative spacing of the vortex-eliminating beams, a non-uniform spacing mapping formula is constructed.

[0135] Based on the experimental characteristics of physical hydraulic models, the local flow surface circulation constant and the relative spacing of the vortex-eliminating beams exhibit an approximately linear decreasing trend within the control range. Based on this physical relationship, a constraint model is established that establishes a direct proportional relationship between the local design target circulation reduction and the local circulation contribution density, leading to the derivation of a linear mapping calculation model suitable for non-uniform layout scenarios.

[0136] Optionally, the numerical values ​​of the comprehensive envelope distribution corresponding to each spatial location are substituted into the non-uniform spacing mapping formula, and the local relative spacing is calculated point by point by combining the pre-configured minimum relative spacing and maximum relative spacing.

[0137] In other embodiments, the numerical values ​​of the comprehensive envelope distribution corresponding to each spatial location are substituted into the non-uniform spacing mapping formula, and the local relative spacing is calculated point by point by combining the pre-configured minimum relative spacing and maximum relative spacing, as well as the maximum distribution value and minimum distribution value.

[0138] The following formula can be used to perform point-by-point calculations:

[0139] s _ratio =s _max -(s _max -s _min )×(ρ _env -ρ _min ) / (ρ _max -ρ _min );

[0140] Among them, s _ratio s represents the local relative spacing of the current coordinate nodes. _max For the pre-configured maximum relative spacing, s _min For the pre-configured minimum relative spacing, ρ _env ρ represents the comprehensive envelope distribution value of the current coordinate node. _max To extract the maximum distribution value, ρ _min This is the minimum distribution value extracted.

[0141] Optionally, the initial installation position and spacing distribution sequence can be determined by taking the starting position of the vortex-eliminating beam's effective range as a reference and arranging it sequentially according to the local relative spacing.

[0142] Accordingly, using the coordinates of the first end of the vortex-eliminating beam as the base point, the local relative spacing value corresponding to the current physical position is extracted as the geometric step size. The spatial placement coordinates of the next adjacent vortex-eliminating beam are accumulated and deduced until the accumulated coordinate axis crosses the end boundary of the range of action of the vortex-eliminating beam, generating a complete sequence containing several discrete arrangement coordinates. The execution logic of this scheme is relatively straightforward and is mainly suitable for engineering layout scenarios where the flow field distribution gradient in front of the inlet is gentle and there are no extreme density extrema.

[0143] In other embodiments, feasible implementation methods for physically adaptive iteration and anti-divergence mechanisms are provided, particularly anti-divergence spatial mask filtering, residual density homogenization dual-criteria logic, power-law adaptive adjustment, and normalization algorithms. A relaxation iteration scheme based on numerical deviation is also provided. Specifically, this includes:

[0144] Step 501: Apply a wake space mask filter to the beam-containing flow field obtained from each numerical simulation.

[0145] The flow field region within the preset beam width range downstream of each vortex-eliminating beam is set as the mask blind zone. The vortex data in the mask blind zone is forcibly set to zero to shield the wake vortex separation characteristics induced by the flow around the vortex-eliminating beam from participating in the subsequent residual density statistics.

[0146] In other words, the flow field region within the preset beam width range downstream of each vortex-eliminating beam is set as the mask blind zone, and the vorticity data within the mask blind zone is forcibly set to zero to obtain the filtered beam-containing flow field. Based on the filtered beam-containing flow field, the residual circulation contribution density of each beam interval is calculated.

[0147] In this step, the solid vortex-eliminating beam, as a rigid water-blocking component, is placed in the fluid environment, inducing alternating positive and negative Kármán vortex street characteristics downstream on its back side. This wake-separated vortex structure contains vorticity components with the same rotation direction as the macroscopic main vortex. If the vorticity of the entire flow field is directly extracted, the underlying same-sign filtering algorithm will misjudge the wake features as residual components of the main vortex that have not been sufficiently eliminated, causing the iterative algorithm to issue an update command for further densification of the arrangement in this region. As the spacing value continues to decrease, the wake water-blocking effect increases nonlinearly, and the induced pseudo-same-sign vorticity continues to increase, triggering positive feedback divergence at the physical mechanism level.

