Distributed dynamic load identification method and system based on time variable separation

Abstract

The invention discloses a distributed dynamic load identification method and system based on time variable separation. According to the method, distributed loads are divided into a distributed load function capable of directly separating space-time variables and a distributed load function in a general form, time-domain loads are converted into frequency-domain loads under two conditions based on orthogonal polynomial fitting one-dimensional and two-dimensional function theories, and the time-domain loads are converted into frequency-domain loads. A Chebyshev orthogonal polynomial is used to complete coefficient identification of the orthogonal polynomial in a frequency domain, and a time domain-frequency domain Fourier transform theory and a least square method are combined to further complete distribution time domain dynamic load identification of the continuous beam. According to the method, a large amount of discrete time domain data is not needed for modeling, and the identified distributed load can be directly and clearly represented; only the column vector matrix needs to be identified, so that the identification efficiency is improved and the identification precision is ensured on the basis of greatly reducing identification calculation data.

Description

Technical Field

[0001] In the field of vibration analysis, the input part, the output part and the system characteristic part together constitute a complete vibration system. Finding the third parameter under the premise of knowing two parameters of the vibration system constitutes three important research contents of structural dynamics. The present invention belongs to the field of dynamic load identification, and finds the input load under the condition of known output response and system characteristics, and specifically relates to a distributed dynamic load identification method and system based on time variable separation. Background Art

[0002] In engineering applications, the impact of loads on structures cannot be ignored. Traditionally, most loads are considered as concentrated loads for analysis and identification, but in actual situations, the loads acting on structures often present distributed characteristics. The existence of these distributed dynamic loads makes the identification of dynamic responses more complicated and difficult. The identification of distributed dynamic loads is an inverse problem, and its solution is of great significance to the safety and reliability of structures. The existing technology in this field mainly focuses on the dynamic identification of concentrated loads, while there are relatively few studies on distributed loads, resulting in the immaturity of related technical means. As the second type of inverse problem research in vibration problems, dynamic load identification started late and is more difficult, but it is becoming more and more important in actual engineering problems and is a difficult problem that needs to be solved urgently. From the end of the 20th century to the present, the theory of dynamic load identification has gradually matured, among which the time domain method and the frequency domain method have been widely used. The time domain method is generally suitable for loads with a short action time, and the identification accuracy after time discretization is poor; the frequency domain method started early and developed maturely, and is suitable for identifying excitations for a long time, but the accuracy is insufficient when dealing with short-term excitations.

[0003] The main deficiencies of the existing technology are: the frequency domain method cannot represent the distributed load as clearly and directly as the time domain method; the time domain method needs to process a large amount of discrete time data, the matrix is ​​too large, and it is easy to cause the matrix ill-posed problem, and the fitting form of the distributed load also requires a sufficient number of orders, that is, a large coefficient matrix to complete the identification of the distributed load. Both require the calculation of a large amount of data, which seriously leads to the problem of low calculation efficiency. Therefore, there is an urgent need for a new method that can not only clearly and intuitively represent the distributed load, but also reduce the calculation data and improve the identification efficiency under the premise of ensuring the identification accuracy. Summary of the invention

[0004] Purpose of the invention: In view of the shortcomings of the prior art, the present invention proposes a distributed dynamic load identification method and system based on time variable separation, which greatly reduces the identification calculation data, improves the identification efficiency, and also ensures the identification accuracy.

[0005] Technical solution: To achieve the above purpose, the present invention adopts the following technical solution:

[0006] A distributed dynamic load identification method based on time variable separation includes the following steps:

[0007] Step 1: According to the distributed load f(x, t) loaded on the identification object, it is divided into a distributed load that can directly separate the time variable and a distributed load in a general form. For the distributed load in the general form, the time variable and the position variable are separated by the principle of two-dimensional Chebyshev orthogonal polynomial fitting function. For the distributed load that can directly separate the time variable, the position function is fitted by the principle of one-dimensional Chebyshev orthogonal polynomial fitting function;

[0008] Step 2: According to the form of distributed load, for the distributed load formula with separated time variables, select appropriate order position orthogonal polynomials to fit the frequency domain force F(x,ω), and calculate the frequency domain displacement response matrix W and frequency response function H respectively;

