Method for rapidly identifying dynamic load time domain of sluice structure and inverting dynamic response field

Abstract

The invention discloses a flood discharge gate structure dynamic load time domain rapid identification and dynamic response field inversion method comprising the following steps: 1) carrying out a vibration test of a flood discharge gate structure, and obtaining a multi-source vibration response signal of the flood discharge gate structure; 2) constructing a unit impulse response matrix H and a load shape function matrix N; 3) establishing an inverse analysis model X = HNA for dynamic load time domain identification of the flood discharge gate structure; 4) inversely solving the load shape function coefficient matrix A to obtain a dynamic identification load of the flood discharge gate structure; and 5) inputting the identification load as an external load of the sluice structure to obtain an overall dynamic response field of the sluice structure.The method is based on the multi-source actual measurement vibration response of the sluice structure, the improved dynamic load time domain identification method based on the shape function is adopted, the time domain load of the sluice structure can be rapidly and effectively identified, and the identification accuracy is high.

Description

Technical Field

[0001] The present invention belongs to the technical field of sluice structure safety monitoring and performance evaluation, and specifically relates to a method for rapid time-domain identification of dynamic loads and dynamic response field inversion of a flood discharge gate structure. Background Art

[0002] my country leads the world in the scale of its water conservancy and hydropower projects. Water conservancy hubs in southwestern my country, in particular, are characterized by high operating heads, high flow velocities, and high discharge volumes and powers. Consequently, various flow-induced vibration issues have become a growing concern. The enormous energy carried by high-speed water flows can easily cause intense vibrations and even damage to spillway structures. Therefore, comprehensively capturing the dynamic loads and overall dynamic response fields of spillway structures under discharge excitation is crucial for operational safety monitoring and condition assessment. Due to operational constraints, it is often difficult to fully capture the dynamic loads and overall dynamic response fields of spillway structures through direct monitoring. Therefore, identifying the dynamic loads acting on the spillway structure using structural dynamic response signals from a limited number of measurement points, and thereby deriving the overall dynamic response field distribution of the spillway structure, is crucial for comprehensively assessing the vibration safety status of the spillway structures at these hubs. Dynamic load identification is crucial.

[0003] In the field of engineering structures, the main methods for dynamic load identification are frequency domain and time domain methods. The frequency domain method is mostly used to identify steady dynamic loads, but its transformation matrix calculation and sample cutoff length result in low load identification accuracy. Traditional time domain methods, on the other hand, primarily calculate loads by deconvolving the measured structural response with a pre-acquired system impulse response function. However, when the sampling time is long or the sampling frequency is high, the coefficient matrix of the impulse response function becomes very large, resulting in low computational efficiency and making it difficult to achieve real-time, rapid inversion of dynamic loads on large structures.

[0004] The time-domain identification method for dynamic loads based on shape functions can significantly reduce the dimensionality of the impulse function coefficient matrix, improving computational efficiency. The continuity of shape functions can also, to a certain extent, eliminate the effects of noise and enhance the robustness of the inverse analysis process to noise. Its identification accuracy and efficiency are determined by the frequency of the shape function, which can be obtained through spectral analysis of the measured structural response. However, the structure of a floodgate is complex, and to avoid loss of structural vibration characteristics, dynamic response signals at different degrees of freedom are typically acquired simultaneously through multiple measurement points. Therefore, extracting the frequency of the shape function from the multi-source response signals is key to achieving rapid time-domain identification of dynamic loads on the floodgate structure. Summary of the Invention

[0005] The present invention addresses the shortcomings of existing technologies by providing a method for rapid time-domain identification of dynamic loads on a floodgate structure and inversion of its dynamic response field. Based on the multi-source measured vibration responses of the floodgate structure, this method employs multi-source information fusion technology to improve the shape function-based time-domain dynamic load identification method for rapid identification of the floodgate structure's time-domain loads. The method also employs variational mode decomposition (VMD) to decompose the fused measured vibration response signals and determine the analytical frequencies of the shape functions used to identify the loads. Based on the identified loads, the modal superposition method is employed to invert the overall dynamic response field of the floodgate structure. This method provides a novel approach for rapid time-domain identification of loads on floodgate structures and inversion of their dynamic response fields.

[0006] In order to achieve the above-mentioned purpose, the present invention adopts the following technical solutions.