[0148] By applying a wake space mask filter, the preset beam width range can be specifically set to a rectangular geometric space extending 1.0 to 3.0 times the width of the solid beam in the direction of water flow. Overwriting the vorticity values ​​of all discrete grid nodes falling within the geometric space to zero allows subsequent statistical algorithms to extract only the low-frequency physical field features characterizing the macroscopic vortex intensity.

[0149] Optionally, based on the beam-containing flow field obtained from each numerical simulation, the residual circulation contribution density of each inter-beam interval is calculated.

[0150] The filtered spatial distribution index of the flow field within the beams is obtained. Curve integration is performed within each independent physical space interval formed by two adjacent anti-vortex beams. The calculated integral result is then divided by the current span parameter of that independent physical space interval. This metric quantitatively characterizes the potential circulation formation intensity that has not been reduced within each local water area under the current arrangement of the physical components.

[0151] Optionally, the global average residual density and coefficient of variation of the residual circulation contribution density are calculated, and the coefficient of variation is used as the physical convergence criterion.

[0152] In this step, the residual circulation contribution density values ​​of all inter-beam intervals are extracted, and their arithmetic mean is calculated to obtain the global average residual density. The standard deviation of the numerical set is extracted, and the ratio of the standard deviation to the global average residual density is calculated to output the coefficient of variation. This coefficient of variation characterizes the dispersion of local vortex dissipation capacity within the global range of action; the lower the value, the more spatially uniform the allocation of vortex dissipation and protection resources tends to be.

[0153] Alternatively, the spatial distribution index is recalculated based on the beam-containing flow field obtained from each numerical simulation. The new spacing for the current round is determined based on the recalculated spatial distribution index. The relative deviation between the new spacing and the old spacing of the previous round is calculated, and the relative deviation is used as the numerical convergence criterion.

[0154] Furthermore, the iteration terminates when a preset convergence criterion is met, specifically by employing a dual-criteria convergence determination, as shown below:

[0155] The relative deviation between the calculated new spacing in the current round and the old spacing in the previous round is less than the preset numerical convergence threshold.

[0156] The coefficient of variation is less than the preset physical convergence threshold;

[0157] When both the relative deviation condition and the coefficient of variation condition are met, the iteration is considered to have converged and the iteration is terminated.

[0158] In other words, when the relative deviation of the spacing distribution sequence before and after the update is less than the numerical convergence threshold;

[0159] When the coefficient of variation is less than the physical convergence threshold;

[0160] The iteration converges and terminates.

[0161] In this application, the preset numerical convergence threshold can be set to 0.05, and the preset physical convergence threshold can be set between 0.15 and 0.20. Relying solely on a single numerical criterion that the coordinate arrangement no longer shifts cannot reveal issues of over-design or insufficient protection in local areas of the engineering scheme. In other words, relying solely on a single, fixed coordinate arrangement criterion to determine displacement is insufficient to detect over-design or weak protection in local areas of the engineering scheme. By employing a combination of logic and computational criteria, the output spacing distribution sequence achieves a globally uniform configuration of physical eddy current reduction effects while simultaneously reaching the numerical calculation fixed point.

[0162] Furthermore, for the problem of multiple local optima that may exist in nonlinear fluid dynamics coupled systems, if the global eddy-elimination physical parameters of the evaluation output do not meet the engineering design standards after the iteration reaches the dual-criteria convergence state, a linear scaling perturbation of a preset ratio can be applied to all values ​​of the initial spacing distribution sequence to construct a new initial value sequence and restart the iteration process. The flow field parameters of multiple convergence results are compared to output a better arrangement scheme.

[0163] When the coefficient of variation does not meet the preset physical convergence threshold, the power-law adaptive adjustment is performed based on the ratio of the residual circulation contribution density of each beam interval to the global average residual density, so as to obtain the updated spacing distribution sequence.

[0164] The pre-adjusted spacing is obtained by multiplying the old spacing of each beam interval by the adjustment exponent of the ratio of residual circulation contribution density to global average residual density.

[0165] Optionally, the pre-adjustment spacing can be derived, and the calculation formula can be expressed as:

[0166] s _pre =s _old ×(ρ _res / ρ _mean ) -γ ;

[0167] Among them, s _pre For pre-adjustment spacing, s _old For the old spacing, ρ _res ρ contributes density to the residual circulation in the current interval. _mean γ represents the global average residual density, and γ is a preset adjustment index.