[0009] Step 3: Substitute the fitted frequency domain force into the displacement response and excitation relationship of the continuous system, set the integral matrix of the frequency response function and the orthogonal polynomial as the dynamic calibration matrix HT, obtain the simplified formula W = (HT)(GA), and obtain the coefficient matrix GA in the frequency domain;

[0010] Step 4: According to the form of distributed load classification, for loads that can directly separate time variables, perform Fourier transform on the time variable function g(t) at a fixed frequency to obtain the coefficients For general loads, choose a time orthogonal polynomial of appropriate order Similarly, the coefficient matrix is ​​obtained according to Fourier transform at a fixed frequency q is the order of the time orthogonal polynomial;

[0011] Step 5: According to the coefficient matrix GA in the frequency domain and the coefficient matrix under fixed frequency and The relationship between them is combined with the least squares method to obtain the original time domain coefficients of the distributed load f(x, t) and complete the identification of f(x, t).

[0012] A distributed dynamic load identification system based on time variable separation, comprising:

[0013] The distributed load time-position separation module is used to separate the distributed load f(x, t) loaded on the identification object into a distributed load that can directly separate the time variable and a general distributed load. For the general distributed load, the time variable and the position variable are separated by the principle of two-dimensional Chebyshev orthogonal polynomial fitting function. For the distributed load that can directly separate the time variable, the position function is fitted by the principle of one-dimensional Chebyshev orthogonal polynomial fitting function.

[0014] The time-frequency conversion module is used to classify the distributed load according to its form, select the appropriate order of position orthogonal polynomials to fit the frequency domain force F(x, ω) for the distributed load formula with separated time variables, and calculate the frequency domain displacement response matrix W and frequency response function H respectively;

[0015] The frequency domain coefficient solving module is used to substitute the fitted frequency domain force into the displacement response and excitation relationship of the continuous system, and set the integral matrix of the frequency response function and the orthogonal polynomial as the dynamic calibration matrix HT, and obtain the simplified formula W = (HT)(GA), and obtain the coefficient matrix GA in the frequency domain;

[0016] The coefficient solving module under fixed frequency is used to classify the distributed load according to its form. For the load that can directly separate the time variable, the time variable function g(t) is transformed under fixed frequency to obtain the coefficient. For general loads, choose a time orthogonal polynomial of appropriate order Similarly, the coefficient matrix is ​​obtained according to Fourier transform at a fixed frequency q is the order of the time orthogonal polynomial;

[0017] The inverse reconstruction module is used to calculate the coefficient matrix GA in the frequency domain and the coefficient matrix in the fixed frequency and The relationship between them is combined with the least squares method to obtain the original time domain coefficients of the distributed load f(x, t) and complete the identification of f(x, t).

[0018] The present invention also provides a computer device, comprising: one or more processors; a memory; and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by the one or more processors, and when the programs are executed by the processors, the steps of the distributed dynamic load identification method based on time variable separation as described above are implemented.

[0019] The present invention also provides a computer storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the steps of the distributed dynamic load identification method based on time variable separation as described above are implemented.

[0020] Beneficial effects: The distributed dynamic load identification method and system based on time variable separation proposed in the present invention combines time domain theory with frequency domain identification, and divides distributed dynamic loads into two types: those that can directly separate time variables and general forms. In two cases, the distributed dynamic loads are identified in the frequency domain in combination with orthogonal polynomials, and the Fourier transform theory of the time domain-frequency domain is used to restore the time domain distributed force. Identification in the frequency domain does not require the use of a large amount of data after time discretization for modeling, thus avoiding the matrix pathological problems that may occur in time domain identification; restoration and completion of the final identification in the time domain can more clearly and intuitively characterize the distributed dynamic loads than simple frequency domain identification. At the same time, in the identification process, the two-dimensional orthogonal polynomial fitting coefficient matrix of the time domain force that was previously identified in the time domain alone resulted in a too large calculation matrix, too much response data, and low calculation efficiency. The present invention transforms the solution of the two-dimensional matrix into a matrix that only requires the solution of the column vector matrix, greatly reducing the identification calculation data, improving the identification efficiency, and ensuring the identification accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 It is a flow chart of the distributed dynamic load identification method based on time variable separation of the present invention;

[0022] Figure 2 is a simply supported beam model under distributed load in an embodiment of the present invention;

[0023] Figure 3 is a comparison of the time history recognition results in an embodiment of the present invention;

[0024] Figure 4 is a comparison of load distribution identification results in an embodiment of the present invention;

[0025] Figure 5 is the absolute error of the time history recognition result in the embodiment of the present invention;

[0026] Figure 6 is the absolute error of the load distribution identification result in the embodiment of the present invention. DETAILED DESCRIPTION

[0027] The technical solution of the present invention is further described below in conjunction with the accompanying drawings.

