Quick prediction method and system for axial force load of variable cross-section beam

Through the theory of Temuxinke beam and the frequency inversion mapping mechanism, combined with the differential product method and the inverse matrices, the rapid prediction problem of axial force load of variable cross-section beams is solved, real-time monitoring requirements for complex structures are realized, and calculation efficiency and accuracy are improved.

CN120509194APending Publication Date: 2025-08-19CHONGQING TSINGSHAN IND
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202510627015.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-05-15
Publication Date
2025-08-19

AI Technical Summary

Technical Problem

The prior art is difficult to achieve rapid and accurate monitoring of axial force loads in variable cross-section beam structures, especially under dynamic conditions, and traditional methods consume large computing resources and have serious accuracy losses.

Method used

The frequency inversion mapping mechanism is adopted, combined with the Temuxinke beam theory, and through the differential product method and the inverse operation of the accompanying matrix, an axial force-frequency correlation matrix model is established to achieve rapid prediction of the axial force of variable cross-section beams.

Benefits of technology

It realizes rapid prediction of axial force load of variable cross-section beams, shortens the time from minute to milliseconds, and is highly adaptable. It is suitable for complex structures such as bridge cables and aero engine blades, reducing calculation costs and hardware transformation costs.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN120509194A_ABST
    Figure CN120509194A_ABST
Patent Text Reader

Abstract

The invention discloses a rapid prediction method and system for an axial force load of a variable cross-section beam. The method comprises the following steps: 1) establishing a coupling dynamics differential equation and corresponding boundary conditions of the variable cross-section beam under the action of an axial force; 2) performing spatial discretization processing through a differential quadrature method; (3) geometric parameters and material parameters of the target variable cross-section beam and actually-measured inherent frequency values of the first m orders are obtained; and 4) resolving the value and direction of the axial force of the variable cross-section beam in real time by using the adjoint matrix inverse operation. According to the method, a dynamic prediction scheme based on the Tiangshencaro beam theory is provided, a frequency inversion mapping mechanism is adopted in a breakthrough mode, and rapid prediction of the axial tension / pressure under the variable cross-section beam condition is achieved. Meanwhile, the system can be integrated on equipment such as a bridge cable monitoring terminal and an aero-engine blade load analyzer, and the problem of real-time force measurement of a complex variable cross-section beam structure is effectively solved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of engineering structure mechanics analysis and health monitoring, and in particular to a method and system for quickly predicting the axial force load of a variable-section beam. Background Art

[0002] In the field of engineering structural mechanics analysis and health monitoring, variable-section beams serve as the core load-bearing elements of key structures such as bridge cables and aircraft engine blades. Real-time monitoring of their axial loads is crucial for structural safety assessment. Existing technologies for monitoring the axial loads of variable-section beams rely on static testing and numerical analysis methods based on finite element simulation. This existing method for monitoring the axial loads of variable-section beams has significant limitations: First, static testing relies on applied loads to simulate actual operating conditions, making it difficult to capture changes in axial force under dynamic conditions. Furthermore, testing cycles are long and costly. Second, while numerical analysis methods based on finite element simulation can handle complex geometries, the modeling process is cumbersome and computationally resource-intensive. Especially for beam structures with continuously varying cross-sectional parameters, mesh discretization can easily lead to a loss of accuracy, making it difficult to meet real-time monitoring requirements. Furthermore, existing theoretical studies often rely on simplified models based on Euler-Bernoulli beam theory, ignoring shear deformation effects. Consequently, they are unable to accurately characterize the coupling relationship between axial force and dynamic response in variable-section scenarios. Although the improved Timoshenko beam theory introduces shear deformation correction, the solution of the dynamic equations for variable cross-section characteristics still faces the problem of high-dimensional nonlinearity, and traditional modal analysis methods make it difficult to construct an efficient axial force-frequency mapping model. Summary of the Invention

[0003] In view of the above-mentioned deficiencies in the prior art, the technical problem to be solved by the present invention is: how to provide a method for quickly predicting the axial force load of a variable-section beam by using a frequency inversion mapping mechanism to realize rapid prediction of axial tension / compression under variable-section beam conditions.