[0007] A method for rapid time-domain identification of dynamic loads and dynamic response field inversion of a floodgate structure includes the following steps:

[0008] Step S1: placing vibration sensors on the flood gate structure, conducting a vibration test of the flood gate structure, and obtaining multi-source vibration response signals of the flood gate structure;

[0009] Step S2: constructing a unit impulse response matrix H and a load shape function matrix N according to the multi-source vibration response signals of the floodgate structure obtained in step S1;

[0010] Step S3: establishing an inverse analysis model X=HNA for time-domain identification of dynamic loads on the floodgate structure based on the unit impulse response matrix H and the load shape function matrix N constructed in step S2;

[0011] Step S4: Based on the established inverse analysis model X=HNA for time-domain identification of dynamic loads of the spillway structure, inversely solve the load shape function coefficient matrix A to obtain the dynamic identification load of the spillway structure;

[0012] Step S5: Using the identified load as an external load input of the flood gate structure to further obtain the overall dynamic response field of the flood gate structure.

[0013] Specifically, the construction of the unit impulse response matrix H in step S2 is based on the finite element analysis method, and the steps are as follows:

[0014] Step S211: establishing a finite element model of the flood gate structure, using the multi-source vibration response signal of the flood gate structure obtained in step S1 as input;

[0015] Step S212: Identify the operating modal parameters of the flood gate structure, and adjust the material parameters of the finite element model of the flood gate structure so that the modal frequency and corresponding vibration shape of the finite element model of the flood gate structure are consistent with the identified structural operating modal frequency and vibration shape;

[0016] Step S213: Perform modal analysis using the adjusted finite element model of the flood gate structure to obtain complete modal information of the flood gate structure and construct a unit impulse response matrix H.

[0017] Specifically, the number of shape functions n′ is determined by the frequency ω of the shape function f Usually, the dynamic load acting on the floodgate structure contains several frequency components. Only when the frequency of the shape function is higher than the highest cutoff frequency of the dynamic load can the load be accurately identified. When the shape function frequency is lower than the highest cutoff frequency of the dynamic load, the identification result has poor accuracy. However, too high a shape function frequency will increase the dimension of the system matrix and greatly reduce the computational efficiency.

[0018] Therefore, how to select the frequency of the shape function is the key to achieving rapid time-domain identification of dynamic loads on the floodgate. For the load to be identified, given that the response of a linear structure depends linearly on the external load, the required shape function frequency can be determined by spectral analysis of the measured structural response:

[0019] Assuming the Fourier transform of the response signal is F(ω), the response content with a frequency less than ω in the measured response is defined as r(ω):

[0020]

[0021] The traditional method of selecting the frequency of the shape function for load identification is to use r(ω c )=95% corresponding frequency ω c As the highest possible frequency of the load, that is, ω c The analysis frequency ω as the shape function required to identify the load f , from which the total number of shape functions n′ can be calculated:

[0022] n′=2(2Tω f +1)=2(2Tω c +1)

[0023] However, this method lacks theoretical basis and is only applicable to a single system response. In dynamic monitoring of floodgate structures, to avoid the loss of dynamic response information, a multi-source vibration signal acquisition method is typically used, with multiple measuring points and multiple orientations. Dynamic response signals acquired at different measuring points contain different structural dynamic information. Therefore, traditional methods for selecting shape function frequencies for load identification are not suitable for selecting the shape function reference frequencies required for dynamic load identification of floodgate structures.

[0024] The present invention fuses the multi-source vibration response signals of the flood discharge gate into a vibration signal containing all response information and having overall characteristics, then obtains the inherent mode functions (IMFs) of each order in the fused response through variational mode decomposition (VMD), and finally determines the shape function frequency according to the energy contribution of each order IMF.

[0025] Specifically, the steps for constructing the load shape function matrix N in step S2 are as follows:

[0026] Step S221: performing energy normalization processing on the multi-source vibration response signals of the flood discharge gate structure obtained in step S1 to ensure that the sampled signals can be analyzed at the same energy level;

[0027] Energy-normalized signal sequence Expressed as:

[0028]

[0029] In the above formula, X(t) is the multi-source vibration response signal sequence of the floodgate structure, X(t) = [x1 x2…x i ];x i is the vibration response signal collected by the i-th sensor; n is the signal sampling length;

[0030] Step S222: Calculate the energy contribution of the response signals at different measurement points at the same time after energy normalization;

[0031] Assume that N vibration sensors are arranged on the floodgate structure, and the signal sampling length of each sensor is n. Let the energy of the jth data collected by sensor i be normalized to Its energy contribution is expressed as:

[0032]

[0033] In the above formula, is x i The mean of S ni is the standard deviation of the vibration response signal collected by the i-th sensor;

[0034] Step S223: Based on the energy contribution of the vibration sampling signals of different measuring points at the same time, the signals after energy normalization of each measuring point are fused to obtain a fused signal.