[0028] The present invention divides the distributed load into a distributed load function that can directly separate the time and space variables and a general form of distributed load function. Based on the theory of orthogonal polynomial fitting one-dimensional and two-dimensional functions, the time domain load is converted into the frequency domain load in two cases, and the coefficients of the orthogonal polynomial are identified in the frequency domain using Chebyshev orthogonal polynomials. The time domain-frequency domain Fourier transform theory and the least squares method are combined to further identify the distributed time domain dynamic load of the continuous beam. The specific technical solution includes:

[0029] 1. Separation method of time variable and position variable of distributed dynamic load

[0030] Taking the distributed load f(x, t) of a one-dimensional structure as an example, it can be divided into a form that can directly separate the time variable and a general form of distributed load function. When encountering a general form function that cannot directly separate the time variable, the two-dimensional Chebyshev orthogonal polynomial fitting function theory can be used to separate the distributed force using orthogonal polynomials.

[0031] f(x,t)=f(x)g(t) (1)

[0032]

[0033] Formula (1) is a distributed load that can directly separate the time function, where f(x) is the position function and g(t) is the time function. Formula (2) is a general form of the distributed load function, Normalized Chebyshev orthogonal polynomials, represents the position polynomial, p is the order of the position polynomial, represents the time polynomial, q is the order of the time polynomial, a ij (ij=1,2,…,pq) are coefficients of a two-dimensional orthogonal polynomial.

[0034] Similarly, the f(x) position function in the separable time variable distribution load is fitted using the one-dimensional Chebyshev orthogonal polynomial fitting function principle:

[0035]

[0036] In formula (3), is the normalized one-dimensional Chebyshev position orthogonal polynomial, a i (i=1,2,…,r) are the coefficients of the one-dimensional polynomial. r is the order of the position orthogonal polynomial.

[0037] In addition, the time term in equation (2) is expanded separately, and we have:

[0038]

[0039] In this way, the time variables in the distributed load are separated from the position variables.

[0040] 2. Construct the time domain load-frequency domain load conversion relationship

[0041] The two types of separated time variables in time domain distributed loads (Equation (3) and Equation (4)) are converted into frequency domain forms according to Fourier time-frequency transform, so that the frequency domain force is obtained as follows:

[0042]

[0043] Formula (5) is the frequency domain force that can directly separate the time variable, and formula (6) is the general form of the frequency domain force. The matrix forms are:

[0044]

[0045] In equations (7) and (8), at a known excitation frequency, the coefficients obtained by Fourier transform are With coefficient are all known complex constants.

[0046] 3. Distributed dynamic load identification method based on time variable separation

[0047] Taking the continuous beam as an example, combined with the transfer relationship between the displacement response w and the dynamic load F in the frequency domain, the frequency domain forces of equations (7) and (8) are substituted into the relationship between the n-point response and the dynamic load of the continuous system in the frequency domain to obtain:

[0048]

[0049] In the above two equations, x is the spatial position distribution of the load, H xk (k=1,2,…,n) is the frequency response function of the corresponding position, w k (k=1,2,…,n) is the frequency domain response of the response point. Let the integral matrix of the frequency response function and the orthogonal polynomial be the dynamic calibration matrix HT, which can be simplified as follows:

[0050] W=(HT)(GA) (11)

[0051] In formula (11), W is the frequency domain displacement response matrix, and the matrix GA is the coefficient matrix of the Fourier transform coefficients of the time function and the fitting coefficients of the time domain load. Under the condition that the response and the dynamic calibration matrix HT are known, the coefficient matrix can be obtained by combining matrix inversion or generalized inverse method.

[0052] 4. Inverse reconstruction of time-domain distributed dynamic loads

[0053] After obtaining the coefficient matrix GA in the frequency domain, due to the fixed frequency and As the coefficients are known, the coefficients of the time domain expressions of the two types of distributed forces can be obtained respectively by combining the least squares method:

[0054]

[0055]

[0056] The coefficient matrix a i (i=1,2,…,r) and a ijSubstituting (ij=1,2,…,pq) into equation (3) and equation (2) respectively can complete the restoration of the two types of distributed forces.

