A flow field sensing and stress analysis system for a temporary support frame structure of a flood drainage channel

Abstract

This invention discloses a flow field perception and stress analysis system for a temporary support frame structure of a flood discharge channel, relating to the field of flood discharge engineering technology. It constructs a support structure layout model; collects velocity vector field and pressure field data in real time at key node locations to form time-series flow field information; constructs a local disturbance distribution function based on the flow field data and extracts vortex structure and shear strength variation indices; inputs the disturbance characteristics into the node stress estimation model to calculate the instantaneous stress response; judges abrupt force changes and identifies instability trends, establishes a local reinforcement calculation domain, refines the flow field and node response, and iteratively predicts the instability moment; finally, it outputs a time-varying force spectrum and risk distribution map. This invention realizes dynamic analysis and early warning of support structures under flood discharge conditions, improves the accuracy and response efficiency of structural safety assessment, and has good engineering application value.

Description

Technical Field

[0001] This invention relates to the field of flood drainage engineering technology, specifically to a flow field sensing and force analysis system for a temporary support frame structure of a flood drainage channel. Background Technology

[0002] Drainage channels are crucial infrastructure for the rapid discharge of urban water during rainstorms, and their structural safety has a critical impact on urban drainage efficiency and public safety. During the construction and emergency maintenance of drainage channels, temporary support structures (such as steel pipe supports and truss supports) are often erected to maintain the stability of the channel walls or culverts. However, in the event of sudden rainstorms or flash floods, the flood flow velocity changes rapidly, and the local flow field where the support structure is located is easily subject to severe disturbances, forming complex unsteady flow structures such as eddies and shear layers. This causes unexpected stress on the support structure, leading to deformation or even instability and failure.

[0003] Current stress safety assessments of temporary support structures are primarily based on static mechanics or simplified flow field assumptions (such as steady-state uniform flow) and theoretical analysis. These analyses fail to adequately consider the dramatic temporal and spatial variations in the flow field within actual flood discharge environments, particularly in localized areas such as irregular nodes and junctions of the support structure, making it difficult to capture critical flow-induced stress peaks. Furthermore, due to site limitations in the arrangement of temporary support structures, significant force coupling effects exist between their supports, making it impossible for traditional analysis methods to establish a dynamic correlation model between flow field disturbances and local structural stresses. Summary of the Invention

[0004] The purpose of this invention is to provide a flow field sensing and force analysis system for temporary support frame structures of flood drainage channels, so as to overcome the shortcomings of the prior art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a flow field sensing and force analysis system for a temporary support frame structure of a flood drainage channel, comprising:

[0006] The structural modeling module obtains the initial layout parameters of the flood discharge channel section where the support frame structure is located, including the location of each node, the connection relationship of the support members, the material properties and geometric parameters of the members, and constructs the layout model of the support structure.

[0007] The data acquisition module selects a set of key nodes in the support structure layout model and collects the velocity vector field V(t) and pressure field P(t) data at each node location in real time to form a time series flow field information set F(t).

[0008] The disturbance identification module constructs the local flow field disturbance distribution function Φ on the surface of the support frame structure based on F(t), and discretizes Φ into segments according to time sliding windows to extract the main vortex structure and shear strength change index in each segment;

[0009] The stress calculation module inputs the shear strength change index corresponding to each node for the time period into the node stress estimation model, and calculates the instantaneous stress response σi(t) of the node in the time period Δt based on the disturbance distribution, structural connection state and material parameters.

[0010] The mutation detection module obtains the peak stress of all key nodes based on σi(t). Its corresponding time point Determine if there is a region of sudden change in stress;

[0011] Instability trend identification module, if it exists If the growth rate exceeds the preset threshold for two consecutive time periods, it is determined that there is a flow field-induced instability trend at the node, and its position Pi is recorded, and the process enters the local intensification analysis module; otherwise, the process enters the result output module.

[0012] The local reinforcement analysis module establishes a local reinforcement computational domain based on Pi, refines the flow field disturbance function Φ in the region, increases the spatial node density and time sampling frequency, updates σi(t), and iteratively predicts the instability time.

[0013] The output module outputs the time-varying force spectrum of the complete support structure σi(t) and generates a risk distribution map of key nodes.

[0014] Preferably, the construction of the support structure layout model includes:

[0015] Based on the structural cross-sectional drawings of the flood discharge channel and the on-site measurement data, the spatial boundary conditions of the supporting structure layout area were obtained, and a three-dimensional coordinate system was established.

[0016] In a three-dimensional coordinate system, the spatial coordinates of each node are determined according to the engineering layout requirements, and a node set is generated and numbered.

[0017] Based on structural design drawings and material database, the connection type, cross-sectional dimensions and elastic modulus parameters of each support member are obtained and associated with the corresponding nodes;

[0018] A support structure layout model is generated based on node sets and member parameters using finite element modeling.

[0019] Preferably, the local flow field disturbance distribution function Φ on the surface of the support frame structure is constructed based on F(t), including:

[0020] The velocity vector field data and pressure field data at key nodes are mapped onto the surface of the support frame structure according to the spatial location of the nodes. A continuous set of discrete flow field points is obtained through three-dimensional interpolation, so that the set of discrete points can cover the entire area of ​​the structure surface.

[0021] The discrete point set is segmented into sliding time windows over a continuous time period. A Gaussian kernel-based time-domain smoothing method is used to calculate the velocity gradient change and pressure gradient change within each time window, thereby obtaining an intermediate perturbation matrix that characterizes the degree of local perturbation.