[0004] In order to solve the above technical problems, the present invention adopts the following technical solutions:

[0005] A rapid prediction method for the axial force load of a variable cross-section beam comprises the following steps:

[0006] Step 1) establishing the coupled dynamic differential equations and corresponding boundary conditions of the variable cross-section beam under axial force;

[0007] Step 2) spatially discretizing the coupled dynamic differential equations and corresponding boundary conditions obtained in step 1) by using the differential quadrature method, constructing an axial force-frequency correlation matrix model, and forming a homogeneous equation system for the axial force;

[0008] Step 3) obtaining the geometric parameters, material parameters and measured first m-order natural frequency values of the target variable-section beam;

[0009] Step 4) Use the adjoint matrix inverse operation to solve the axial force value and direction of the variable cross-section beam in real time.

[0010] Preferably, in step 1), based on the Timoshenko beam theory, the axial elongation, bending, and shear deformation of the variable-section beam are comprehensively considered, and the cross-sectional geometric parameters of the variable-section beam are introduced to establish the coupled dynamic differential equations of the variable-section beam under the action of axial force and the corresponding boundary conditions.

[0011] Preferably, in step 1), the cross-sectional geometric parameters of the variable-section beam include a cross-sectional area function A(x) and a moment of inertia function I(x), wherein x is a dimensionless coordinate along the axial direction of the variable-section beam, and the value range of x is [0, 1].

[0012] Preferably, the method for performing spatial discretization processing in step 2) is:

[0013] The variable-section beam is divided into n grid points along the axial direction. The coupled dynamic differential equations in the continuous domain are transformed into algebraic equations at n grid points by the differential quadrature method, and the generalized characteristic matrix equation including the axial force p is generated, which is expressed as:

[0014]

[0015] Where: K is the stiffness matrix, M is the mass matrix, Ω is the natural frequency, is the modal displacement vector.

[0016] Preferably, the axial force-frequency correlation matrix constructed in step 2) generates an implicit nonlinear equation about the axial force p by introducing an adjoint matrix conditional expression, and a homogeneous equation group is formed after substituting natural frequency values of different orders.

[0017] Preferably, in step 3), the homogeneous equations are converted into an overdetermined linear equations in combination with the axial force-frequency correlation matrix model constructed in step S2).

[0018] Preferably, in step 3), a cross-sectional variation function is input, and the cross-sectional variation function is a preset polynomial function, exponential function or piecewise linear function, and its high-order derivatives satisfy continuity conditions at the boundary.

[0019] Preferably, in step 4), the overdetermined linear equations are solved by adjoint matrix inverse operation, and the numerical value and direction information of the axial force load are inverted based on the minimum singular value decomposition or orthogonal triangular decomposition technology of the matrix to realize dynamic identification of axial tension and pressure.

[0020] Preferably, in step 4), the method of the adjoint matrix inverse operation is: regularizing the coefficient matrix in the overdetermined linear equations, extracting the non-zero least squares solution using the orthogonal triangular decomposition technique, and determining the direction of the axial force by sign discrimination.

[0021] Preferably, in step 4), the accuracy of the non-zero least squares solution is numerically corrected by the Newton iteration method. The correction method is: using the axial force calculated in the previous solution as the initial value, constructing the Jacobian matrix for iterative approximation until the relative error of two adjacent non-zero least squares solutions is less than 10 -5 .

[0022] A system for realizing the above-mentioned rapid prediction method of the axial force load of the variable cross-section beam comprises:

[0023] A parameter acquisition module, which is used to obtain geometric parameters, material parameters and measured first m-order natural frequency values of the variable-section beam;

[0024] a matrix construction module, the matrix construction module being connected to the parameter acquisition module, the matrix construction module discretizing the coupled dynamic differential equations and corresponding boundary conditions based on the differential quadrature method, and generating an axial force-frequency correlation matrix model in real time;

[0025] A numerical solution module, which is connected to the matrix construction module and has an adjoint matrix operation unit integrated therein, and is used for inversely solving the axial force value and direction of the variable-section beam;

[0026] An output module is connected to the numerical solution module, and is used to transmit the solution result to an external monitoring terminal through a visual interface or a wireless communication unit.

[0027] Compared with the prior art, the present invention has the following advantages:

[0028] 1. This invention addresses the low efficiency and poor adaptability of traditional axial force detection methods, which rely on static testing and finite element simulation. It proposes a dynamic prediction scheme based on Timoshenko beam theory and employs a groundbreaking frequency inversion mapping mechanism to rapidly predict axial tension and compression forces under variable-section beam conditions. Furthermore, the system can be integrated into equipment such as bridge cable monitoring terminals and aircraft engine blade load analyzers, effectively solving the real-time force measurement challenge for complex variable-section beam structures.