[0057] Based on the above inventive concept, the method implementation steps of the present invention include:

[0058] Step 1: According to the distributed load f(x, t) loaded on the identification object, its frequency domain displacement response matrix W and frequency response function H are obtained; in the verification of the simulation identification calculation, the distributed load f(x, t) is first loaded on the identification object through forward calculation to obtain its frequency domain displacement response matrix W and frequency response function H; in practical applications, it can be directly measured through experiments;

[0059] Step 2: According to the form of the distributed load, select a position orthogonal polynomial of appropriate order to fit the frequency domain force F(x,ω); the selection of the position order: to ensure the solution of equation (11), the order of the position polynomial (r or p) must be less than or equal to the number of response measurement points n;

[0060] Step 3: Substitute the fitted frequency domain force in step 2 into the displacement response and excitation relationship of the continuous system to obtain the simplified formula W = (HT)(GA), and obtain the coefficient matrix GA in the frequency domain;

[0061] Step 4: According to the classification of distributed forces, directly perform Fourier transform on the function of time variables at a fixed frequency to obtain the coefficients Or use a time orthogonal polynomial of appropriate order and obtain the coefficient matrix according to Fourier transform at a fixed frequency After verification by simulation calculation, it is generally only necessary to take the time orthogonal polynomial to be above 3rd order and within 10th order to ensure the accuracy of recognition. Taking it too low will lead to insufficient fitting order and too low recognition accuracy; taking it too high will affect the recognition accuracy due to too many unknown coefficients. In practical applications, the cross-validation method can be used to compare and select an order with the best recognition effect.

[0062] Step 5: Combine the coefficient matrices in step 3 and step 4, and use equation (12) or equation (13) to obtain the original coefficients of the distributed load f(x, t), and complete the identification of f(x, t).

[0063] The following is an example of the practical application of the method of the present invention combined with a specific model, and verifies the performance of the method. Figure 2 The simply supported beam structure shown in the figure is subjected to distributed dynamic load, and the parameters of the simply supported beam are shown in Table 1:

[0064] Table 1 Geometric parameters and material properties of simply supported beams

[0065]

[0066] Taking the more computationally complex general form of distributed load identification as an example, Figure 2 On the simply supported beam model shown in the figure, a general distributed load is applied, and the relationship between the load and time is: The loading time is 10 s, 10 points evenly distributed on the beam are taken as response measurement points, and 50 identification points evenly distributed on the beam are taken for dynamic load identification.

[0067] Step 1: Calculate the distributed load in the frequency domain Frequency domain displacement response matrix W and frequency response function H under excitation;

[0068] Step 2: Select the 8th-order position orthogonal polynomial The fitting form of the frequency domain force F(x,ω) is obtained;

[0069] Step 3: For the continuous beam structure, substitute the frequency domain force into equation (10) to calculate the coefficient matrix in the frequency domain and obtain the coefficient matrix GA with a matrix size of 8 × 1;

[0070] Step 4: Select a 10th-order time orthogonal polynomial Perform Fourier transform at a fixed frequency and get Column vector matrix of ;

[0071] Step 5: According to formula (13), combine the obtained matrix GA with The distributed force coefficient a in formula (2) is calculated n The identification of the general form of distributed load can be completed by using the matrix form of (n=1,2,…,pq). Taking the time history identification result and the load distribution identification result of the beam with x=0.15m and the loading time t=5.5s as an example, the identification effect is as follows Figure 3 to Figure 6 As shown, it can be seen that the load error identified by the method of the present invention is very small and has high accuracy.

[0072] Based on the same technical concept as the method embodiment, the present invention also provides a distributed dynamic load identification system based on time variable separation, comprising:

[0073] The distributed load time-position separation module is used to separate the distributed load f(x, t) loaded on the identification object into a form in which the time variable can be directly separated and a general form of the distributed load function. For the general form of the distributed load, the time variable and the position variable are separated by the principle of the two-dimensional Chebyshev orthogonal polynomial fitting function. For the distributed load in which the time variable can be directly separated, the position function is fitted by the principle of the one-dimensional Chebyshev orthogonal polynomial fitting function.

