| Bonded structure has been widely used in many important sectors such as aerospace, aviation and military industries for its superior performance. However kissing bond, weak bonding interface defects produced in the process of manufacture and working at times not only affect the reliability and safety of materials but also lead to catastrophic accidents. So it is significant to make accurate detection for interface strength and insure products working safety. This article proposes characterization of adhesive strength between metal and nonmetal components by nonlinear ultrasonic methods, designs immersion ultrasonic detecting system of high-energy ultrasonic signal, which can stimulate nonlinear harmonics, and studies the methods and related de-noising method for nonlinear ultrasound to represent adhesive strength of metal and nonmetal bonding structure using the nonlinear ultrasonic characteristic parameters.The main contents and innovative work in this paper can be summarized as follows.(1) By exploring nonlinear elastic wave equation, this paper investigates the relationship between ultrasonic signal characteristics and nonlinearity parameter of materials, analyzes its physical meaning and other related parameters of nonlinear coefficient, which lays the foundation for testing metal and nonmetal adhesive strength with nonlinear parameter. For steel-synthetic glass adhesive structure, the quantitative effect of adhesive layer thickness and excitation frequency on secondary harmonic was investigated considering the case of sound attenuation, which provides an experimental basis for selecting the optimum excitation frequency. Hydrogen bonds at steel-layer interface combined with nonlinear viscoelastic constitutive model of layer were introduced into the nonlinear interaction between ultrasound and media and functional relationship of nonlinear coefficient and adhesive strength was established providing a theoretical explanation for nonlinear ultrasonic testing the adhesive strength of the metal and nonmetal components.(2) The quantitative relationship between nonlinear parameter of material and voltages for different frequency components of signal acquired by piezoelectric transducer was deduced proving that it is feasible to characterize nonlinear coefficient of media by ultrasonic signal characteristics. The influence on secondary harmonic amplitude of nonlinear ultrasonic by different impulse width was researched through the theoretical analysis and simulation experiment indicating that apparent harmonic could only be inspired if the impulse width of excitation signal was greater than 1.2us, and the larger the impulse width of excitation signal was, the greater the amplitude of harmonics would be. To overcome time domain ultrasonic echoes aliasing effect,9 cycles burst and 6 cycles burst weighted by Hanning window are chosen as excitation signal for 6mm steel-synthetic glass specimen and 4mm steel-synthetic glass specimen respectively, which has solved the problem how to stimulate apparent nonlinear harmonics in materials.(3) Effects of excitation signal waveform on ultrasound spectrum precision was analyzed, the problem that window function truncates nonlinear ultrasound bringing about spectrum leakage and harmonic amplitude calculation error was discussed. By deducing the energy centroid of discrete spectrum of frequency, energy centrobaric correction method for nonlinear ultrasound frequency spectrum was proposed to measure fundamental and harmonic amplitudes precisely. Nonlinear ultrasound signal is non-stationary, nonlinear and apt to be interfered by noise, experimental results reveal that instrument nonlinearity could be successfully reduced by using piezoelectric crystal plate natural filtering effect.(4) Nonlinear characteristic parameter extraction method based on dual-tree complex wavelet scale coefficients correlation and Bayesian maximum a posteriori estimation (MAP) filtering algorithm was proposed. Firstly, dual-tree complex wavelet transform (DT-CWT) was used to decompose the nonlinear ultrasound into several components with different frequency bands. Because frequency aliasing existed to a certain degree in the components, wavelet scale coefficients correlation was introduced to de-noise the components and eliminate the frequency aliasing as many as possible. Secondly, a further de-noising process was performed on the amended wavelet coefficients using MAP filtering algorithm so as to obtain second harmonic. The influence of different filtering algorithm on through-transmission ultrasound was inspected from the aspects of SNR, MSE and second harmonic error. Stimulation results indicate that signal-to-noise ratio of second harmonic was improved effectively after processed by the algorithm mentioned above.(5) The testing system for adhesive strength of metal and non-metal components using nonlinear ultrasonic method was established. The feasibility and validity of this method to reveal the adhesive strength of metal and non-metal components using nonlinear parameter which is calculated precisely by ultrasonic velocity, acoustic attenuation, the amplitudes of fundamental-frequency wave and second harmonic is confirmed by experimental results and comparison with the laws of adhesive strength changing with curing agent content.The work of this paper is of profound theoretical and practical significance on the application in tine-frequency processing technique of nonlinear ultrasonic signal in new field, and also characterizing adhesive strength between metal and nonmetal components. |
PDF Full Text Request