[0168] The preset adjustment index controls the sensitivity of the spacing variable to the non-uniform distribution of residual density, and its specific value range can be set from 0.3 to 0.5. If the residual density in a local area is higher than the global average, it indicates that the area is insufficiently protected. The base is greater than 1, and after negative exponentiation, the multiplier is less than 1, resulting in a smaller output pre-adjusted spacing to achieve component densification. Conversely, the pre-adjusted spacing is increased to release redundant resources.

[0169] In other embodiments, when the coefficient of variation does not meet the preset physical convergence threshold, the current spacing of each beam interval is multiplied by the preset adjustment exponent power of the ratio of the residual circulation contribution density to the global average residual density to obtain the pre-adjusted spacing of each beam interval. The pre-adjusted spacing is then normalized to obtain the updated spacing distribution sequence.

[0170] Furthermore, the preset total effective range length of the vortex-eliminating beam is obtained, and all pre-adjusted spacings are scaled and normalized proportionally so that the sum of the normalized spacings is equal to the total effective range length of the vortex-eliminating beam, thus obtaining an updated spacing distribution sequence that satisfies the total length conservation.

[0171] The pre-adjusted spacing sequence derived through nonlinear power operations will inevitably deviate from the preset total effective range length of the anti-vortex beam, causing dimensional drift in the engineering project. Global proportional scaling and normalization can be performed using the following formula:

[0172] s _new =s _pre ×(L_s / S _pre_sum );

[0173] Among them, s _new To ensure the updated spacing while maintaining total length conservation, L_s is the preset total effective range length of the anti-vortex beam, and S... _pre_sum This is the sum of all pre-adjusted spacings. This scaling operation restores the geometric constraints of the macroscopic layout boundary of the project while maintaining the relative density ratio topological relationship of each sub-interval.

[0174] As an alternative to physical adaptive iteration, when the nonlinear coupling of the flow field is low, a pure numerical relaxation iteration method can be used to calculate the update scheme.

[0175] Furthermore, based on the beam-containing flow field obtained from each numerical simulation, the new spacing for the current round is extracted, and the relative deviation between the current spacing and the old spacing from the previous round is calculated. The relative deviation is used as the numerical convergence criterion.

[0176] If the relative deviation does not meet the preset convergence threshold, the preset relaxation factor is used to perform relaxation weighting calculation on the new spacing and the old spacing to obtain the updated spacing distribution sequence and enter the next round of iteration.

[0177] Accordingly, the relaxation weighting can be calculated using the following formula:

[0178] s _updated =β×s _new +(1-β)×s _old ;

[0179] Among them, s _updated For the updated spacing distribution sequence elements, s _new To extract the new spacing for the current round, s _old The previous spacing is represented by β, which is a preset relaxation factor. This preset relaxation factor is introduced to suppress numerical oscillations caused by the nonlinear flow field response of the water-blocking component. Specific parameter selection can be based on the local maximum blockage ratio of the vortex-eliminating beam, using adaptive matching. The local maximum blockage ratio is set as the ratio of the solid beam width to the minimum spacing limit. When the calculated local maximum blockage ratio is greater than 0.3, the flow deflection effect is sensitive, and the preset relaxation factor is set to 0.3. When the blockage ratio is less than or equal to 0.3, the preset relaxation factor is set to 0.5, ensuring computational stability while accelerating the convergence process.

[0180] In this invention, the one-time design and point-by-point mapping of open loop are abandoned. Instead, an inversion arrangement and adaptive feedback correction iteration mechanism based on the principle of cumulative circulation division are introduced to solve the problem of flow field deflection caused by water obstruction of solid components and the induction of secondary vortices. This makes the flow field of the non-uniform arrangement scheme physically self-consistent after the intervention of solid water obstruction.

[0181] Because the iterative optimization process is prone to divergence, the scheme also establishes a physical homogenization double convergence criterion based on the residual density variation coefficient, and forcibly applies a wake space mask filter to shield against spurious vortices generated by the Kármán vortex street. This combination solves the divergence dead loop problem caused by simple numerical approximation, ensuring the robustness of the iterative algorithm and the globally optimal allocation of anti-vortex resources.