[0022] The spatial scale of the structural surface is refined based on the intermediate perturbation matrix. The three-dimensional vortex extraction algorithm is used to identify the main vortex structural features at each location, and the spatial perturbation intensity distribution is constructed by combining the gradient change.

[0023] The temporal disturbance intensity sequence and spatial distribution characteristics are jointly normalized to form a local flow field disturbance distribution function Φ that dynamically changes with spatial coordinates and time.

[0024] Preferably, Φ is discretized into segments according to a time sliding window, and the main vortex structure and shear strength variation indices within each segment are extracted, including:

[0025] Set the sliding time window parameter ΔT and the sliding step size δT, and process the disturbance function Φ in segments according to ΔT;

[0026] In each time period, the three-dimensional vorticity of the velocity vector field corresponding to Φ is calculated, the curl of the velocity field is solved using the central difference scheme, the local rotation core region is identified, and the vortex scale and principal axis direction are determined.

[0027] By combining vorticity intensity with its spatial gradient, a set of characteristic parameters for local vortex intensity is constructed.

[0028] Simultaneously extract the shear rate tensor within the corresponding time period, calculate its principal shear strain rate change, and use it as an indicator of shear strength change.

[0029] Preferably, the shear strength variation index corresponding to each node over a given time period is input into the node stress estimation model, including:

[0030] By spatially correlating the shear strength change index obtained by the key node within the corresponding time period with the disturbance distribution area where the node is located, the local direction and distribution range of the disturbance on the node can be determined.

[0031] Based on the connection topology of the support frame structure, the connection type, spatial angle and geometric parameters of all support members connected to the node are extracted to construct the local structural response feature set of the node;

[0032] By combining the material's elastic modulus, cross-sectional properties, and local structural response characteristic set, a three-dimensional mechanical equilibrium equation set for the nodes within the time interval Δt is constructed using the finite element method based on the generalized Hooke's law.

[0033] Solving the system of equations yields the instantaneous force response σi(t) of the node under disturbance, which includes three components: axial force, shear force, and bending moment.

[0034] Preferably, the peak force of all key nodes is obtained based on σi(t). Its corresponding time point To determine whether there is a region of sudden change in stress, including:

[0035] The force response sequence σi(t) of each key node during the complete analysis period is traversed, and the local peak points are extracted using a sliding extremum search algorithm, with the maximum peak value recorded. and their corresponding time points ;

[0036] By performing differential calculations on the peak change amplitudes within two adjacent sliding time intervals, the force growth rate of the node over a continuous time interval can be obtained. ;

[0037] Set a threshold λ for determining sudden changes in force, when When the value is greater than λ, it is determined that the node has an abnormal upward trend of force within the corresponding time period.

[0038] Nodes that meet the criteria are marked as potential mutation nodes, and their spatial distribution is combined to identify whether there are concentrated stress mutation regions.

[0039] Preferably, if there are If the growth rate exceeds a preset threshold for two consecutive time periods, it is determined that there is a flow field-induced instability trend at the node, and its location Pi is recorded, including:

[0040] The peak stress of the key node in two consecutive time periods. Substitute the force growth rate into the formula, obtain the force growth rate sequence through the difference method, and compare it with the preset threshold item by item to filter out the nodes that meet the growth rate conditions.

[0041] The temporal continuity of the force growth rate sequence of the selected nodes is tested. When the force growth rate is greater than the preset threshold for more than two adjacent time periods, the node is marked as a node with a flow field-induced instability trend.

[0042] Based on the spatial position of the node in the three-dimensional coordinate system and the corresponding perturbation function distribution region Φ, the direction of the local perturbation that leads to the instability trend and the affected structural range are identified.

[0043] The spatial coordinates of nodes that satisfy the instability trend condition are recorded as Pi, and used as input for the subsequent construction of the local reinforcement computational domain.

[0044] Preferably, the local enhancement computational domain is established based on Pi, and the flow field perturbation function Φ is refined again for the region, including:

[0045] A local enhancement computational domain is constructed with node location Pi as the center, and a spherical spatial region with Pi as the center and a radius of a preset value r is selected as the local analysis range;

[0046] The perturbation function Φ is locally reconstructed within the computational domain. A three-dimensional high-order interpolation method is used to increase the spatial node density and the time sampling frequency is increased to twice the original frequency.

[0047] The instantaneous force response of the local critical nodes is recalculated using the refined perturbation function. ;

[0048] Based on the rate of change of the stress response curve and the characteristics of the second derivative, the minimum curvature fitting method is used to iteratively predict the estimated time when the node may become unstable, and the estimated time is used for structural risk early warning output.

[0049] The technical effects and advantages provided by the present invention in the above technical solution are as follows:

[0050] 1. This invention achieves dynamic stress analysis of temporary support structures for flood discharge channels under complex flow environments by introducing real-time flow field disturbance perception, instantaneous stress response modeling of nodes, and instability trend prediction methods. Compared with traditional static or steady-state assumption models, this invention can accurately capture the local vortex influence and shear stress changes of the support structure under actual flood disturbance conditions, significantly improving the accuracy of identifying sudden stress on nodes and potential instability points, and providing a reliable basis for structural safety assessment.

