Investigations On Damage Detection Of The PE/PE Self-Reinforced Composites Based On Acoustic Emission Technique

Ultra-high molecular weight polyethylene (UHMWPE) fibers have excellent properties, such as high tensile strength, high tensile modulus, high toughness, well abrasion resistance and fine anti-impact performance. The self-reinforced composites of UHMWPE can obtain excellent integrated performances and have been widely used in wide engineering fields. However, the imperfect processing systems and unreasonable processing conditions may result in composites defects, such as pores, cracks, slack and ill-cohesion of the interface. The factors such as loading, mechanical damage, fatigue, creep and other complex conditions which composites endured in the process of active service may cause the damage of the composite, and the serious damage will lead the entire failure of the material which may cause serious accidents. So it is necessary to check the composite in the process of active service in order to make out the safety alarm and avoid the occurrence of accidents.Acoustic emission technique is one of the best methods to detect the damage process of the composites in real time. It may play an important role in avoiding the occurrence of the mass accidents. Many research have been carried out by acoustic emission, which mainly focus on the thermosetting materials such as carbon fibre/epoxide or glass fibre/epoxide resin composite, however they were scarcely about the detection of damage process of thermoplastic materials. With the characteristics of poor plasticity and lower elastic deformation, thermosetting materials can produce marked signals in the process of damage, however that of thermoplastic materials are different evidently because of theirs better plasticity and higher elastic deformation. The damage process of composites are accompanied with the occurring and propagating of signals, the complex composite structure also affect the propagation of signals, and the different damage mode of each time exhibits various signal characteristics. So, to detect the damage processes of the thermoplastic materials in real-time, master the signal active instances of PE/PE self-reinforced composites of various structures, study the propagating characteristics, analyze waveform of different signals are important to reveal the damage mechanism of composites, locate the signal source and confirm the characteristic of source.At the beginning of this dissertation, five types of specimen which include fiber reinforcement material, matrix material,[0°], [+45°/-45°] and [90°]PE/PE laminates, are tested and monitored by acoustic emission in order to study tension damage process. Then, different types of damage characteristics of specimen were distinguished by AE parameter analysis. The results revealed that various specimen showed distinct different stress-strain relationship during tension process. Meanwhile, the activities and characteristics (including amplitude, energy, duration time) of AE signal are also different between different specimens. Furthermore, according to characteristic of AE parameter, the process of material damage can be divided into several phases, AE signals of each phase is distinct different. At the same time, using SEM to check fracture surface, we found that the damage of composite laminates presented many mechanisms such as fiber breakage, matrix crack, interface debonding and delamination. As a result, AE detecting is an effective method on revealing damage evolution mechanism and damage modes during tension process, which can reflect the damage characteristic of the composite clearly and intuitionally.To overcome the shortcoming of the method of parameter analysis, the technique of mode acoustic emission which was acknowledged is used to study acoustic emission signals. It is based on the plate wave theory and mostly used for detecting the structure of board. In this paper, the propagating model of lamb wave is built based on plate theory and the propagating characteristics in the plate are predicted. According to the different elastic characters, we built the propagating model of lamb wave in unidirectional and symmetric cross-ply laminates and given the lamb wave dispersion equations of lowest symmetric and the lowest antisymmetric modes. In accordance with the control equations and boundary conditions, the lowest mode dispersive curves are predicted in the PE/PE composite laminates with different layered-up. Predictions show that the propagating performances of lamb waves in laminates are related to the propagating direction and the layered angle of the fibers, the velocity of lamb waves is relying on the elastic characters of the laminates.To grasp the propagating characteristics of signals in laminates would provide theory fundaments for the practical acoustic emission detecting. We can also learn what modes of the Lamb waves would occur in the laminates. Analysis shows that the lowest antisymmetrical mode of lamb waves A₀ plays an important role in laminates, which is just the research object of the acoustic emission detection. Experiments in this paper verified the theory predict. The signal source is located by wavelet transformation. The attenuation characteristics of signal in PE/PE composite laminates are investigated as well, which provides a foundation for the arrangement of sensors.Using lead breaking experiments, the attenuation performances of acoustic emission signal are obtained in PE/PE composite laminates. The role which Gabor wavelet analysis played in the signal time-frequency-energy analysis is discussed. The correctness of control equation of laminate theory is also verified by experiments of Gabor wavelet analysis. The issues of source location related to lead breaking signals and laminates fracture signals are investigated. Result shows that signals attenuated seriously in the laminates, especially in the distance of 50mm scope away from signal source. This is because most of the higher frequency data was filtered, while lower frequency (<200 KHz) signal with relatively long wave can propagate within a wide range. The experiments provide foundation of the sensors arrangement for the source location t.Gabor wavelet was applicable to analysis of transient non-stationary signals, and its wavelet coefficient peak value corresponding to arrival time of signals. Gabor wavelet analysis can verify the correctness of control equation and can provide a tool to precisely source location. Both lead breaking and the tension experiments show that Gabor wavelet transformation can determine the source location precisely.Finally, acoustic emission signals are identified by wavelet analysis. Approach of multi-resolution Laplace correlation filtering is proposed to identify both the simulated signal with single-sided damped and signals of fiber fracture in the process of composites damage and the signal parameters are identified. We also propose a method of multi-resolution wavelet time-frequency-energy analysis to study acoustic emission signal of composite material. Experiments are carried out to distinguish the signals produced from specimens of matrix material, [0°], [+45°/-45°] and [90°] PE/PE laminates. Different damage mechanisms of composite laminates are distinguished and summarized.Analysis shows that method of multi-resolution Laplace correlation filtering can accurately identify the parameters of transient signal, while reduce the data of correlation filtering and improve calculating speed at the same time. It can also effectively distinguish acoustic emission signal of the damage in the composites. Wavelet multi-resolution analysis and its reconstruction can extract useful data in the original signal and get rid of noise which goes against analysis. Wavelet time-frequency-energy analysis can effectually distinguish different damage mechanisms of composite laminates. There are several fracture mechanisms of composite laminates, while the fiber lay-ups as well as its layered angle cause major influence to the fracture mode during tension. AE signals are complicated during the tension of PE/PE laminates and characteristic of acoustic emission signals are significantly influenced by damage mechanism. Damage signals were different in the character of waveform and spectrum.