Research On Theory And Key Technologies Of Nonlinear Ultrasonics For Health Monitoring Of Plate-like Metallic Structures

Plate-like metallic structures are widely used in various engineering equipments. Fatigue cracking is one of the main damage types in such structures. Micro-scale crack generating in a structure leads to stress concentration, and propagates under variable loading conditions, if un-prevented, which may form one or several macro-scale cracks, and eventually result in abrupt fracture of the respective component, or even destroy the whole equipment. To ensure structural and operational safety, it is important to research structural health monitoring (SHM) techniques to detect, assess and locate fatigue cracks at the early stage. This is also prerequisite for maintenance. However, initial fatigue cracks in plate-like metallic structures are usually closed, i.e. the interfaces contact each other, and their positions are usually hidden deep in structure. These characteristics are much challenging for current SHM techniques.Nonlinear ultrasonic techniques, which detect damages based on various nonlinear effects of ultrasonic wave propagating in structures, have been proved to be able to identify microscopic material degradation and highly sensitive to micro-scale and closed cracks. Among them, vibro-acoustic modulation (VAM) technique is an emerging one developing rapidly in recent years. This technique excites the inspected structure with a high-frequency ultrasound and a low-frequency vibration during inspection, and then detects the existence of damages according to the VAM effect. This phenomenon has been observed in various materials and shows great potential for nondestructive testing. However, when applied to SHM of plate-like metallic structures, there are still some aspects need further investigation, including the underlying physical mechanism of damage-induced VAM effect, the quantitative relations between damage characteristics and nonlinear ultrasonic responses, signal processing and feature extraction method for damage evaluation, damage location strategy, etc.Aiming at the abovementioned problems, and granted by the Excellent Youth Foundation of Hu'nan Scientific Committee of China, This dissertation carries out the researches on theory and key technologies of VAM approach for SHM of plate-like metallic structures. The main contents and innovative work can be summarized as follows.1. The mechanism of structural damage-induced VAM effect is discussed from the viewpoints of microscopic and macroscopic respectively starting from the nonlinear elastic wave spectroscopy.(1) Firstly, the influence of damage on nonlinearity of material elastic behavior as a whole is characterized by the nonlinear coefficient. Starting from the macroscopic stress-strain constitutive relationship, the nonlinear elastic wave equation is deduced, and then solved through the perturbation approximation method. From the solution, the mechanism of damage-induced VAM phenomenon is explained in theory, and the quantitative relation between nonlinear coefficient and the acoustical responses is obtained. Secondly, the relationship between the nonlinear coefficient and damage variables is summarized. By combining with the solution of nonlinear elastic wave equation, the bridges between damage variables - nonlinear coefficient - acoustical responses are established.(2) A nonlinear spring model is used to describe the nonlinear elasticity of a micro crack. Based on this model, the acoustical response feature of a crack with rough contact interfaces to a high-frequency ultrasound and a low-frequency vibration mixing wave fields is deduced. And the quantitative relation between local damage variables and nonlinear acoustical responses is obtained.2. A time-frequency analysis algorithm for VAM signal processing is researched and proposed.VAM signal is nonlinear, non-stationary and modulated. To extract appropriate features for damage assessment, a time-frequency and AM-FM demodulation algorithm is proposed. Firstly, the principle of the Hilbert-Huang transform for signal demodulation and time-frequency analysis is summarized. Then the limitations of the Hilbert transform (HT) are remarked. As an enhancement of the HT, an algorithm named empirical AM-FM demodulation is investigated and improved. By combining the Empirical Mode Decomposition (EMD) with the improved empirical AM-FM demodulation, an algorithm suitable for instantaneous characteristics extraction of complicated multi-component signals is proposed.Simulation studies show that the EMD can decompose any multi-component signal into mono-component ones, and then the instantaneous amplitude and instantaneous frequency of a mono-component signal can be calculated via the improved empirical AM-FM demodulation algorithm. As compared with the HT, the improved algorithm can avoid approximation error of the HT, and the influence of noise can be also diminished.3. Experimental researches on the VAM technique for initial fatigue crack detection and progressive crack assessment in plate-like metallic structures are performed.(1) Facing to engineering application of the VAM technique, one practical method for measuring the nonlinear coefficient of materials is proposed, and a testing system based on surface-bonded low-profile piezoceramic transducers is designed and built.(2) The experimental research on micro fatigue crack detection in plate-like metallic structures is performed and verified.(3) By analyzing the VAM response signals in time-frequency domain and AM-FM demodulating via the above proposed algorithm, the coupling characteristics of the two wave fields for VAM technique are determined. Then a damage index for assessing the crack progression is extracted based on the AM features of the VAM responses. Its validity is confirmed by experimental signals.Experimental researches show that, the VAM technique can detect micro fatigue crack with high sensitivity; both amplitude and frequency modulations are present in the acoustical responses for a cracked plate-like metallic structure; the damage index extracted from the AM features of VAM responses can be used to track and assess the damage state.4. A VAM-based damage location and imaging approach for plate-like metallic structures is proposed and verified.Based on the sparse distributed transducer array technique and the delay-and-sum focusing algorithm, a damage imaging method for plate-like metallic structures is proposed and verified. Experimental results show that, from the generated image the location and other characteristics of damages can be determined in an intuitive manner.5. A signal refocusing and enhanced damage imaging method based on time reversal (TR) of VAM signals is investigated and experimentally verified.(1) The transmission modes of time reversed linear guided waves in a plate-like structure are analyzed by means of transfer function. The time reversal spatial-temporal adaptive focusing effect is investigated in theory. Meanwhile, the time reversal focusing effect of nonlinear guided waves is discussed qualitatively.(2) The refocusing effect of VAM-TR process is verified experimentally by means of laser scanning. And the enhanced imaging strategy based on VAM-TR using position-fixed transducer array is discussed.Studies show that, the acoustical energy can be focalized on damage position adaptively through the VAM-TR process, therefore the received signals are amplified and the signal-to-noise ratio increases. According to the spatial distribution of time reversal acoustical energy, an enhanced image representation can be obtained.