[0051] 2. This invention establishes a closed-loop analysis process from data acquisition and physical modeling to visualization output by constructing a time-varying force spectrum and a risk distribution map of key nodes. This process features high spatiotemporal resolution, adaptability to drastic changes in the flow field, and support for concurrent processing of multiple nodes. It is suitable for structural risk early warning and decision support under various flood discharge channel conditions and has good engineering application value. Attached Figure Description

[0052] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0053] Figure 1 This is a flowchart of the system modules of the present invention. Detailed Implementation

[0054] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0055] For examples, please refer to Figure 1 As shown in this embodiment, a flow field sensing and force analysis system for a temporary support frame structure of a flood drainage channel includes:

[0056] The structural modeling module obtains the initial layout parameters of the flood discharge channel section where the support frame structure is located, including the location of each node, the connection relationship of the support members, the material properties and geometric parameters of the members, and constructs the layout model of the support structure.

[0057] First, based on the structural cross-sectional drawings and on-site measurement data of the area where the drainage channel is located, the boundary of the supporting structure layout area is determined, including parameters such as channel bottom width, sidewall height, channel wall slope, and flow direction. By extracting the above geometric information, a Cartesian three-dimensional coordinate system is established, with the channel bottom centerline set as the origin, the water flow direction as the X-axis, the direction perpendicular to the channel bottom as the Z-axis, and the transverse direction as the Y-axis. This three-dimensional coordinate system serves as the reference basis for subsequent node layout and support member connection modeling.

[0058] Based on the established three-dimensional coordinate system, and according to the arrangement of supporting structures and stress requirements in the engineering layout design drawings, the spatial coordinates of each structural connection node are determined. Nodes are assigned numbers sequentially according to a numbering rule, forming a node set. Node numbers are assigned in ascending natural numbers, starting from 1. The spatial position of each node is recorded in the form of "XYZ", accurate to the millimeter level, to ensure the spatial accuracy of subsequent calculations. The node set is an ordered set of triples, representing the spatial distribution of all key connection points.

[0059] Based on the structural design drawings and a standard library of structural components, the starting and ending nodes connected to each supporting member are extracted to determine their connection relationships, and a connection matrix is ​​constructed. Each record in the connection matrix represents the node numbers connected to a supporting member. Subsequently, material data for the corresponding model is retrieved from the structural material database to obtain the member's cross-sectional shape, cross-sectional area, moment of inertia, elastic modulus, Poisson's ratio, and other physical and mechanical parameters. For each member, its material parameters are bound to its spatial connection relationships to establish a complete structural parameter dataset.

[0060] Using the established node set and support member parameter data, a support structure layout model is constructed using the finite element method. The modeling process employs a member-by-member addition approach, treating each member as a one-dimensional elastic rod element, spatially positioned through its start and end nodes. The entire structure is discretized using three-dimensional spatial beam elements. The element stiffness matrix is ​​calculated using material parameters and geometric dimensions, expressed as follows: the axial stiffness of the element is calculated as the product of the elastic modulus and the cross-sectional area divided by the member length; the bending stiffness is calculated as the product of the elastic modulus and the moment of inertia divided by the cube of the length. These parameters are then assembled to form the global stiffness matrix of the entire structure. The final support structure layout model is a discrete structural model jointly defined by the node set, member connection set, and stiffness parameter matrix.

[0061] The data acquisition module selects a set of key nodes in the support structure layout model and collects the velocity vector field V(t) and pressure field P(t) data at each node location in real time to form a time series flow field information set F(t).

[0062] In the support structure layout model, the set of key nodes is determined according to the following two principles:

[0063] The flow field disturbance is significant at the location of the node, such as on the upstream side of the supporting structure, at irregular connection points, or in the transition region between flow field contraction and expansion.

[0064] The stress states of the nodes are complex, providing representative data for subsequent structural response calculations.

[0065] Following the above principles, at least 30% of the total number of structural nodes are selected as critical nodes. The set of critical nodes is represented as follows: Each node has unique spatial coordinates.

[0066] A flow field sensing unit is deployed at each key node location. This unit includes one set of miniature multi-axis velocity sensors and one set of piezoelectric pressure sensors, all of which are waterproof and shock-resistant. The sensors are embedded in the surface of the support rods and are fixed to the support material structure by bonding or bolting.

[0067] The velocity sensor is used to acquire the velocity components of the fluid in three dimensions, with a sampling frequency set to 100 Hz to meet the dynamic response requirements under rapid flow conditions. The pressure sensor is used to acquire the instantaneous static pressure value acting on the node surface, with a sampling frequency set to 50 Hz, and has an automatic temperature compensation function to improve data stability.

[0068] Using the aforementioned sensors, each key node Ni can obtain three-dimensional velocity vector data V(t) = [vx(t), vy(t), vz(t)] and the corresponding scalar pressure value P(t) at any time t. The collected velocity and pressure data are continuously recorded in time sequence to form the flow field data sequence corresponding to node Ni. ,in The start time of data collection. The data collection ends at the designated time. The flow field data sets of all key nodes constitute the flow field information set of the entire structure, denoted as: , where n is the total number of critical nodes.

[0069] The disturbance identification module constructs a local flow field disturbance distribution function Φ(x,y,z,t) on the surface of the support frame structure based on F(t). Φ is discretized into time sliding windows to extract the main vortex structure and shear strength change index in each segment.

[0070] First, the velocity vector field data and pressure field data collected at key nodes are mapped onto the surface of the support frame structure according to their spatial positions in the three-dimensional coordinate system. In the mapping step, the actual spatial coordinates of each node are used as the interpolation base point. Since the distribution of nodes on the structural surface is discrete, in order to obtain a continuous flow field distribution on the structural surface, this invention uses a three-dimensional interpolation method for data expansion.