[0074] The time-frequency conversion module is used to classify the distributed load according to its form, select the appropriate order of position orthogonal polynomials to fit the frequency domain force F(x, ω) for the distributed load formula with separated time variables, and calculate the frequency domain displacement response matrix W and frequency response function H respectively;

[0075] The frequency domain coefficient solving module is used to substitute the fitted frequency domain force into the displacement response and excitation relationship of the continuous system, and set the integral matrix of the frequency response function and the orthogonal polynomial as the dynamic calibration matrix HT, and obtain the simplified formula W = (HT)(GA), and obtain the coefficient matrix GA in the frequency domain;

[0076] The coefficient solving module under fixed frequency is used to classify the distributed load according to its form. For the load that can directly separate the time variable, the time variable function g(t) is transformed under fixed frequency to obtain the coefficient. For general loads, choose a time orthogonal polynomial of appropriate order Similarly, the coefficient matrix is ​​obtained according to Fourier transform at a fixed frequency q is the order of the time orthogonal polynomial;

[0077] The inverse reconstruction module is used to calculate the coefficient matrix GA in the frequency domain and the coefficient matrix in the fixed frequency and The relationship between them is combined with the least squares method to obtain the original time domain coefficients of the distributed load f(x, t) and complete the identification of f(x, t).

[0078] The present invention also provides a computer device, comprising: one or more processors; a memory; and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by the one or more processors, and when the programs are executed by the processors, the steps of the distributed dynamic load identification method based on time variable separation as described above are implemented.

[0079] The present invention also provides a computer storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the steps of the distributed dynamic load identification method based on time variable separation as described above are implemented.

[0080] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, devices, computer equipment or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0081] The present invention is described with reference to a flowchart of a method according to an embodiment of the present invention. It should be understood that each process in the flowchart and a combination of processes in the flowchart can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the process in the flowchart. Figure 1 These computer program instructions can also be stored in a computer-readable memory that can guide a computer or other programmable data processing device to work in a specific way, so that the instructions stored in the computer-readable memory produce a product including an instruction device, which implements the functions specified in the process. Figure 1 These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the functions specified in the process. Figure 1 The steps of a specified function in a process or multiple processes.

Claims

1. A distributed dynamic load identification method based on time variable separation, characterized in that: The following steps are involved: Step 1: According to the distributed load f(x, t) loaded on the identification object, it is divided into a distributed load that can directly separate the time variable and a distributed load in a general form. For the distributed load in the general form, the time variable and the position variable are separated by the principle of two-dimensional Chebyshev orthogonal polynomial fitting function. For the distributed load that can directly separate the time variable, the position function is fitted by the principle of one-dimensional Chebyshev orthogonal polynomial fitting function; Step 2: According to the form of distributed load, for the distributed load formula with separated time variables, select appropriate order position orthogonal polynomials to fit the frequency domain force F(x,ω), and calculate the frequency domain displacement response matrix W and frequency response function H respectively; Step 3: Substitute the fitted frequency domain force into the displacement response and excitation relationship of the continuous system, set the integral matrix of the frequency response function and the orthogonal polynomial as the dynamic calibration matrix HT, obtain the simplified formula W = (HT)(GA), and obtain the coefficient matrix GA in the frequency domain; Step 4: According to the form of distributed load classification, for loads that can directly separate time variables, perform Fourier transform on the time variable function g(t) at a fixed frequency to obtain the coefficients For general loads, choose a time orthogonal polynomial of appropriate order Similarly, the coefficient matrix is ​​obtained according to Fourier transform at a fixed frequency n=1,2,…q, q is the order of the time orthogonal polynomial; Step 5: According to the coefficient matrix GA in the frequency domain and the coefficient matrix under fixed frequency and The relationship between them is combined with the least squares method to obtain the original time domain coefficients of the distributed load f(x, t) and complete the identification of f(x, t).

2. The method according to claim 1, characterized in that In step 1, for a general form of distributed load, the time variable and the position variable are separated by the principle of a two-dimensional Chebyshev orthogonal polynomial fitting function, including: According to the two-dimensional Chebyshev orthogonal polynomial fitting function theory, the distributed load is separated using orthogonal polynomials: In the formula, is the normalized Chebyshev orthogonal polynomial, represents the position polynomial, i = 1, 2, ..., p, p is the order of the position polynomial, represents the time polynomial, j = 1, 2, ..., q, q is the order of the time polynomial, a ij are the coefficients of a two-dimensional orthogonal polynomial.