[0035]

[0036] Step S224: Use VMD algorithm to transform the fused signal Decompose into p IMF components and calculate the energy contribution of each order IMF component

[0037]

[0038] In the above formula, is the variance of a p-order IMF component; is the total variance;

[0039] Step S225: Select the frequency corresponding to the qth order IMF component whose energy contribution is greater than or equal to 95% of the sum of the first qth order IMF components. Shape function frequencies as load identification;

[0040] Step S226: Calculate the number of shape functions required for floodgate structure load identification by determining the shape function frequency. Construct the load shape function matrix N=[N1 N2 … N n′ ].

[0041] Specifically, the inverse analysis model X=HNA for time-domain identification of dynamic loads on the floodgate structure described in step S3 is established as follows:

[0042] Step S31: Assume that the time beam is divided into n equal segments, with a total of n+1 nodes. Since only the deformation of the time beam in the xoy plane needs to be considered, each node has only two shape functions: unit deflection and unit rotation. According to the load shape function matrix N constructed in step S2, the load time series after being discretized according to the sampling frequency f is expressed as:

[0043] P=NA=[N1 N2 … N n′ ][a1 a2 … a n′ ] T

[0044] In the above formula, P=[p1 p2 … p m ] T , m is the number of time steps, assuming the total duration of the load sequence is T, then m = Tf; a n′ is the coefficient of the n′th shape function;

[0045] Step S32: When the number of nodes in the beam element is too large, high-order terms will appear in the shape function, affecting the calculation efficiency. Therefore, the time beam is divided into n two-node elements. Then the shape function matrix N in the jth (j = 1, 2, ..., n' / 2) element is j It can be expressed as:

[0046]

[0047] In the above formula, T l is the sampling time point in the jth unit, l=1,2,…,k, k is the total number of sampling time points in the unit; t j and t j+1are the left and right endpoint moments of the unit respectively;

[0048] Then the shape function matrix N of the entire time beam is composed of the shape function matrices N of all double-node elements j constitute:

[0049]

[0050] Step S33: Based on the Duhamel integral, the response X(t) caused by any dynamic load P(t) in the system is expressed as a linear combination of a series of unit pulse load responses:

[0051]

[0052] The continuous dynamic load function P(t) is approximated by a series of step functions superimposed by rectangular pulses. Let Δt be the discrete sampling time interval, p i′ is the load value to be identified at time t = (i′-1)Δt, the dynamic response convolution relationship of the linear combination can be discretized and converted into a set of linear equations:

[0053]

[0054] Further simplification leads to the inverse analysis model for time-domain identification of dynamic loads on the floodgate structure:

[0055] X=HNA=BA

[0056] Let B = HN be the response matrix of the load shape function, so that the number of columns of the system matrix is reduced from m to n′, and the dimensionality reduction increases with the increase of the sampling sequence length.

[0057] Specifically, the overall dynamic response field of the flood gate structure is further obtained as described in step S4, and the obtained identified load is used as the external load input of the flood gate structure. The overall dynamic response field of the flood gate structure and its distribution law are obtained by positive analysis of the flood gate structure based on the modal superposition method.

[0058] Compared with the prior art, the present invention has the following beneficial effects:

[0059] 1. The method of the present invention is based on the multi-source measured vibration response of the spillway gate structure and adopts an improved dynamic load time-domain identification method based on shape function. It can quickly and effectively identify the time-domain load of the spillway gate structure with high identification accuracy.

[0060] 2. The method of the present invention uses the variational mode decomposition method (VMD) to decompose the fused measured vibration response signal, determine the analysis frequency of the identified load shape function, and based on the identified load, use the modal superposition method to invert the overall dynamic response field of the flood gate structure, providing a new idea for the rapid time domain identification of the flood gate structure load and the inversion of the dynamic response field. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the following briefly introduces the drawings required for use in the embodiments. It should be understood that the following drawings only illustrate certain embodiments of the present disclosure and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other relevant drawings can be obtained based on these drawings without any creative work.

[0062] Figure 1 It is a flow chart of the method for rapid time-domain identification of dynamic loads and dynamic response field inversion of the flood discharge gate structure of the present invention;

[0063] Figure 2 This is a diagram of the signal collection arrangement of the flood discharge gate structure in an embodiment of the present invention;

[0064] Figure 3 Schematic diagram of the operating mode shape of the flood discharge gate structure according to an embodiment of the present invention;

[0065] Figure 4 1 is a schematic diagram of the modal frequencies and vibration shapes of the finite element model of the flood discharge gate structure according to an embodiment of the present invention;

[0066] Figure 5 1 is an equivalent model and a schematic diagram of the action position of the structural load in an embodiment of the present invention;

[0067] Figure 6 2. Schematic diagram of the fusion response of the flood discharge gate structure in an embodiment of the present invention;