[0071] The described three-dimensional interpolation method employs a volume interpolation approach based on inverse distance weighting. The flow field values ​​at each node are weighted according to the reciprocal of their spatial distance from the target point, ensuring that the velocity and pressure values ​​at the interpolation points smoothly decrease as the node distance increases. This ultimately forms a discrete set of flow field points covering the entire surface of the support frame structure. The spatial distribution density of this discrete set is greater than three times the number of nodes to ensure that the spatial resolution meets the computational requirements for vortex extraction.

[0072] After obtaining the spatial discrete point set, time-domain processing is performed on this discrete point set over a continuous time period. This invention sets the sliding time window length to ΔT and the sliding step size to δT, enabling the time window to sequentially cover all flow field acquisition periods. For the velocity and pressure field data within each sliding time window, a time-domain smoothing method based on a Gaussian kernel function is used to calculate the changes in velocity and pressure gradients.

[0073] The Gaussian kernel time-domain smoothing method is implemented as follows: Using the center time of the time window as the base point, all time point data within the time window are weighted according to a time weight constructed in the form of an exponential function, giving higher weights to flow field data closer to the center time. Then, the rate of change of velocity vector components and pressure values ​​in the time direction is calculated using a central difference method, and the rate of change is used as the change in velocity gradient and pressure gradient, respectively. The smoothed gradient features constitute an intermediate perturbation matrix used to characterize the degree of local disturbance.

[0074] The intensity distribution of local disturbances on the structural surface within each time window was obtained using the intermediate perturbation matrix. To further identify the specific flow structure of the perturbation, this invention refines the spatial scale of the support frame structure surface. During the refinement process, the density of discrete points is adaptively increased according to the degree of change in the local curvature of the structural surface and the flow field gradient, so that the refined spatial grid can form a higher resolution in the vortex center region.

[0075] Subsequently, a three-dimensional vortex extraction algorithm was used to identify the main vortex structure in the refined space. The vortex calculation method is as follows: based on the three directional components of the velocity vector field, the vortex vector is obtained by solving for the curl of the velocity field. The magnitude of the vortex vector is used to determine the vortex intensity. The vortex center is determined based on the vortex peak position, and the vortex morphology and scale are identified by combining the principal directions and circumferential distribution of the vortex. By comprehensively analyzing the vortex intensity and gradient changes, a spatial disturbance intensity distribution is constructed, enabling the disturbance intensity to simultaneously reflect local velocity structure changes and pressure changes.

[0076] After completing the spatial perturbation intensity distribution, the perturbation intensities are sequentially arranged across continuous time windows to form a perturbation intensity sequence that varies over time. This invention employs a joint normalization method based on the mapping relationship between minimum and maximum values ​​to perform overall normalization of the perturbation intensity sequence in the time dimension and the perturbation feature distribution in the spatial dimension. The normalization process maps each perturbation value to the range of zero to one by subtracting the minimum perturbation value from the entire sequence and then dividing by the difference between the maximum and minimum perturbation values, thus achieving consistency in the expression of perturbation features across the spatial and temporal dimensions.

[0077] Finally, by recombining the normalized disturbance intensity values ​​according to their corresponding spatial location and time series, a local flow field disturbance distribution function Φ that varies with the three-dimensional spatial coordinates and time is constructed. This function serves as the input basis for subsequent nodal force response calculations.

[0078] First, a sliding time window parameter ΔT is set to define the length of the time segment interval of the perturbation function Φ; a sliding step size parameter δT is set to control the forward movement of the time window on the perturbation time series. In this invention, the value of ΔT ranges from 0.2 seconds to 1 second, and the value of δT is no greater than half of ΔT, preferably 1 / 5 of ΔT, to ensure the continuity of the perturbation characteristics within the time overlap segment.

[0079] The disturbance function Φ is processed using a sliding window method based on the aforementioned parameters, uniformly dividing the entire disturbance time series into multiple overlapping time periods. The data within each time period is treated as an approximate steady-state flow field and participates as an independent data block in subsequent vorticity and shear feature extraction processing.

[0080] Within each time interval, the velocity vector field V corresponding to the perturbation function Φ is extracted. The curl of the velocity field is then solved using a central difference scheme in three-dimensional space, thus obtaining the vorticity vector field ω(x, y, z). This calculation method involves calculating the partial derivatives of the velocity components with respect to spatial coordinates at each spatial point. For example, the partial derivative of the velocity component vx in the x-direction with respect to the y-direction is expressed as... Combine the partial derivatives in the three directions into a curl vector: In the formula, This represents the partial derivative of the velocity component vz in the z-direction with respect to the y-direction. This represents the partial derivative of the velocity component vy in the y-direction with respect to the z-direction.

[0081] In the calculated vortex vector field, locations with higher moduli represent regions with significant local rotation, indicating the potential presence of vortex cores. This invention uses points with vortex moduli greater than twice the average vortex intensity as candidate vortex cores and further filters stable vortices based on the consistency of vortex vector directions. Each identified vortex structure includes its vortex center coordinates, vortex principal axis direction (determined by the local vortex vector direction), and vortex influence radius (the distance determined by vortex decay to its original value 1 / e) as its basic parameters.

[0082] After identifying each vortex structure, its disturbance intensity characteristics are further quantified. This invention uses three factors—vortex modulus, spatial gradient rate of change, and principal axis stability—corresponding to each vortex structure as evaluation indicators to construct a set of local vortex intensity characteristic parameters.

[0083] Eddy modulus: used to measure the rotational speed of a vortex;

[0084] The rate of change of vorticity spatial gradient is determined by calculating the partial derivative of the vorticity modulus with respect to the radial direction, reflecting the changing trend of the vortex boundary.