3. The method according to claim 2, characterized in that For distributed loads that can directly separate time variables, the position function is fitted by the one-dimensional Chebyshev orthogonal polynomial fitting function principle, including: The distributed load with directly separable time function is expressed as f(x,t)=f(x)g(t), where f(x) is the position function and g(t) is the time function. The position function f(x) in the separable time variable distributed load is fitted using the one-dimensional Chebyshev orthogonal polynomial fitting function principle: In the formula, is the normalized Chebyshev position orthogonal polynomial, a i are the coefficients of the one-dimensional polynomial, i = 1, 2, …, r, and r is the order of the position orthogonal polynomial.

4. The method according to claim 3, characterized in that In step 2, for the distributed load formula of the separated time variable, appropriate order position orthogonal polynomials are selected to fit the frequency domain force F(x,ω), including: The two types of distributed load formulas with separated time variables are converted into frequency domain forms according to Fourier time-frequency transform to obtain frequency domain force expressions; Select appropriate order of position orthogonal polynomials to fit the frequency domain force F(x,ω).

5. The method according to claim 4, characterized in that For distributed loads that can directly separate time variables, the frequency domain force expression after Fourier time-frequency transform is: The matrix form is: For the general form of distributed load, the time term is expanded separately, and we have: Then the frequency domain force expression after Fourier time-frequency transform is: The matrix form is:

6. The method according to claim 1, characterized in that In step 5, for the distributed load that can directly separate the time function, according to The coefficient a of the time domain expression of this type of distributed force is obtained by using the least squares method. i , i=1,2,…,r, change a i Bringing it back into the load expression of the one-dimensional Chebyshev orthogonal polynomial expansion, we can get the distributed load identification result.

7. The method according to claim 1, characterized in that In step 5, for the general form of distributed load, according to The coefficient a of the time domain expression of this type of distributed force is obtained by using the least squares method. ij , a ij Bringing it back into the load expression of the two-dimensional Chebyshev orthogonal polynomial expansion, we can obtain the distributed load identification result.

8. A distributed dynamic load identification system based on time variable separation, characterized in that: include: The distributed load time-position separation module is used to separate the distributed load f(x, t) loaded on the identification object into a distributed load that can directly separate the time variable and a general distributed load. For the general distributed load, the time variable and the position variable are separated by the principle of two-dimensional Chebyshev orthogonal polynomial fitting function. For the distributed load that can directly separate the time variable, the position function is fitted by the principle of one-dimensional Chebyshev orthogonal polynomial fitting function. The time-frequency conversion module is used to classify the distributed load according to its form, select the appropriate order of position orthogonal polynomials to fit the frequency domain force F(x, ω) for the distributed load formula with separated time variables, and calculate the frequency domain displacement response matrix W and frequency response function H respectively; The frequency domain coefficient solving module is used to substitute the fitted frequency domain force into the displacement response and excitation relationship of the continuous system, and set the integral matrix of the frequency response function and the orthogonal polynomial as the dynamic calibration matrix HT, and obtain the simplified formula W = (HT)(GA), and obtain the coefficient matrix GA in the frequency domain; The coefficient solving module under fixed frequency is used to classify the distributed load according to its form. For the load that can directly separate the time variable, the time variable function g(t) is transformed under fixed frequency to obtain the coefficient. For general loads, choose a time orthogonal polynomial of appropriate order Similarly, the coefficient matrix is ​​obtained according to Fourier transform at a fixed frequency n=1,2,…q, q is the order of the time orthogonal polynomial; The inverse reconstruction module is used to calculate the coefficient matrix GA in the frequency domain and the coefficient matrix in the fixed frequency and The relationship between them is combined with the least squares method to obtain the original time domain coefficients of the distributed load f(x, t) and complete the identification of f(x, t).

9. A computer device, characterized in that: include: one or more processors; Memory; and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by the one or more processors, and when the programs are executed by the processors, the steps of the distributed dynamic load identification method based on time variable separation as described in any one of claims 1 to 7 are implemented.

10. A computer storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the distributed dynamic load identification method based on time variable separation according to any one of claims 1 to 7 are implemented.

Images