[0029] 2. The present invention discretizes the coupled dynamic differential equations with high precision through the differential quadrature method, avoiding the grid dependence of the traditional finite element method, and combines the adjoint matrix inverse operation to realize the rapid solution of the overdetermined equation group, so that the axial force prediction time is shortened from the minutes level of traditional simulation to the millisecond level, and the computational efficiency is significantly improved, meeting the real-time monitoring needs under complex working conditions.

[0030] 3. The coupled dynamic model based on Timoshenko beam theory comprehensively considers the shear deformation and axial-bending mode coupling effects, supports various cross-sectional variation forms such as polynomials, exponentials, and piecewise functions, and has an axial force inversion accuracy within ±3%. It effectively covers typical scenarios such as bridge cables (gradual cross-sections) to aviation blades (special-shaped surfaces), and its adaptability is greatly enhanced.

[0031] 4. The present invention proposes a universal construction method for the frequency-axial force mapping matrix. Only the first m-order (m≥2) measured natural frequencies need to be input to dynamically identify the load direction (tension / compression) and value. The system is compatible with existing vibration sensor networks and can be seamlessly integrated into the structural health monitoring platform, reducing hardware modification costs and greatly enhancing engineering practicality. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Attachment Figure 1 This is a flow chart of a method for rapidly predicting the axial force load of a variable-section beam according to the present invention;

[0033] Attachment Figure 2 This is a system connection block diagram of the rapid prediction system for the axial force load of a variable-section beam of the present invention;

[0034] Attachment Figure 3 A geometric model diagram of a cantilever beam with a variable cross-section in an embodiment;

[0035] Attachment Figure 4 This is a comparison diagram of the results between the method of the present invention and finite element simulation. DETAILED DESCRIPTION

[0036] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. The components of the embodiments of the present invention generally described and shown in the drawings herein can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the invention claimed for protection, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0037] In this specific embodiment, a method for quickly predicting the axial force load of a variable cross-section beam is provided. Figure 1 As shown, the following steps are included:

[0038] Step 1) Establish the coupled dynamic differential equations and corresponding boundary conditions of the variable cross-section beam under axial force.

[0039] Specifically, based on Timoshenko beam theory, the axial elongation, bending, and shear deformation of variable-section beams are comprehensively considered, and the cross-sectional geometric parameters of variable-section beams are introduced to establish the coupled dynamic differential equations and corresponding boundary conditions for variable-section beams under axial force. The cross-sectional geometric parameters of variable-section beams include the cross-sectional area function A(x) and the moment of inertia function I(x), where x is the dimensionless coordinate along the axial direction of the variable-section beam and has a value range of [0, 1].

[0040] Based on Timoshenko theory and Hamilton's principle, the vibration equation of the variable cross-section Timoshenko beam can be deduced as follows:

[0041]

[0042] And the boundary conditions are:

[0043]

[0044] The dimensionless variables are introduced as follows:

[0045]

[0046] Substituting into equations (1)-(6), we obtain the following dimensionless equations:

[0047]

[0048] And the corresponding boundary conditions:

[0049]

[0050] Where γ1=A0 / L 2 ,γ2=GA0 / EL 2 , p=P / EL 2 ,α1=P / EA0,α2=P / GA0,A0 and I0 come from the area and moment of inertia of a certain cross section.

[0051] Step 2) The coupled dynamic differential equations and corresponding boundary conditions obtained in step 1) are spatially discretized by the differential quadrature method, an axial force-frequency correlation matrix model is constructed, and a homogeneous set of equations about the axial force is formed.

[0052] Specifically, the constructed axial force-frequency correlation matrix generates an implicit nonlinear equation about the axial force p by introducing the adjoint matrix conditional expression, and a homogeneous set of equations is formed by substituting the natural frequency values of different orders.

[0053] The method for spatial discretization is:

[0054] The variable-section beam is divided into n grid points along the axial direction. The coupled dynamic differential equations in the continuous domain are transformed into algebraic equations at n grid points by the differential quadrature method, and the generalized characteristic matrix equation including the axial force p is generated, which is expressed as:

[0055]

[0056] Where: K is the stiffness matrix, M is the mass matrix, Ω is the natural frequency, is the modal displacement vector.