[0182] In other embodiments, the specific implementation process of the underlying automated mesh execution engine is described, including the loading mechanism of the globally unified background mesh in the beamless state, the resistance equivalent mapping mechanism of the momentum source term in porous media, and the meshless reconstruction iterative calculation method based on the modification of the activation coordinate distribution by automated scripts. Specifically, this includes:

[0183] Step 601: Load the pre-built global uniform background mesh for the beamless state;

[0184] In numerical simulations of fluid mechanics, traditional physical modeling methods require rebuilding a 3D physical mesh containing the geometric boundaries of the solid components after each change in the spacing distribution sequence of the anti-vortex beams. Since the spacing parameters output in each feedback update are different, frequent mesh reconstruction can easily lead to distortion of discrete mesh nodes or failure of boundary layer mesh generation, causing interruptions in the automated iterative calculation process.

[0185] Therefore, a pre-constructed global uniform background mesh for the beam-free state can be obtained. This global uniform background mesh retains only the solid hydrodynamic boundaries of the inlet, diversion channels, and hydraulic structures' foundations within the entire computational fluid domain. The internal water areas are continuously geometrically discretized using either structured hexahedral meshes or unstructured tetrahedral meshes. Because the physical geometric walls of the solid vortex-eliminating beams are eliminated, this global uniform background mesh maintains a fixed geometric topology in all subsequent feedback update calculations, avoiding flow field divergence caused by frequent changes in geometric boundaries.

[0186] Step 602: Using the equivalent model of porous media with momentum source terms, the physical resistance of the anti-vortex beam to the water flow is equivalently mapped to the corresponding local momentum sink term on the global unified background grid.

[0187] In this embodiment, the physical flow obstruction effect of the solid anti-vortex beam on the water flow is the form resistance and surface friction resistance exerted by the rigid component surface on the fluid micro-particles. To reproduce this physical flow obstruction characteristic in a mesh system without solid geometric boundaries, a porous media equivalent model with momentum source terms is introduced.

[0188] Specifically, in the momentum conservation equation of the Navier-Stokes equations for fluids, adding a negative momentum sink term transforms the local space where solid components were originally placed into a porous medium region with anisotropic permeability resistance properties. The resistance component of the local momentum sink term is calculated using the following formula:

[0189] S _i =-(0.5×ρ×C _d ×A _f ×v _i ×abs(v _i )) / V _cell ;

[0190] Among them, S _i For the local momentum sink term along the specified spatial coordinate axis, ρ is the fluid density, and C is the fluid density. _d A is the dimensionless drag coefficient of the vortex-reducing beam section. _f v is the projected area of ​​the grid cell perpendicular to the flow velocity direction. _i To specify the fluid velocity components along the spatial coordinate axes, abs(v _i V is a mathematical function that takes the absolute value of the corresponding fluid velocity component. _cellThis represents the spatial volume of the corresponding discrete grid cell.

[0191] Building upon this, a numerical drag mechanism is added to the momentum equation. Without altering the underlying physical geometry, the momentum of the fluid flowing through a predetermined grid region is forcibly consumed through equation rewriting. At the flow field scale, the water-blocking effect of the rigid body component and the unsteady flow characteristics around the component are equivalently reconstructed. This equivalent mapping transforms the structural spatial arrangement problem into a problem of assigning attribute matrices to grid nodes.

[0192] Step 603: During feedback updates, the global uniform background grid remains fixed, and an automated script is used to modify the local momentum sink term and the activation coordinate distribution on the global uniform background grid to realize the update calculation of the iterative flow field.

[0193] In a physically adaptive iterative closed-loop system, upon receiving an updated spacing distribution sequence and corresponding initial installation position parameters, control commands are executed to maintain the coordinate attributes of all nodes in the globally uniform background mesh unchanged. The system reads a pre-configured automated script, parses the updated spatial coordinate parameters, and performs a geometric centroid search within the entire three-dimensional fluid computational domain matrix. It then determines whether the spatial centroid coordinates of each independent mesh element fall within the theoretical design space of the vortex-reducing beam defined by the spacing distribution sequence.