[0085] Main axis stability: By statistically analyzing the variation amplitude of the main axis of vorticity within multiple time windows, higher stability indicates the persistence of the vortex.

[0086] The parameter set is used to express the perturbation behavior of each vortex in the current time period and to distinguish it from non-rotational perturbation behavior in subsequent shear strength assessment.

[0087] To comprehensively reflect the shear behavior caused by non-rotational flow in a disturbed flow field, this invention simultaneously calculates the shear rate tensor and obtains the principal shear strain rate variation index based on the eigenvalues ​​of the tensor. The shear rate tensor S is the symmetric part of the velocity gradient tensor ∇V, and its calculation method is as follows:

[0088] Find the gradient tensor ∇V of the velocity vector field; divide the sum of the gradient tensor and its transpose by 2 to obtain the shear rate tensor S, i.e.: T is the transpose; then the shear rate tensor S is decomposed into eigenvalues, and the largest principal eigenvalue is taken as the principal shear strain rate εs, which represents the maximum shear deformation rate at that point in that time period.

[0089] By comparing the variation trends of principal shear strain rate at various spatial locations over a continuous time period, a shear strength variation index is constructed. Areas with large variations are identified as risk areas where local stress abrupt changes may occur.

[0090] The stress calculation module inputs the shear strength change index corresponding to each node for the time period into the node stress estimation model, and calculates the instantaneous stress response σi(t) of the node in the time period Δt based on the disturbance distribution, structural connection state and material parameters.

[0091] First, the shear strength change index calculated for the key node within a certain sliding time window is mapped one-to-one with the spatial coordinates of the node in the perturbation function Φ. By retrieving spatial cells containing the node's Ni coordinates within the three-dimensional perturbation field defined by Φ, the direction and magnitude of the perturbation intensity vector in these cells are extracted. This allows for the determination of the local action direction (based on the perturbation vector direction) and the range of its effect on the node (based on the spatial expansion region where the perturbation field intensity exceeds a set threshold). This spatial location association operation provides constraints for the subsequent directionality of force on the node and the force boundary conditions.

[0092] Based on the identified disturbance area, the structural information of all support members connected to node Ni is further extracted. This step includes the following three sub-processes:

[0093] Connection type identification: Based on the construction method of the connection nodes and rods in the support layout model, they are classified into rigid connections, hinged connections, or sliding connections;

[0094] Spatial angle calculation: Using the geometric information of each connecting rod in the three-dimensional coordinate system, calculate the angle between them and the disturbance direction vector to determine the projection effect of the disturbance on each rod;

[0095] Geometric parameter extraction: including member length, cross-sectional shape, cross-sectional area, moment of inertia, etc., used to establish the member stiffness characteristics.

[0096] The above information together constitutes the local structural response feature set of the node, denoted as Ri={connection type, included angle θk, member parameter k}, where k is the number of members connected to the node.

[0097] Combining the aforementioned feature set Ri and the direction of disturbance, a three-dimensional static equilibrium equation for the nodes within the time interval Δt is constructed. This model uses the generalized Hooke's law as its theoretical basis to calculate the stress-strain relationship of each member under disturbance. For each member, let fk(t) be the component of the disturbance projection along the member's axis; then the formula for calculating its internal force is: Where Nk(t) is the axial force, E is the elastic modulus, A is the cross-sectional area, and εk(t) is the axial strain of the member, calculated by dividing the displacement difference between the member's endpoints by the member length. By summarizing the internal forces of all connected members and the projections of external forces caused by disturbances, a three-dimensional equilibrium equation system for node Ni is established, including:

[0098] Force balance in the X direction: ∑Fxk + external disturbance Fx = 0;

[0099] Force balance in the Y direction: ∑Fyk + external disturbance Fy = 0;

[0100] Force balance in the Z direction: ∑Fzk + external disturbance Fz = 0;

[0101] When there are rigid constraints in the connection, a moment equilibrium equation also needs to be introduced.

[0102] By solving the above equilibrium equations, the instantaneous force response σi(t) of node Ni during the time interval Δt under the disturbance is obtained. The force response includes the following three directional components:

[0103] Axial force component: represents the force exerted by the disturbance in the direction of the node link axis;

[0104] Shear force component: Represents the tangential force of the disturbance in the direction perpendicular to the axis;

[0105] Bending moment component: Under rigid connection conditions, the internal moment formed by the tendency of node rotation caused by disturbance.

[0106] The obtained σi(t) serves as the mechanical state index of the node under the current disturbance environment.

[0107] The mutation detection module obtains the peak stress of all key nodes based on σi(t). Its corresponding time point To determine whether there is a region of sudden change in stress.

[0108] For each key node Ni, its instantaneous force response σi(t) is recorded as a continuous time series throughout the complete analysis period. This invention employs a sliding extreme value search algorithm for data processing, which involves setting the sliding window length... (Preferably 0.5 seconds) Move sequentially on the σi(t) sequence, compare the maximum value within each time window, and remove spurious peaks caused by local fluctuations.

[0109] The current maximum force value is recorded in each sliding window, and the peak force value of that node during the entire analysis period is extracted through global comparison. and their corresponding time points This operation ensures the time accuracy of peak extraction and the sensitivity of force response identification.

[0110] In obtaining key nodes and Next, the rate of force change over consecutive time periods is further analyzed. The maximum force value between sliding time windows is calculated using differential methods, and the rate of force increase at node Ni between two consecutive time periods is defined as follows: The specific calculation formula is as follows: ;in The interval between adjacent time periods is preferably set to 0.2 to 0.5 seconds. This indicator is used to quantify the drastic changes in force experienced by a node within a short period of time and is the basis for mutation identification.