[0068] Figure 7 : is a time history line and power spectrum density diagram of the IMF component of the fusion response signal in an embodiment of the present invention;

[0069] Figure 8 1 is a time history line and a power spectrum density diagram for identifying loads in an embodiment of the present invention;

[0070] Figure 9 This is a root mean square cloud diagram of the dynamic displacement of the gate pier in an embodiment of the present invention;

[0071] Figure 10 1 is a time history comparison diagram of the inverted displacement and the measured displacement in an embodiment of the present invention;

[0072] Figure 11 3 is a comparison diagram of the inverted displacement and the measured displacement root mean square in an embodiment of the present invention. DETAILED DESCRIPTION

[0073] In order to facilitate those skilled in the art to understand and implement the present invention, each step of the method proposed in the present invention is described in detail below. It should be understood that these embodiments are only used to illustrate the present invention and are not intended to limit the scope of the present invention. In addition, it should be understood that after reading the content taught by the present invention, those skilled in the art can make various changes or modifications to the present invention, and these equivalent forms also fall within the scope defined by the claims appended hereto.

[0074] Example

[0075] The present invention discloses a method for rapid time-domain identification of dynamic loads of a flood discharge gate structure and inversion of a dynamic response field, comprising the following steps:

[0076] Step S1: placing vibration sensors on the flood gate structure, conducting a vibration test of the flood gate structure, and obtaining multi-source vibration response signals of the flood gate structure;

[0077] Step S2: constructing a unit impulse response matrix H and a load shape function matrix N according to the multi-source vibration response signals of the floodgate structure obtained in step S1;

[0078] Step S3: establishing an inverse analysis model X=HNA for time-domain identification of dynamic loads on the floodgate structure based on the unit impulse response matrix H and the load shape function matrix N constructed in step S2;

[0079] Step S4: Based on the established inverse analysis model X=HNA for time-domain identification of dynamic loads of the spillway structure, inversely solve the load shape function coefficient matrix A to obtain the dynamic identification load of the spillway structure;

[0080] Step S5: Using the identified load as an external load input of the flood gate structure to further obtain the overall dynamic response field of the flood gate structure.

[0081] Specifically, the construction of the unit impulse response matrix H in step S2 is based on the finite element analysis method, and the steps are as follows:

[0082] Step S211: establishing a finite element model of the flood gate structure, using the multi-source vibration response signal of the flood gate structure obtained in step S1 as input;

[0083] Step S212: Identify the operating modal parameters of the flood gate structure, and adjust the material parameters of the finite element model of the flood gate structure so that the modal frequency and corresponding vibration shape of the finite element model of the flood gate structure are consistent with the identified structural operating modal frequency and vibration shape;

[0084] Step S213: Perform modal analysis using the adjusted finite element model of the flood gate structure to obtain complete modal information of the flood gate structure and construct a unit impulse response matrix H.

[0085] Specifically, the number of shape functions n′ is determined by the frequency ω of the shape function f Usually, the dynamic load acting on the floodgate structure contains several frequency components. Only when the frequency of the shape function is higher than the highest cutoff frequency of the dynamic load can the load be accurately identified. When the shape function frequency is lower than the highest cutoff frequency of the dynamic load, the identification result has poor accuracy. However, too high a shape function frequency will increase the dimension of the system matrix and greatly reduce the computational efficiency.

[0086] Therefore, how to select the frequency of the shape function is the key to achieving rapid time-domain identification of dynamic loads on the floodgate. For the load to be identified, given that the response of a linear structure depends linearly on the external load, the required shape function frequency can be determined by spectral analysis of the measured structural response:

[0087] Assuming the Fourier transform of the response signal is F(ω), the response content with a frequency less than ω in the measured response is defined as r(ω):

[0088]

[0089] The traditional method of selecting the frequency of the shape function for load identification is to use r(ω c )=95% corresponding frequency ω c As the highest possible frequency of the load, that is, ω c The analysis frequency ω as the shape function required to identify the load f , from which the total number of shape functions n′ can be calculated:

[0090] n′=2(2Tω f +1)=2(2Tω c +1)

[0091] However, this method lacks theoretical basis and is only applicable to a single system response. In dynamic monitoring of floodgate structures, to avoid the loss of dynamic response information, a multi-source vibration signal acquisition method is typically used, with multiple measuring points and multiple orientations. Dynamic response signals acquired at different measuring points contain different structural dynamic information. Therefore, traditional methods for selecting shape function frequencies for load identification are not suitable for selecting the shape function reference frequencies required for dynamic load identification of floodgate structures.