[0194] If the judgment result is true, the corresponding independent mesh cell is marked as active, and the local momentum sink term parameter is loaded and written into its fluid momentum equation matrix;

[0195] If the judgment result is false, the corresponding independent grid cell is marked as inactive, and its momentum equation matrix retains the standard control form without additional source terms.

[0196] On the other hand, a specific calculation example of the spatial discrete addressing assignment process is provided. Let X be the center abscissa of a certain anti-vortex beam to be deployed, and B be the corresponding physical parameter of the structural width. An automated script traverses the computational domain mesh to obtain a set of discrete mesh elements whose abscissas are between (X-0.5×B) and (X+0.5×B), and whose corresponding ordinates and vertical elevations are all within the theoretical coverage limits of the component. All nodes within this extracted set of discrete mesh elements are assigned the local momentum sink term parameters of the porous medium equivalent model.

[0197] By updating the numerical boundary parameters of the geometric spatial coordinate determination conditions, the reconstructed calculation of the drag distribution properties of the entire flow field after the spacing change is realized. This automated script modification mechanism decouples the spatial arrangement optimization algorithm layer from the underlying mesh construction layer, compressing the computer time overhead of a single iteration into the numerical solution process of the equation system, thus supporting the computational feasibility of the iterative algorithm under the condition of limited computing power in engineering.

[0198] On the other hand, this paper provides possible implementation methods for the two-dimensional variable spacing extension of double-layer anti-vortex beams, particularly how to utilize two-dimensional spatial distribution index data sources to deduce the independent arrangement parameters of mutually orthogonal double-layer solid components. Specifically, this can be:

[0199] Under extreme operating conditions in some intake projects, relying solely on a single-layer vortex-suppressing component may be insufficient to weaken high-intensity three-dimensional intake vortices. In such complex scenarios, a double-layer vortex-suppressing structure is typically constructed by deploying a first layer of horizontal beams and a second layer of vertical beams. The geometric axes of the first-layer horizontal beams are arranged parallel to the horizontal axis of space, while the geometric axes of the second-layer vertical beams are arranged parallel to the vertical axis of space, with the two layers of components orthogonally stacked in the spatial plane. Because the vortex generation zone at the intake exhibits non-uniform distribution of physical quantities in both the horizontal and vertical spatial dimensions, a spacing arrangement scheme varying along a single dimension cannot achieve a globally optimal configuration of structural interception resources.

[0200] Based on this, the system extracts the two-dimensional comprehensive envelope distribution matrix without one-dimensional projection reduction. This two-dimensional comprehensive envelope distribution matrix inherits the discrete spatial node attributes of the two-dimensional circulation contribution density field calculated based on directed vortex statistics. To balance the consistency of anti-vortex physical properties with the decoupling of the algorithm in the computer optimization process, independent order-reduction spatial integration operations are performed on the two-dimensional comprehensive envelope distribution matrix along both the horizontal and vertical axes.

[0201] For example, the lateral dimension projection density data used for beam layout design can be calculated using the following formula:

[0202] ρ _env_x (x)=integral _y (ρ _env (x,y))dy;

[0203] Where, ρ _env_x (x) represents the calculated lateral projection density at the corresponding abscissa position, ρ _env (x,y) represents the two-dimensional integrated envelope distribution value of the current coordinate node, integral _y This is a spatial integration operation performed along the vertical axis within a preset vertical physical boundary, where dy is the differential element of the vertical axis.

[0204] Accordingly, the longitudinal dimension projection density data used for the longitudinal beam layout design can be calculated simultaneously using the following formula:

[0205] ρ _env_y (y)=integral _x (ρ _env dx (x,y);

[0206] Where, ρ_env_y (y) represents the calculated longitudinal projection density at the corresponding ordinate position, integral _x This refers to the spatial integration operation performed along the horizontal axis within a preset horizontal physical boundary.