[0057] According to the differential quadrature method, the g-th order derivative of the function V(x) is approximately:

[0058]

[0059] in is the weighting coefficient of the g-order derivative, V(x j ) is the coordinate x of the discrete point j The corresponding function value is represented by a vector V = [V1, V2, ..., V n ] T ; n is the number of discrete points, and the discrete points are obtained using the Chebyshev–Gauss–Lobatto formula, that is:

[0060]

[0061] Using the Lagrange interpolation polynomial as the test function, the weighted coefficient matrix of the first-order derivative can be obtained as follows

[0062]

[0063] in:

[0064] The weighting coefficients of higher-order derivatives are calculated as follows:

[0065]

[0066] According to the above method, the dynamic equation can be written in the following discrete form:

[0067]

[0068] The corresponding boundary condition discrete form is:

[0069]

[0070] in and are weighted coefficient matrices of the first-order and second-order derivatives respectively; I is the identity matrix; S=(A1,A2,…,A n ) T ,Q=(I1,I2,…,I n ) T ,U=(u1,u2,…,u n ) T ,θ=(θ1,θ2,…,θ n ) T And W=(w1,w2,…,w n ) T In addition, It is the Hadamard product operation, and its operation rules can be briefly described as follows: if there are two matrices R = [r ij ]∈C M×N and S = [s ij ]∈C M×N , then Similarly, the k-th Hadamard product operation of matrix R can be expressed as

[0071] The discretized linear equations are corrected by the discrete boundary equations and expressed in matrix form as follows:

[0072]

[0073] Where X = [U, θ, W] T , K and M are the stiffness and mass matrices respectively.

[0074] set up Substitute it into formula (25), and the solution of formula (25) is transformed into the eigenvalue problem, as follows:

[0075]

[0076] Where Ω represents the natural frequency, is the corresponding vibration mode.

[0077] Step 3) Obtain the geometric parameters, material parameters and measured first m-order natural frequency values of the target variable-section beam.

[0078] Specifically, the homogeneous equations are transformed into an overdetermined linear equations by combining the axial force-frequency correlation matrix model constructed in step S2).

[0079] At the same time, in step 3), a cross-sectional variation function is input, which is a preset polynomial function, exponential function or piecewise linear function, and its high-order derivatives meet continuity conditions at the boundary.

[0080] Step 4) Use the adjoint matrix inverse operation to solve the axial force value and direction of the variable cross-section beam in real time.

[0081] Specifically, the overdetermined linear equations are solved by adjoint matrix inverse operation, and the numerical value and direction information of the axial force load are inverted based on the minimum singular value decomposition or orthogonal triangular decomposition technology of the matrix, thereby realizing the dynamic identification of axial tension and pressure.

[0082] Among them, the method of adjoint matrix inverse operation is: regularize the coefficient matrix in the overdetermined linear equation system, use the orthogonal triangular decomposition technology to extract the non-zero least squares solution, and determine the direction of the axial force through sign discrimination.

[0083] Secondly, the accuracy of the non-zero least squares solution is numerically corrected by the Newton iteration method. The correction method is as follows: the axial force calculated in the previous solution is used as the initial value, and the Jacobian matrix is constructed for iterative approximation until the relative error between two adjacent non-zero least squares solutions is less than 10 -5 .

[0084] In addition, this specific embodiment also provides a system for realizing the rapid prediction method of the axial force load of the variable cross-section beam, as shown in the attached figure. Figure 2 As shown, it includes parameter acquisition module, matrix construction module, numerical solution module and output module.

[0085] Among them, the parameter acquisition module is used to obtain the geometric parameters, material parameters and measured first m-order natural frequency values of the variable-section beam; specifically, the parameter acquisition module receives the geometric parameters (such as cross-sectional area function, material parameters) and measured natural frequency data of the variable-section beam input by external sensors or preset databases through a data interface; the parameter acquisition module supports multiple data input methods, including: 1) directly reading a preset cross-sectional change function (such as a polynomial function, an exponential function); 2) receiving real-time geometric parameters collected by a laser displacement sensor and a strain gauge array; 3) integrating a material parameter database to automatically match material density and elastic modulus.

[0086] The matrix construction module is connected to the parameter acquisition module. The matrix construction module discretizes the coupled dynamic differential equations and corresponding boundary conditions based on the differential quadrature method, and generates an axial force-frequency correlation matrix model in real time.

[0087] Specifically, the matrix construction module discretizes the coupled dynamic equations based on the differential quadrature method and generates the axial force-frequency correlation matrix in real time. The matrix construction module divides the beam into n non-uniform grid points along the axial direction according to the input geometric parameters, and uses the Lagrange interpolation basis function to calculate the stiffness matrix K(p) and mass matrix M at each grid point to construct a matrix of the form The generalized characteristic equation of .