[0092] The present invention fuses the multi-source vibration response signals of the flood discharge gate into a vibration signal containing all response information and having overall characteristics, then obtains the inherent mode functions (IMFs) of each order in the fused response through variational mode decomposition (VMD), and finally determines the shape function frequency according to the energy contribution of each order IMF.

[0093] Specifically, the steps for constructing the load shape function matrix N in step S2 are as follows:

[0094] Step S221: performing energy normalization processing on the multi-source vibration response signals of the flood discharge gate structure obtained in step S1 to ensure that the sampled signals can be analyzed at the same energy level;

[0095] Energy-normalized signal sequence Expressed as:

[0096]

[0097] In the above formula, X(t) is the multi-source vibration response signal sequence of the floodgate structure, X(t) = [x1 x2…x i ];x i is the vibration response signal collected by the i-th sensor; n is the signal sampling length;

[0098] Step S222: Calculate the energy contribution of the response signals at different measurement points at the same time after energy normalization;

[0099] Assume that N vibration sensors are arranged on the floodgate structure, and the signal sampling length of each sensor is n. Let the energy of the jth data collected by sensor i be normalized to Its energy contribution is expressed as:

[0100]

[0101] In the above formula, is x i The mean of S ni is the standard deviation of the vibration response signal collected by the i-th sensor;

[0102] Step S223: Based on the energy contribution of the vibration sampling signals of different measuring points at the same time, the signals after energy normalization of each measuring point are fused to obtain a fused signal.

[0103]

[0104] Step S224: Use VMD algorithm to transform the fused signal Decompose into p IMF components and calculate the energy contribution of each order IMF component

[0105]

[0106] In the above formula, is the variance of a p-order IMF component; is the total variance;

[0107] Step S225: Select the frequency corresponding to the qth order IMF component whose energy contribution is greater than or equal to 95% of the sum of the first qth order IMF components. Shape function frequencies as load identification;

[0108] Step S226: Calculate the number of shape functions required for floodgate structure load identification by determining the shape function frequency. Construct the load shape function matrix N=[N1 N2 … N n′ ].

[0109] Specifically, the inverse analysis model X=HNA for time-domain identification of dynamic loads on the floodgate structure described in step S3 is established as follows:

[0110] Step S31: Assume that the time beam is divided into n equal segments, with a total of n+1 nodes. Since only the deformation of the time beam in the xoy plane needs to be considered, each node has only two shape functions: unit deflection and unit rotation. According to the load shape function matrix N constructed in step S2, the load time series after being discretized according to the sampling frequency f is expressed as:

[0111] P=NA=[N1 N2 … N n′ ][a1 a2 … a n′ ] T

[0112] In the above formula, P=[p1 p2 … p m ] T , m is the number of time steps, assuming the total duration of the load sequence is T, then m = Tf; a n′ is the coefficient of the n′th shape function;

[0113] Step S32: When the number of nodes in the beam element is too large, high-order terms will appear in the shape function, affecting the calculation efficiency. Therefore, the time beam is divided into n two-node elements. Then the shape function matrix N in the jth (j = 1, 2, ..., n' / 2) element is j It can be expressed as:

[0114]

[0115] In the above formula, T l is the sampling time point in the jth unit, l=1,2,…,k, k is the total number of sampling time points in the unit; t j and t j+1 are the left and right endpoint moments of the unit respectively;

[0116] Then the shape function matrix N of the entire time beam is composed of the shape function matrices N of all double-node elements j constitute:

[0117]

[0118] Step S33: Based on the Duhamel integral, the response X(t) caused by any dynamic load P(t) in the system is expressed as a linear combination of a series of unit pulse load responses:

[0119]

[0120] The continuous dynamic load function P(t) is approximated by a series of step functions superimposed by rectangular pulses. Let Δt be the discrete sampling time interval, p i′ is the load value to be identified at time t = (i′-1)Δt, the dynamic response convolution relationship of the linear combination can be discretized and converted into a set of linear equations:

[0121]

[0122] Further simplification leads to the inverse analysis model for time-domain identification of dynamic loads on the floodgate structure:

[0123] X=HNA=BA

[0124] Let B = HN be the response matrix of the load shape function, so that the number of columns of the system matrix is reduced from m to n′, and the dimensionality reduction increases with the increase of the sampling sequence length.

[0125] Specifically, the overall dynamic response field of the flood gate structure is further obtained as described in step S4, and the obtained identified load is used as the external load input of the flood gate structure. The overall dynamic response field of the flood gate structure and its distribution law are obtained by positive analysis of the flood gate structure based on the modal superposition method.

[0126] The technical effect of the method of the present invention is further illustrated below through an actual engineering case.