[0207] The above process involves performing independent integration in orthogonal directions to extract two one-dimensional discrete variable sequences corresponding to the spatial characteristics of the first-layer horizontal beams and the second-layer vertical beams, respectively. For the iterative calculation of the spacing of the first-layer horizontal beams, the maximum and minimum extreme values ​​in the transverse projection density array are extracted and substituted into the spacing mapping rule equation of the horizontal beam dimension. The local relative spacing of the horizontal beams can be calculated using the following formula, specifically expressed as:

[0208] s _x (x)=d×(R _x_max -(R _x_max -R _x_min )×(ρ _env_x (x)-ρ _min_x ) / (ρ _max_x -ρ _min_x ));

[0209] Among them, s _x (x) represents the local physical spacing of the crossbeams at the x-coordinate, d represents the characteristic diameter of the inlet of the flow field, and R _x_max R is the pre-configured maximum relative beam spacing ratio. _x_min ρ is the pre-configured minimum relative beam spacing ratio. _env_x (x) represents the extracted current lateral projection density value, ρ _max_x ρ is the absolute maximum extremum extracted from the transverse projection density sequence. _min_x The absolute minimum extremum is extracted from the transverse projection density sequence.

[0210] Furthermore, the same computational framework is used to process the second-layer structure. For the spacing calculation of the second-layer longitudinal beams, the maximum and minimum extreme values ​​in the longitudinal projected density array are extracted and substituted into the corresponding longitudinal beam dimensional spacing mapping law equation. The calculated local relative spacing of the longitudinal beams can be described by the following formula:

[0211] s _y (y)=d×(R _y_max -(R _y_max -R _y_min )×(ρ _env_y (y)-ρ _min_y ) / (ρ _max_y -ρ _min_y ));

[0212] Among them, s _y (y) represents the local physical spacing of the longitudinal beams at the ordinate y, R_y_max R is the pre-configured maximum relative longitudinal beam spacing ratio. _y_min ρ is the pre-configured minimum relative longitudinal beam spacing ratio. _env_y (y) represents the extracted current longitudinal projection density value, ρ _max_y ρ is the absolute maximum extremum extracted from the longitudinal projection density sequence. _min_y It is the absolute minimum extreme value extracted from the longitudinal projection density sequence.

[0213] In this implementation scheme, both the derived beam spacing distribution sequence and the longitudinal beam spacing distribution sequence are derived from the same underlying two-dimensional flow field physical data. This algorithmic architecture ensures that the orthogonally arranged double-layer solid components maintain self-consistency at the physical logic level of intercepting three-dimensional spatial circulation, while simultaneously achieving mutual decoupling of the two coordinate dimension mapping relationships at the computer discrete layout execution level. Compared to the strategy of merging the beam spacing into indeterminate parameters and directly inputting them into a high-dimensional coupled iterative network for joint optimization, this reduced-order independent expansion scheme reduces computer memory consumption and computation cycle, while maintaining the consistency of the circulation stripping mechanism and taking into account the computational convenience of engineering deployment.

[0214] As another example, for the discretized arrangement calculation of the first layer of horizontal beams and the second layer of vertical beams, not only can the extreme value linear mapping formula be used, but the principle of cumulative circulation equal division can also be reused to perform the inversion arrangement. In the specific implementation, after obtaining the separated and dimension-reduced horizontal projection density array and vertical projection density array, the background noise matrix is ​​independently injected into the two sets of sequences to construct strictly monotonically increasing horizontal cumulative contribution functions and vertical cumulative contribution functions.

[0215] Based on the pre-configured limits on the total number of horizontal and vertical beams, the target values ​​for horizontal and vertical equal divisions are calculated and extracted separately. In the corresponding horizontal and vertical cumulative function networks, a binary search root-finding inversion operation is performed in parallel. The discrete continuous physical coordinate arrays output by the two independent equal-division inversion root-finding modules directly determine the optimal initial spatial station locations for the horizontal and vertical solid structure groups.

[0216] According to one aspect of this application, after the adaptive feedback correction iterative calculation satisfies the preset dual-criteria convergence condition and outputs the layout scheme, it is necessary to quantitatively determine the actual vortex-reducing performance of the non-uniform spacing configuration. Next, a verification system based on the same total number of components is established to maintain the objectivity and fairness of the flow field performance evaluation. The system extracts the total number of solid vortex-reducing beams in the layout scheme, using this total number of beams as a hard constraint. Within the same preset vortex-reducing beam action range, a physical model or numerical simulation mesh for a traditional equidistant arrangement is constructed.