[0111] To determine whether a sudden change trend exists at a node, a stress growth rate threshold λ is set, measured in megapascals per second (MPa / s). The value of λ is determined based on the upper limit of the node material strength, the safety margin factor, and the stress variation range under typical flood discharge disturbance conditions. The preferred value range for λ is 5 to 20 MPa / s, which can be dynamically adjusted according to different engineering requirements.

[0112] When the rate of increase of the force at any node Ni If the stress exceeds the preset threshold λ, the system determines that the node has an abnormal upward trend in stress during the current time period, and records the node number, location coordinates and abrupt change time period as input for subsequent reinforcement analysis.

[0113] For the set of nodes marked as potential mutation nodes, their clustering in the spatial distribution of the supporting structure is further analyzed. A three-dimensional spatial neighborhood clustering method is adopted, and a spatial distance threshold d (preferably twice the minimum distance between nodes) is set. All mutation nodes are clustered according to their spatial adjacency to identify whether there are regions where multiple mutation nodes are concentrated.

[0114] If a single cluster contains three or more mutation nodes and the overlap of their stress mutation time periods exceeds 80%, then the spatial region is determined to be a stress mutation region and needs to proceed to the subsequent instability trend prediction process.

[0115] Instability trend identification module, if it exists If the growth rate exceeds the preset threshold for two consecutive time periods, it is determined that there is a flow field-induced instability trend at the node, and its position Pi is recorded, and the process enters the local enhancement analysis module; otherwise, the process enters the result output module.

[0116] To identify potential instability trends at key nodes under the influence of disturbed flow fields, this invention utilizes the instantaneous force responses of each node. and its peak sequence Based on this, by dynamically analyzing the force variation characteristics over a continuous time period, it is determined whether there is a tendency for flow field-induced instability at the nodes. The specific steps are as follows:

[0117] First, select the maximum force value of the key node Ni in two consecutive time periods. and ,in and These are the center times of adjacent sliding time windows, respectively. The force growth rate Gi is calculated using the following formula: The growth rate Gi is measured in megapascals per second (MPa / s) and reflects the rate of change of peak stress at a node over time. This invention sets a threshold λ for assessing instability risk, the value of which is determined based on the actual structure's resistance to disturbances and engineering experience parameters, with a preferred range of 10 to 20 MPa / s. The stress growth rate Gi of each critical node is compared with λ, and nodes with Gi greater than λ are preliminarily identified as nodes that may experience sudden stress changes, proceeding to the next step of time continuity testing.

[0118] A time series analysis of the force growth rate is performed on the selected nodes. Let Gi(t) be the growth rate of node Ni in each consecutive time period. If Gi(t) continuously satisfies the following conditions in two or more adjacent time periods... If so, then node Ni is considered to be in a state of continuous force abrupt change.

[0119] This condition is used to eliminate the influence of occasional force fluctuations, thereby identifying the true instability trend driven by continuous flow field disturbances. Nodes that meet this condition will be marked as nodes with flow field-induced instability trends.

[0120] For each node marked as having an unstable trend, its positional relationship in the perturbation distribution function Φ is further analyzed. First, the position (xi, yi, zi) of node Ni in the three-dimensional spatial coordinates is retrieved, and then the perturbation vector direction and perturbation intensity value corresponding to that position are found in the perturbation function Φ.

[0121] By extracting the main disturbance direction (i.e., the principal axis direction of the disturbance vector field) of the disturbance cell where the node is located, the direction of the current disturbance's effect on the node can be clearly determined. At the same time, the region where the disturbance intensity value exceeds the mean of the disturbance distribution is defined as the "significant disturbance region," and a node falling into this region is considered to be within the main disturbance's influence range.

[0122] The spatial coordinates (xi, yi, zi) of the node Ni that is determined to have a tendency to become unstable due to the flow field are denoted as Pi, and Pi is written into the structural risk identification dataset as the input basis for the subsequent construction of the local reinforcement computational domain.

[0123] The local reinforcement analysis module establishes a local reinforcement computational domain based on Pi, refines the flow field disturbance function Φ in the region, increases the spatial node density and time sampling frequency, updates σi(t), and iteratively predicts the instability time.

[0124] Based on the location coordinates of key nodes A local reinforcement computational domain is constructed centered on this domain. This computational domain is set as a three-dimensional spherical region, with a radius r being an adjustable parameter, preferably ranging from 2 to 4 times the average distance between nodes. Spatial points within the spherical boundary will serve as the target region for local perturbation reconstruction.

[0125] This spherical analysis domain has strong local focusing ability, which can effectively capture high gradient flow regions caused by concentrated disturbances, while avoiding redundant consumption of overall computing resources.

[0126] Within the aforementioned local computational domain, the original perturbation distribution function Φ is locally reconstructed. To improve spatial resolution, a three-dimensional high-order Lagrange interpolation method is used to densify the spatial nodes of the original perturbation data. The interpolation order is preferably 3 or higher to preserve the detailed changes in the boundary of the perturbation structure.

[0127] Meanwhile, the sampling frequency of the perturbation function is increased to twice the original frequency in the time dimension to ensure sufficient time sampling density during the rapid change phase of the perturbation for high-frequency dynamic analysis of the subsequent nodal force response.

[0128] After the above processing, a local perturbation function Φ′ containing high-density spatial points and high-frequency time sampling is obtained, which is used to refine the force changes in the modeling region.