[0127] 1. Project Overview

[0128] like Figure 2 As shown in (a), a large hydropower station in my country consists of concrete non-overflow dams on both banks, surface overflow dams, bottom overflow dams, and riverbed powerhouses. The maximum dam height is 48.80 m, and the total storage capacity is 475 million m3. 3 Among them, the surface overflow dam is arranged with five flood discharge gates from H1 to H5 from left to right. Figure 2 As shown in (b), each spillway has a net width of 12m. The bottom adopts a WES weir design, with a crest elevation of 47.0m and a pier thickness of 3.0m. Seven measuring points were set at the tops of piers 2#, 3#, 4#, and 5# (at 65.5m), corresponding to stake numbers 0+38.9m, 0+28.5m, 0+26.9m, 0+24.4m, 0+21.8m, 0+16.9m, and 0+11.8m.

[0129] like Figure 2As shown in (c), a total of 28 horizontal dynamic displacement sensors are arranged on the four gate piers, numbered D1 to D28, of which D1, D2, D8, D9, D15, D16, D22, and D23 are located at the bottom of the gate room (61.5m). The sensor is a BY-S07 sensor with a frequency response range of 0.25 to 100Hz and a sensitivity of 2.4v·s / m. The signal acquisition and processing system is a DASP analysis system, which focuses on dynamic displacement and automatically converts the velocity signal into a displacement signal with a sampling frequency of 200Hz. During on-site sampling, the upstream water level was 56.02m, the downstream water level was 38.5m, the 2# and 4# surface holes were fully open, and the flow rate of each hole was 650m 3 / s.

[0130] 2. Establishment of accurate finite element model

[0131] The dynamic response signals of 28 measuring points are used as system input to identify the operational modal parameters of the floodgate structure. The first four modal frequencies of the floodgate structure identified are 3.88Hz, 4.20Hz, 4.53Hz, and 4.80Hz. Figure 3 Shown are the mode shapes corresponding to each frequency order.

[0132] According to the engineering survey and design data, a 1:1 finite element model of the floodgate structure was established in the finite element software. The foundation was a massless foundation (density 0.001kg / m 3 ), the foundation simulation range is 55.5m deep, 115m long upstream and downstream, and 153m wide left and right. The upstream side of the gate pier is equipped with a gantry track beam, a traffic bridge, a pipeline beam, etc., which are set to friction contact with the gate pier, and the friction coefficient is 0.3. Fixed constraints are used at the bottom of the foundation, and normal constraints are used around the foundation. The modified Westergaard formula is used to simulate the effect of reservoir water by applying additional mass on the overflow weir and the gate pier. The finite element model is mainly divided into hexahedral grids, with a total of 223,973 units and 280,110 nodes, such as Figure 4 By adjusting the material parameters of the finite element model of the floodgate structure, the modal frequencies and corresponding vibration modes of the finite element model of the floodgate structure are basically consistent with the identified structural operating modal frequencies and vibration modes. The material parameters of the finite element model of the floodgate structure are shown in Table 1 below.

[0133] Table 1. Material parameters of the finite element model of the floodgate structure

[0134]

[0135]

[0136] 3. Rapid identification of dynamic loads in the time domain and inversion of dynamic response fields of flood discharge gate structures

[0137] During the flood discharge process, the vibration of the structure is mainly induced by the pulsation of the high-speed water flow passing through the gate. The pulsation of the water flow passing through the gate is mainly divided into three parts, namely the pulsating pressure acting on the bottom plate, the left gate pier and the right gate pier. The pulsating load of the water flow on the gate chamber structure of a single flood discharge gate can be equivalent to two lateral loads acting on the gate piers on both sides and one load acting on the bottom plate of the overflow weir. Under this test condition, if Figure 5 As shown in Figure 1, when the two gates release water at the same time, the vibration of the flood discharge gate structure can be regarded as the result of the combined action of six equivalent loads P1 to P6.

[0138] The dynamic response signals from 20 measuring points were used for load inversion, with the dynamic response signals from 8 measuring points reserved as a control group to verify the inversion results. To ensure the authenticity and comprehensiveness of the control group, the 8 reserved measuring points were selected based on the principle of selecting 2 for each pier, and all 8 measuring points must be located across all pile numbers. Therefore, the selected measuring points for the verification group were: D1, D5, D11, D14, D16, D20, D24, and D28.

[0139] After normalizing the dynamic response signals of the 20 measurement points other than the reserved measurement points, the 20 groups of displacement responses are fused using a data-level fusion method based on vibration energy contribution, such as Figure 6 The fusion result of the 20 sets of displacement responses is shown in Figure 2. The fused displacement responses are then decomposed by VMD, as shown in Figure 2. Figure 7 As shown, a total of 6-order IMF components are obtained, and the vibration energy contribution of each order IMF is shown in Table 2 below.