[0217] Optionally, for the target beam-containing flow field corresponding to the non-uniform spacing arrangement scheme, and the benchmark beam-containing flow field corresponding to the control group, key hydrodynamic evaluation indicators are extracted under all typical operating conditions. These key hydrodynamic evaluation indicators specifically include the average vortex intensity in the vortex region, the circulation constant of the water flow surface, the maximum tangential velocity, and the maximum radial velocity. The discrete hydrodynamic data of the two flow fields in four dimensions can be compared to verify the effectiveness of the solid structure based on the adaptive arrangement of circulation contribution density in intercepting water vortices and reducing local fluid kinetic energy.

[0218] If, under all fortification conditions, all four indicators of the target beam-containing flow field are lower than those of the control group's baseline beam-containing flow field, then the output layout scheme is determined to meet the engineering fortification standards and is saved as construction layout parameters.

[0219] In this invention, the values ​​and ranges of parameters and thresholds can be flexibly adjusted by engineers according to the on-site working conditions, and are not limited to a single value.

[0220] The optional embodiments of the present invention have been described in detail above. However, the present invention is not limited to the specific details of the above embodiments. Within the scope of the technical concept of the present invention, various equivalent transformations can be made to the technical solution of the present invention, and these equivalent transformations all fall within the protection scope of the present invention.

Claims

1. A method for adaptive variable spacing arrangement of vortex-eliminating beams for side-mounted water inlets, characterized in that, include: Acquire flow field data in front of the inlet under multiple operating conditions. The flow field data includes time-series velocity fields of the vortex generation zone under multiple typical operating conditions. Based on the flow field data, vorticity characteristics are extracted, and spatial distribution indicators are calculated to characterize the contribution of each spatial location in the vortex generation zone to the vortex circulation. Envelope processing is performed on spatial distribution indices under multiple typical operating conditions to obtain a comprehensive envelope distribution for the design of vortex-reducing beams. Based on the comprehensive envelope distribution, the initial installation position and spacing distribution sequence of the vortex-eliminating beam are determined within the preset range of action of the vortex-eliminating beam. The flow field containing beams is constructed based on the initial spacing distribution sequence and numerical simulation is performed. The spacing is updated based on the evaluation results of the flow field containing beams. The iteration is terminated when the preset convergence criterion is met, and the adaptive variable spacing arrangement scheme of the vortex-eliminating beams is output. The process involves constructing a beam-containing flow field based on the initial spacing distribution sequence and performing numerical simulations. The spacing is then updated based on the evaluation results of the beam-containing flow field. The iteration terminates when the convergence criterion is met. This includes: extracting the new spacing for the current round based on the beam-containing flow field obtained from each numerical simulation; calculating the relative deviation between the new and old spacings from the previous round; using the relative deviation as the numerical convergence criterion; and if the relative deviation does not meet the convergence threshold, using a relaxation factor to perform a relaxation weighted calculation on the new and old spacings to obtain the updated spacing distribution sequence and proceeding to the next iteration. The process involves constructing a beam-containing flow field based on the initial spacing distribution sequence and performing numerical simulations. The spacing is then updated based on the evaluation results of the beam-containing flow field. The iteration terminates when the convergence criterion is met. This includes: calculating the residual circulation contribution density of each beam interval based on the beam-containing flow field obtained from each numerical simulation; calculating the global average residual density and coefficient of variation of the residual circulation contribution density, using the coefficient of variation as the physical convergence criterion; and when the coefficient of variation does not meet the physical convergence threshold, using an adjustment exponent to perform power-law adaptive adjustment based on the ratio of the residual circulation contribution density of each beam interval to the global average residual density, to obtain the updated spacing distribution sequence.

2. The method according to claim 1, characterized in that, Based on flow field data, vorticity characteristics are extracted, and spatial distribution indices are calculated to characterize the contribution of each spatial location within the vortex generation region to vortex circulation, including: Calculate the instantaneous vorticity field at each time step based on the time-series velocity field in the flow field data; By statistically analyzing the instantaneous vorticity field over the time dimension, we obtain the time-averaged vorticity field and the standard deviation field of vorticity fluctuation. The absolute value of the time-averaged vorticity field is superimposed with the standard deviation of vorticity fluctuation multiplied by the risk coefficient to obtain the two-dimensional circulation contribution density field. By integrating and projecting the two-dimensional circulation contribution density field along the length of the vortex-reducing beam, a one-dimensional circulation contribution density is obtained as a spatial distribution index.