[0129] The refined perturbation function Φ′ is input into the original force calculation model, and the instantaneous force response is recalculated for all key nodes included in the computational domain. .

[0130] In this process, the connection relationships between nodes and the structure, material parameters, and boundary conditions are kept unchanged; only the spatial distribution and intensity data of the disturbance sources are updated, so that the recalculated data are consistent with the original data. It possesses higher time resolution and disturbance sensitivity. The obtained high-precision stress curves are used to identify nonlinear trends in stress changes and potential instability points.

[0131] To determine whether critical nodes will experience structural instability in the future, this invention uses high-resolution stress response curves. A fitting analysis was performed, and the trend of its change was modeled using a method based on the minimum curvature fitting principle.

[0132] This method first calculates The first and second derivatives (rate of force increase) are used to identify sections where the rate of change of the curve continuously increases. Then, the least squares method is used to perform nonlinear fitting on the curve of this section, and the inflection point of the fitted function is extracted as the potential instability moment tc. Inflection point identification is based on... The point of minimum curvature of a curve is defined as the time point corresponding to the minimum curvature of the function, indicating the key position where the trend of the force curve is about to change.

[0133] The final obtained tc is used as the prediction time of potential node instability, and is used to link the output of the structural risk early warning module with the safety response mechanism.

[0134] The output module outputs the time-varying force spectrum of the complete support structure σi(t) and generates a risk distribution map of key nodes.

[0135] This invention collects the instantaneous force response σi(t) of all key nodes Ni to construct a complete time-varying force spectrum data matrix Σ, where each row represents the force change curve of a node within the analysis time range, and each column represents the force state of all nodes at a certain moment.

[0136] This data matrix uses time as the horizontal axis and node number as the vertical axis. Force values ​​are encoded using color intensity or curve height. The visualization output is a two-dimensional heatmap or a three-dimensional force evolution diagram, forming a complete force spectrum. The force spectrum reflects the overall structural response during flow field disturbances, including: the force change trend of each node; the cooperative or coupled force characteristics of multiple nodes; and the time periods when local force extrema occur synchronously. The force spectrum is output with equal time intervals, sampling at least 20 frames per second to ensure the capture of response peaks caused by high-frequency disturbances.

[0137] Based on obtaining σi(t) and the predicted instability time tc, this invention classifies the risk levels and visualizes the spatial distribution of all key nodes. The specific method is as follows:

[0138] Risk level assessment criteria:

[0139] Level 1 risk node: σi(t) shows continuous growth, and the predicted instability time tc is less than 5% of the end of the analysis period;

[0140] Secondary risk node: σi(t) has multiple stress abrupt changes, but no significant upward trend has been formed;

[0141] Level 3 risk node: σi(t) fluctuates little and is stable under stress.

[0142] Risk map construction method: Based on the 3D support structure layout diagram, all key nodes are spatially mapped according to the above risk levels. Each node is identified by three colors: red (high risk), orange (medium risk), and green (low risk), and an instability prediction time label is superimposed.

[0143] For Level 1 risk nodes, the figure also displays parameters such as the maximum stress value σmax(i), growth rate Gi, and the fitted instability time tc, which support the judgment of structural maintenance or emergency response.

[0144] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. A flow field sensing and force analysis system for a temporary support frame structure of a flood drainage channel, characterized in that: include: The structural modeling module obtains the initial layout parameters of the flood discharge channel section where the support frame structure is located, including the location of each node, the connection relationship of the support members, the material properties and geometric parameters of the members, and constructs the layout model of the support structure. The data acquisition module selects a set of key nodes in the support structure layout model and collects the velocity vector field V(t) and pressure field P(t) data at each node location in real time to form a time series flow field information set F(t). The disturbance identification module constructs the local flow field disturbance distribution function Φ on the surface of the support frame structure based on F(t), and discretizes Φ into segments according to time sliding windows to extract the main vortex structure and shear strength change index in each segment; The stress calculation module inputs the shear strength change index corresponding to each node for the time period into the node stress estimation model, and calculates the instantaneous stress response σi(t) of the node in the time period Δt based on the disturbance distribution, structural connection state and material parameters. The mutation detection module obtains the peak stress of all key nodes based on σi(t). Its corresponding time point Determine if there is a region of sudden change in stress; Instability trend identification module, if it exists If the growth rate exceeds the preset threshold in two consecutive time periods, it is determined that there is a flow field-induced instability trend at the node, and its position Pi is recorded, and the local intensification analysis module is entered. Otherwise, proceed to the result output module; The local reinforcement analysis module establishes a local reinforcement computational domain based on Pi, refines the flow field disturbance function Φ in the region, increases the spatial node density and time sampling frequency, updates σi(t), and iteratively predicts the instability time. The output module outputs the time-varying force spectrum of the complete support structure σi(t) and generates a risk distribution map of key nodes.

2. The flow field sensing and force analysis system for a temporary support frame structure of a flood drainage channel according to claim 1, characterized in that: The construction of the support structure layout model includes: Based on the structural cross-sectional drawings of the flood discharge channel and the on-site measurement data, the spatial boundary conditions of the supporting structure layout area were obtained, and a three-dimensional coordinate system was established. In a three-dimensional coordinate system, the spatial coordinates of each node are determined according to the engineering layout requirements, and a node set is generated and numbered. Based on structural design drawings and material database, the connection type, cross-sectional dimensions and elastic modulus parameters of each support member are obtained and associated with the corresponding nodes; A support structure layout model is generated based on node sets and member parameters using finite element modeling.