[0088] The numerical solution module is connected to the matrix construction module, and an adjoint matrix operation unit is integrated in the numerical settlement module. The numerical solution module is used to inversely solve the axial force value and direction of the variable-section beam.

[0089] Specifically, the numerical solution module solves the axial force value and direction through inverse calculation of adjoint matrix operations; the numerical solution module extracts the least squares solution based on QR decomposition and uses residual analysis to determine the uniqueness of the solution. When the equation system has multiple solutions, the axial force value is corrected by Newton iteration until the convergence tolerance is less than 10 -5 Direction discrimination is based on sensitivity analysis of the sign of the matrix eigenvalues.

[0090] The output module is connected to the numerical solution module, and the output module is used to transmit the solution result to the external monitoring terminal through a visual interface or a wireless communication unit.

[0091] Specifically, the output module pushes the calculated results to a visualization interface or remote monitoring terminal, triggering early warning logic. The output module dynamically displays the axial force time history curve through the HMI interface and supports 4G / 5G wireless transmission to a cloud server. A preset threshold detection unit compares the axial force with the safety range in real time. If the threshold is exceeded, an audible and visual alarm or shutdown signal is triggered.

[0092] The above system can be deployed on an embedded hardware platform and integrated with a bridge cable monitoring terminal or an aircraft engine blade load analyzer to feed back the real-time solution results to the early warning system and trigger an alarm signal when the axial force exceeds the preset threshold.

[0093] Example:

[0094] The following is a specific example, taking a cantilever beam with exponentially varying cross-sectional width and subjected to an axial force as an example, and verifying the effectiveness and accuracy of this method by comparing it with finite element results. Figure 3 The finite element model of the variable cross-section cantilever beam in this embodiment has a length of L = 0.3m, a span-to-height ratio of L / h = 15, and a width expressed as an exponential function b0e -5X Change, where b0 = 0.04, the geometric parameters obtained are A(X) = b0he -5X , I(X)=b0h 3 e -5X / 12. This example is a cantilever beam, and its corresponding discrete boundary conditions are:

[0095] I i U=0,I i W=0,I i θ=0,i=1(27)

[0096]

[0097] The six equations representing the cantilever boundary conditions, namely equations (27) and (28), are substituted for the corresponding six equations in equations (19)-(21), and the discrete dynamic equations of the exponentially variable cross-section cantilever beam modified by the boundary conditions are obtained. Through the adjoint matrix operation, the numerical value and direction information of the axial force load are obtained by inversion. Figure 4 3 is a comparison between the method of the present invention and the finite element simulation results. It can be seen that the method of the present invention is consistent with the finite element simulation results.

[0098] Compared with the existing technology, the present invention addresses the problems of low efficiency and poor adaptability of traditional axial force detection methods that rely on static tests and finite element simulations, proposes a dynamic prediction scheme based on Timoshenko beam theory, and uses a breakthrough frequency inversion mapping mechanism to achieve rapid prediction of axial tension / compression under variable cross-section beam conditions. At the same time, the system of the present invention can be integrated into equipment such as bridge cable monitoring terminals and aircraft engine blade load analyzers, effectively solving the real-time force measurement problem of complex variable cross-section beam structures. The present invention discretizes the coupled dynamic differential equations with high precision through the differential quadrature method, circumvents the grid dependence of the traditional finite element method, and combines the adjoint matrix inverse operation to achieve rapid solution of overdetermined equations, reducing the time consumption of axial force prediction from minutes in traditional simulations to milliseconds, significantly improving computational efficiency, and meeting the real-time monitoring needs under complex working conditions. The present invention comprehensively considers the shear deformation and axial-bending modal coupling effects based on the coupled dynamic model of Timoshenko beam theory, supports multiple cross-sectional variation forms such as polynomials, exponentials, and piecewise functions, and the axial force inversion accuracy is within ±3%, effectively covering typical scenarios such as bridge cables (gradual cross-sections) to aviation blades (special-shaped surfaces), and its adaptability is greatly enhanced. The present invention proposes a universal construction method for the frequency-axial force mapping matrix, which only requires the input of the first m-order (m≥2) measured natural frequencies to dynamically identify the load direction (tension / compression) and value. The system is compatible with existing vibration sensor networks and can be seamlessly integrated into the structural health monitoring platform, reducing hardware modification costs and greatly enhancing engineering practicality.

[0099] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit the technical solutions. Those skilled in the art should understand that modifications or equivalent replacements of the technical solutions of the present invention that do not depart from the purpose and scope of the technical solutions of the present invention should be included in the scope of the claims of the present invention.