[0140] Table 2. Vibration energy contribution of each order IMF component of the fusion response signal

[0141]

[0142] From the VMD decomposition results, it can be seen that the first two order IMF components have occupied the main energy of the fusion response (cumulative vibration energy contribution = 99.518%> 95%). Therefore, the frequency 12.5Hz corresponding to the second order IMF component is selected as the analysis frequency of the shape function required for load identification. The number of shape functions n′=1002 is calculated, and the shape function matrix N is constructed based on this for load identification, which is as follows: Figure 8 Load identification results are shown.

[0143] Then, the identified equivalent load is applied to the structure to perform structural dynamics forward analysis and calculation, and the dynamic response field of the entire floodgate structure can be obtained. Figure 9 The displacement response field of 2#~5# piers is shown. Figure 9It can be seen that under discharge excitation, the displacement response distribution of the flood discharge gate piers shows that the vibration amplitude is small in the areas with lateral constraints, such as the areas near the weir surface, traffic bridge, and working bridge; while the downstream area of the pier top lacks lateral constraints and has additional mass due to the hoisting and closing machine room, so the vibration is more intense.

[0144] In order to verify the effectiveness of the proposed method for rapid identification of dynamic loads in the time domain of the floodgate, the displacement response data obtained from the positive analysis of the floodgate structure dynamics at 28 measuring points were extracted and compared with the measured data. Figure 10 The time histories of the eight measurement points in the validation group are shown. The correlation coefficients between the inverted displacements and the measured displacements at each measurement point are shown in Table 3 below (“*” indicates a measurement point in the validation group, the same below).

[0145] Table 3. Correlation coefficients between inverted displacement and measured displacement

[0146]

[0147] The RMS values between the inverted displacement and the measured displacement at each measuring point are as follows: Figure 11 and as shown in Table 4 below.

[0148] Table 4. RMS values of inverted displacement and measured displacement

[0149]

[0150]

[0151] From the correlation coefficient comparison results, it can be seen that, except for the D11 and D14 measuring points, the correlation coefficients between the inverted displacement and the measured displacement of the other measuring points are all higher than 0.9, indicating that the correlation between the two is high. From the root mean square value comparison results, it can be seen that the energy distribution law of the inverted displacement is consistent with that of the measured displacement, that is, the root mean square value gradually decreases from downstream to upstream. Except for measuring points D10 and D11, the other points with large root mean square value errors are all located at the upstream measuring points. This is because the vibration amplitude on the upstream side of the gate pier is small and the energy is low, so it is more sensitive to errors. The reason for the large root mean square value errors of measuring points D10 and D11 may be that the measured response signal is interfered by noise. Overall, the inverted displacement is relatively consistent with the measured displacement, which verifies that the method proposed in the present invention can realize the time domain rapid and effective identification of the dynamic load of the flood discharge gate structure and the inversion of the dynamic response field, and has high accuracy.

[0152] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any other manner. Any person skilled in the art may utilize the above-disclosed technical content to modify or modify the present invention into equivalent embodiments. However, any simple modifications, equivalent variations, and modifications to the above embodiments that do not depart from the technical content of the present invention and are based on the technical essence of the present invention remain within the scope of protection of the present invention.

Claims

1. A method for rapid time-domain identification of dynamic loads and dynamic response field inversion of a floodgate structure, characterized in that: The following steps are involved: Step S1: placing vibration sensors on the flood gate structure, conducting a vibration test of the flood gate structure, and obtaining multi-source vibration response signals of the flood gate structure; Step S2: constructing a unit impulse response matrix H and a load shape function matrix N according to the multi-source vibration response signals of the floodgate structure obtained in step S1; Step S3: establishing an inverse analysis model X=HNA for time-domain identification of dynamic loads on the floodgate structure based on the unit impulse response matrix H and the load shape function matrix N constructed in step S2; Step S4: Based on the established inverse analysis model X=HNA for time-domain identification of dynamic loads of the spillway structure, inversely solve the load shape function coefficient matrix A to obtain the dynamic identification load of the spillway structure; Step S5: Using the identified load as an external load input of the flood gate structure to further obtain the overall dynamic response field of the flood gate structure.