3. The flow field sensing and force analysis system for a temporary support frame structure of a flood drainage channel according to claim 1, characterized in that: The local flow field disturbance distribution function Φ on the surface of the support frame structure is constructed based on F(t), including: The velocity vector field data and pressure field data at key nodes are mapped onto the surface of the support frame structure according to the spatial location of the nodes. A continuous set of discrete flow field points is obtained through three-dimensional interpolation, so that the set of discrete points can cover the entire area of ​​the structure surface. The discrete point set is segmented into sliding time windows over a continuous time period. A Gaussian kernel-based time-domain smoothing method is used to calculate the velocity gradient change and pressure gradient change within each time window, thereby obtaining an intermediate perturbation matrix that characterizes the degree of local perturbation. The spatial scale of the structural surface is refined based on the intermediate perturbation matrix. The three-dimensional vortex extraction algorithm is used to identify the main vortex structural features at each location, and the spatial perturbation intensity distribution is constructed by combining the gradient change. The temporal disturbance intensity sequence and spatial distribution characteristics are jointly normalized to form a local flow field disturbance distribution function Φ that dynamically changes with spatial coordinates and time.

4. The flow field sensing and force analysis system for a temporary support frame structure of a flood drainage channel according to claim 3, characterized in that: The Φ is discretized into segments using a time sliding window, and the main vortex structure and shear strength variation indices within each segment are extracted, including: Set the sliding time window parameter ΔT and the sliding step size δT, and process the disturbance function Φ in segments according to ΔT; In each time period, the three-dimensional vorticity of the velocity vector field corresponding to Φ is calculated, the curl of the velocity field is solved using the central difference scheme, the local rotation core region is identified, and the vortex scale and principal axis direction are determined. By combining vorticity intensity with its spatial gradient, a set of characteristic parameters for local vortex intensity is constructed. Simultaneously extract the shear rate tensor within the corresponding time period, calculate its principal shear strain rate change, and use it as an indicator of shear strength change.

5. The flow field sensing and force analysis system for a temporary support frame structure of a flood discharge channel according to claim 1, characterized in that: The shear strength variation index corresponding to each node over a given time period is input into the node stress estimation model, including: By spatially correlating the shear strength change index obtained by the key node within the corresponding time period with the disturbance distribution area where the node is located, the local direction and distribution range of the disturbance on the node can be determined. Based on the connection topology of the support frame structure, the connection type, spatial angle and geometric parameters of all support members connected to the node are extracted to construct the local structural response feature set of the node; By combining the material's elastic modulus, cross-sectional properties, and local structural response characteristic set, a three-dimensional mechanical equilibrium equation set for the nodes within the time interval Δt is constructed using the finite element method based on the generalized Hooke's law. Solving the system of equations yields the instantaneous force response σi(t) of the node under disturbance, which includes three components: axial force, shear force, and bending moment.

6. The flow field sensing and force analysis system for a temporary support frame structure of a flood drainage channel according to claim 5, characterized in that: The peak force of all key nodes is obtained based on σi(t). Its corresponding time point To determine whether there is a region of sudden change in stress, including: The force response sequence σi(t) of each key node during the complete analysis period is traversed, and the local peak points are extracted using a sliding extremum search algorithm, with the maximum peak value recorded. and their corresponding time points ; By performing differential calculations on the peak change amplitudes within two adjacent sliding time intervals, the force growth rate of the node over a continuous time interval can be obtained. ; Set a threshold λ for determining sudden changes in force, when When the value is greater than λ, it is determined that the node has an abnormal upward trend of force within the corresponding time period. Nodes that meet the criteria are marked as potential mutation nodes, and their spatial distribution is combined to identify whether there are concentrated stress mutation regions.

7. The flow field sensing and force analysis system for a temporary support frame structure of a flood drainage channel according to claim 6, characterized in that: If it exists If the growth rate exceeds a preset threshold for two consecutive time periods, it is determined that there is a flow field-induced instability trend at the node, and its location Pi is recorded, including: The peak stress of the key node in two consecutive time periods. Substitute the force growth rate into the formula, obtain the force growth rate sequence through the difference method, and compare it with the preset threshold item by item to filter out the nodes that meet the growth rate conditions. The temporal continuity of the force growth rate sequence of the selected nodes is tested. When the force growth rate is greater than the preset threshold for more than two adjacent time periods, the node is marked as a node with a flow field-induced instability trend. Based on the spatial position of the node in the three-dimensional coordinate system and the corresponding perturbation function distribution region Φ, the direction of the local perturbation that leads to the instability trend and the affected structural range are identified. The spatial coordinates of nodes that satisfy the instability trend condition are recorded as Pi, and used as input for the subsequent construction of the local reinforcement computational domain.

8. The flow field sensing and force analysis system for a temporary support frame structure of a flood drainage channel according to claim 7, characterized in that: The local enhancement computational domain is established based on Pi, and the flow field perturbation function Φ is refined again for the region, including: A local enhancement computational domain is constructed with node location Pi as the center, and a spherical spatial region with Pi as the center and a radius of a preset value r is selected as the local analysis range; The perturbation function Φ is locally reconstructed within the computational domain. A three-dimensional high-order interpolation method is used to increase the spatial node density and the time sampling frequency is increased to twice the original frequency. The instantaneous force response of the local critical nodes is recalculated using the refined perturbation function. ; Based on the rate of change of the stress response curve and the characteristics of the second derivative, the minimum curvature fitting method is used to iteratively predict the estimated time when the node may become unstable, and the estimated time is used for structural risk early warning output.

Images