Claims

1. A rapid prediction method for the axial force load of a variable cross-section beam, characterized in that: The following steps are involved: Step 1) establishing the coupled dynamic differential equations and corresponding boundary conditions of the variable cross-section beam under axial force; Step 2) spatially discretizing the coupled dynamic differential equations and corresponding boundary conditions obtained in step 1) by using the differential quadrature method, constructing an axial force-frequency correlation matrix model, and forming a homogeneous equation system for the axial force; Step 3) obtaining the geometric parameters, material parameters and measured first m-order natural frequency values of the target variable-section beam; Step 4) Use the adjoint matrix inverse operation to solve the axial force value and direction of the variable cross-section beam in real time.

2. The rapid prediction method for axial force load of a variable cross-section beam according to claim 1 is characterized in that: In step 1), based on the Timoshenko beam theory, the axial elongation, bending, and shear deformation of the variable-section beam are comprehensively considered, and the cross-sectional geometric parameters of the variable-section beam are introduced to establish the coupled dynamic differential equations of the variable-section beam under the action of axial force and the corresponding boundary conditions.

3. The rapid prediction method for axial force load of a variable cross-section beam according to claim 2 is characterized in that: In step 1), the cross-sectional geometric parameters of the variable-section beam include the cross-sectional area function A(x) and the moment of inertia function I(x), where x is a dimensionless coordinate along the axial direction of the variable-section beam, and the value range of x is [0, 1].

4. The rapid prediction method for axial force load of a variable cross-section beam according to claim 1 is characterized in that: The method for performing spatial discretization processing in step 2) is: The variable-section beam is divided into n grid points along the axial direction. The coupled dynamic differential equations in the continuous domain are transformed into algebraic equations at n grid points by the differential quadrature method, and the generalized characteristic matrix equation including the axial force p is generated, which is expressed as: Where: K is the stiffness matrix, M is the mass matrix, Ω is the natural frequency, is the modal displacement vector.

5. The rapid prediction method for axial force load of a variable cross-section beam according to claim 4 is characterized in that: The axial force-frequency correlation matrix constructed in step 2) generates an implicit nonlinear equation about the axial force p by introducing the adjoint matrix conditional expression. After substituting the natural frequency values of different orders into it, a homogeneous equation group is formed.

6. The rapid prediction method for axial force load of a variable cross-section beam according to claim 1 is characterized in that: In step 3), the homogeneous equations are transformed into an overdetermined linear equations in combination with the axial force-frequency correlation matrix model constructed in step S2).

7. The rapid prediction method for axial force load of a variable cross-section beam according to claim 1 is characterized in that: In step 4), the overdetermined linear equations are solved by adjoint matrix inverse operation. Based on the minimum singular value decomposition or orthogonal triangular decomposition technology of the matrix, the numerical value and direction information of the axial force load are inverted to realize the dynamic identification of axial tension and pressure.

8. The rapid prediction method for axial force load of a variable cross-section beam according to claim 7, characterized in that: In step 4), the method of the adjoint matrix inverse operation is: regularize the coefficient matrix in the overdetermined linear equation system, extract the non-zero least squares solution using the orthogonal triangular decomposition technique, and determine the direction of the axial force by sign discrimination.

9. The rapid prediction method for axial force load of a variable cross-section beam according to claim 8, characterized in that: In step 4), the accuracy of the non-zero least squares solution is numerically corrected by the Newton iteration method. The correction method is: the axial force calculated in the previous solution is used as the initial value, and the Jacobian matrix is constructed for iterative approximation until the relative error between two adjacent non-zero least squares solutions is less than 10 -5 .

10. A system for implementing the method for rapidly predicting the axial force load of a variable cross-section beam as claimed in claim 1, characterized in that: include: A parameter acquisition module, which is used to obtain geometric parameters, material parameters and measured first m-order natural frequency values of the variable-section beam; a matrix construction module, the matrix construction module being connected to the parameter acquisition module, the matrix construction module discretizing the coupled dynamic differential equations and corresponding boundary conditions based on the differential quadrature method, and generating an axial force-frequency correlation matrix model in real time; A numerical solution module, which is connected to the matrix construction module and has an adjoint matrix operation unit integrated therein, and is used for inversely solving the axial force value and direction of the variable-section beam; An output module is connected to the numerical solution module, and is used to transmit the solution result to an external monitoring terminal through a visual interface or a wireless communication unit.