Power Flow Characteristic Analysis And Coupling Stiffness Identification Of Elastic Coupled Beam Structure

As a basic structural form,beam structure has been widely used in the fields of architecture,navigation and aerospace.In the actual engineering environment,the beam structure is often in the bad working environment where the excitation load introduced by the internal and external environment of the system is constantly acting.Under these conditions,vibration damage is easy to occur in the beam structure.When analyzing this kind of problem,it can be decomposed into several sub-structure connection models,and the structural damage of the system can be characterized by setting the coupling relationship between the sub-structures,that is,the cracked beam is divided into the spring coupled beam structure form.The spring-coupled beam structure can also be used as a simplified form of other multi-section structures connected by elasticity in engineering,such as ship propulsion shafting,truss,bridge,etc.In addition,due to the complexity and diversity of actual structures in engineering,the coupling forms of the simplified coupling beam structures are also different.The power flow theory,as a vibration research method which takes speed and force as research variables,can study the input,output and transmission characteristics of energy in the system,and has advantages in the study of vibration characteristics of coupled beam structures.In addition,with the expansion of cracks in engineering,the reliability of the structure will decline.If not found in time,it will cause such as bridge collapse,air disaster,shipwreck,resulting in huge economic losses.Therefore,it is necessary to recognize and forecast the cracks in the structure and improve the working safety of the equipment.In this paper,a coupled beam structure model which can be reconstructed rapidly for different structural forms is established,and the characteristics of the structure can be predicted quickly and accurately.A coupling parameter identification model based on artificial neural network is proposed.The main work of this paper is as follows:A vibration analysis model of coupled beam structure with elastic boundary conditions is established.The boundary smooth Fourier series was used to expand the lateral displacement of the structure,and the system equation was constructed according to the energy principle and solved systematically with the Rayleigh-Ritz method.The correctness of the model was verified by comparing with the numerical solution.The effects of coupling system parameters on the natural frequency and dynamic response of coupled beam structures with elastic boundary conditions are studied.The coupling beam model under elastic boundary conditions was applied to the Euler-Bernoulli beam with unilateral cracking,and the effects of crack parameters on the natural frequency,amplitude-frequency response and power flow distribution characteristics of the cracked beam were predicted.The internal coupled beam model under elastic boundary conditions is established for the internal coupled beam structure.Each section of beam is divided into sub-beam structure and the coupled end spring is introduced at the coupling position to solve the problem of numerical instability of vibration power flow near the coupling system in the beam structure due to the discontinuity of shear force and bending moment at the coupling position.The influence of coupling form and coupling parameters on the transmission and distribution characteristics of vibration power flow in the internal coupled beam model under elastic boundary conditions is also predicted.The simulation data of the vibration system of the elastic beam structure with different coupling forms were used as the training samples of BP neural network.The models for the identification of coupling parameters in the end-coupled beam and the internal coupled beam under the elastic boundary conditions were established respectively.In the model mentioned above,the corresponding parameter sensitive region suitable for identification is selected,and a large number of characteristic data required for identification is calculated.The BP neural network was used to predict and compare with the actual values,and the relative errors were calculated to evaluate the effectiveness of BP neural network in identifying the coupling system parameters in the coupled beam structure under elastic boundary conditions.Based on LabVIEW platform,a two-channel signal acquisition program and a four-channel acceleration signal acquisition program with integrated signal scavenging function are written,and the cracked beam dynamic response test bed and coupled two-layer beam dynamic response test bed are built.The dynamic response test of the cracked beam was carried out to change the crack depth in the beam structure and observe the variation trend of the natural frequency of the beam structure with different sizes and crack locations.The dynamic response test of the coupled double beam was carried out to change the spring stiffness of the coupled unit,and the change trend of the excitation frequency corresponding to the partial local peak of the output power flow in the beam structure was observed.Compared with the research in the second and third chapters of the paper,the correctness of the previous research is verified.Based on the structural characteristic parameters obtained in the experiment,the crack parameters and coupling spring stiffness of the above beam structures were identified and predicted respectively.