3. The method according to claim 1, characterized in that, Based on flow field data, vorticity characteristics are extracted, and spatial distribution indices are calculated to characterize the contribution of each spatial location within the vortex generation region to vortex circulation, including: The time-averaged vorticity field is calculated based on the flow field data. The surface integral of the time-averaged vorticity field is performed in the vortex generation region to extract the main rotation direction of the vortex in the vortex generation region. Based on this, the instantaneous vorticity fields corresponding to the flow field data are filtered for the same sign, and the vorticity components that are opposite to the main rotation direction of the vortex are filtered out. The remaining vorticity components are converted into non-negative vorticity intensities with the same sign. The intensity of the same vorticity at each time step is statistically analyzed to obtain the time-averaged vorticity field and the standard deviation field of the vorticity fluctuation. The time-averaged eddy field with the same sign is superimposed with the standard deviation field of the same sign eddy fluctuation multiplied by the risk coefficient to obtain the directed circulation contribution density as a spatial distribution indicator.

4. The method according to claim 3, characterized in that, When obtaining the standard deviation field of vorticity fluctuations of the same sign, extract the turbulent kinetic energy and integral scale corresponding to the flow field data; Based on the isotropic turbulence assumption, the ratio of the square root of the turbulent kinetic energy to the integral scale is calculated. The ratio is then multiplied by an empirical constant to reconstruct the standard deviation field of the same sign vorticity fluctuation.

5. The method according to claim 1, characterized in that, Envelope processing is performed on spatial distribution indices under multiple typical operating conditions to obtain a comprehensive envelope distribution for vortex-reducing beam design, including: Extract the pure maximum value of the spatial distribution index of each spatial location under multiple typical operating conditions; Obtain the probability of occurrence of each typical operating condition, and calculate the probability-weighted expected value of the spatial distribution index of each spatial location under multiple typical operating conditions; By using the conservatism coefficient, the pure maximum value and the probability-weighted expected value are linearly weighted and fused to obtain the comprehensive envelope distribution.

6. The method according to claim 1, characterized in that, Based on the comprehensive envelope distribution, the initial installation positions and spacing distribution sequence of the vortex-eliminating beams are determined within the effective range of the vortex-eliminating beams, including: Extract the maximum and minimum distribution values ​​of the comprehensive envelope distribution within the effective range of the vortex-eliminating beam; Based on the physical mapping law that the circulation constant of the water flow surface has a linear relationship with the relative spacing of the vortex-eliminating beams, a non-uniform spacing mapping formula is constructed. Substitute the numerical values ​​of the comprehensive envelope distribution corresponding to each spatial location into the non-uniform spacing mapping formula, and combine the pre-configured minimum relative spacing and maximum relative spacing to calculate the local relative spacing point by point. Based on the starting position of the vortex-eliminating beam's effective range, the initial installation position and spacing distribution sequence are determined by sequentially deduce the arrangement according to the local relative spacing.

7. The method according to claim 1, characterized in that, Based on the comprehensive envelope distribution, the initial installation positions and spacing distribution sequence of the vortex-eliminating beams are determined within the effective range of the vortex-eliminating beams, including: Preset background noise is injected into the comprehensive envelope distribution, and spatial integration is performed within the range of the vortex-eliminating beam to construct a strictly monotonically increasing cumulative circulation contribution function. Obtain the total number of vortex-eliminating beams, and based on the total number of vortex-eliminating beams, divide the total circulation contribution of the cumulative circulation contribution function into equal parts, and calculate the target cumulative value of each vortex-eliminating beam. The target cumulative value of each antivortex beam is divided into equal parts and inverted in the cumulative circulation contribution function to find the root and calculate the initial installation position. The spacing distribution sequence is derived based on the coordinate difference between adjacent initial installation positions.

8. The method according to claim 3, characterized in that, Before performing the surface integral of the time-averaged vorticity field within the vortex generation region, the following steps are also included: Extract the absolute value of the global maximum vorticity of the time-averaged vorticity field; Multiply the absolute value of the global maximum vorticity by the percentage threshold to obtain the background vorticity threshold; The spatial region in the time-averaged vortex field where the absolute value of vortex is greater than the background vortex threshold is defined as the vortex generation region.

Images