2. A method for rapid time-domain identification of dynamic loads and dynamic response field inversion of a flood discharge gate structure according to claim 1, characterized in that: The construction of the unit impulse response matrix H in step S2 is based on the finite element analysis method, and the specific steps are as follows: Step S211: establishing a finite element model of the flood gate structure, using the multi-source vibration response signal of the flood gate structure obtained in step S1 as input; Step S212: Identify the operating modal parameters of the flood gate structure, and adjust the material parameters of the finite element model of the flood gate structure so that the modal frequency and corresponding vibration shape of the finite element model of the flood gate structure are consistent with the identified structural operating modal frequency and vibration shape; Step S213: Perform modal analysis using the adjusted finite element model of the flood gate structure to obtain complete modal information of the flood gate structure and construct a unit impulse response matrix H.

3. The method for rapid time-domain identification of dynamic loads and dynamic response field inversion of a flood discharge gate structure according to claim 1 is characterized in that: The steps for constructing the load shape function matrix N in step S2 are as follows: Step S221: performing energy normalization processing on the multi-source vibration response signals of the flood discharge gate structure obtained in step S1 to ensure that the sampled signals can be analyzed at the same energy level; Energy-normalized signal sequence Expressed as: In the above formula, X(t) is the multi-source vibration response signal sequence of the floodgate structure, X(t) = [x1 x2…x i ];x i is the vibration response signal collected by the i-th sensor; n is the signal sampling length; Step S222: Calculate the energy contribution of the response signals at different measurement points at the same time after energy normalization; Assume that N vibration sensors are arranged on the floodgate structure, and the signal sampling length of each sensor is n. Let the energy of the jth data collected by sensor i be normalized to Its energy contribution is expressed as: In the above formula, is x i The mean of S ni is the standard deviation of the vibration response signal collected by the i-th sensor; Step S223: Based on the energy contribution of the vibration sampling signals of different measuring points at the same time, the signals after energy normalization of each measuring point are fused to obtain a fused signal. Step S224: Use VMD algorithm to transform the fused signal Decompose into p IMF components and calculate the energy contribution of each order IMF component In the above formula, is the variance of a p-order IMF component; is the total variance; Step S225: Select the frequency corresponding to the qth order IMF component whose energy contribution is greater than or equal to 95% of the sum of the first qth order IMF components. Shape function frequencies as load identification; Step S226: Calculate the number of shape functions required for floodgate structure load identification by determining the shape function frequency. Construct the load shape function matrix N=[N1N2…N n′ ].

4. The method for rapid time-domain identification of dynamic loads and dynamic response field inversion of a flood discharge gate structure according to claim 1 is characterized in that: The steps for establishing the inverse analysis model X=HNA for time-domain identification of dynamic loads on the floodgate structure in step S3 are as follows: Step S31: Assume that the time beam is divided into n equal segments, with a total of n+1 nodes. Since only the deformation of the time beam in the xoy plane needs to be considered, each node has only two shape functions: unit deflection and unit rotation. According to the load shape function matrix N constructed in step S2, the load time series after being discretized according to the sampling frequency f is expressed as: P=NA=[N1 N2…N n′ ][a1 a2a n′ ] T In the above formula, P=[p1 p2…p m ] T , m is the number of time steps, assuming the total duration of the load sequence is T, then m = Tf; a n′ is the coefficient of the n′th shape function; Step S32: When the number of nodes in the beam element is too large, high-order terms will appear in the shape function, affecting the calculation efficiency. Therefore, the time beam is divided into n two-node elements. Then the shape function matrix N in the jth (j = 1, 2, ..., n' / 2) element is j It can be expressed as: In the above formula, T l is the sampling time point in the jth unit, l=1,2,…,k, k is the total number of sampling time points in the unit; t j and t j+1 are the left and right endpoint moments of the unit respectively; Then the shape function matrix N of the entire time beam is composed of the shape function matrices N of all double-node elements j constitute: Step S33: Based on the Duhamel integral, the response X(t) caused by any dynamic load P(t) in the system is expressed as a linear combination of a series of unit pulse load responses: The continuous dynamic load function P(t) is approximated by a series of step functions superimposed by rectangular pulses. Let Δt be the discrete sampling time interval, p i′ is the load value to be identified at time t = (i′-1)Δt, the dynamic response convolution relationship of the linear combination can be discretized and converted into a set of linear equations: Further simplification leads to the inverse analysis model for time-domain identification of dynamic loads on the floodgate structure: X=HNA=BA Let B = HN be the response matrix of the load shape function, so that the number of columns of the system matrix is reduced from m to n′, and the dimensionality reduction increases with the increase of the sampling sequence length.

5. The method for rapid time-domain identification of dynamic loads and dynamic response field inversion of a floodgate structure according to claim 1 is characterized in that: In step S4, the overall dynamic response field of the flood gate structure is further obtained, and the obtained identified load is used as the external load input of the flood gate structure. The overall dynamic response field of the flood gate structure and its distribution law are obtained by positive analysis of the flood gate structure based on the modal superposition method.

Images