Locating acoustic emission sources in complex structures using Gaussian processes

A standard technique in the field of non-destructive evaluation is to use acoustic emissions to characterise and locate the damage events that generate them. The location problem is typically posed in terms of the times of flight of the waves and results in an optimisation problem, which can at times be ill-posed. A method is proposed here for learning the relationship between time of flight differences and damage location using data generated by artificially stimulated acoustic emission (AE)—a classic problem of regression. A structure designed to represent a complicated aerospace component was interrogated using a laser to thermoelastically generate AE at multiple points across the structure's surface. Piezoelectric transducers were mounted on the surface of the structure, and the resulting waveforms were recorded. A Gaussian process (GP) with RBF kernels was chosen for regression. Since during AE monitoring not all events can be guaranteed to be detected by all sensors, a GP was trained on data for all possible combinations (subsets) of sensors. The inputs to the GPs were the differences in time of flight between sensors in the set, and the targets were the locations of the source of ultrasonic stimulation. Subsequent (test) data points were located by every possible GP, given the active sensors. It is shown that maps learned on a given structure can generalise effectively to nominally identical structures.     

On-line data acquisition system for acoustic emission signals

An electronic system for processing and recording some very important parameters of acoustic emission signals at a relatively high speed is described. The system is applicable to almost any process in which acoustic emission is implemented for quality control. Data processing is done in real time. The processed data are ring-down count (threshold crossing), energy, duration, and time of occurrence of each acoustic emission burst.     

A criterion for evaluating the brittle fracturing of glass using acoustic emission signals

Some features of the fracturing of silicate glass materials are considered. The brittle fracturing of glasses is shown to begin under small applied forces that are characterized by low-amplitude acoustic emission (AE) signals. The further propagation of a fracture is shown to be accompanied by a change in the amplitude-frequency characteristics of AE signals, which we propose to describe with a continuous wavelet transform. A criterion for estimating the brittle fracture of glass with the use of attendant AE signals is proposed.     

Source location of acoustic emission in diesel engines

This paper is the third in a series developing methods of mapping acoustic emission (AE) signals and wave propagation in engines and focuses on source location techniques for the multi-source signals on relatively complex structures typical of machinery applications. Two source location techniques, a traditional wave velocity-based and an AE energy-based technique, using triangular sensor arrays, are used to locate source positions on the cylinder head of a 74 kW diesel engine using simulated sources (pencil lead break) and real sources (e.g. injectors (INJs) and exhaust valves during engine running).Source location using both techniques is demonstrated on the cylinder head of a 74 kW four-stroke diesel engine. The velocity-based technique uses AE wave speeds and time-of-flight (wave arrival time) to locate source position and is found to be most effective for single source signals with a sharp rising edge and good signal to noise ratios. The energy-based technique is based on a simple absorption attenuation model and was found to be useful for multiple source signals such as INJ signals, although structure-specific attenuation coefficients need to be measured for accurate source location.     

Developing and testing a laboratory system for recording and analysis of acoustic emission

A laboratory system for recording and analyzing acoustic emission was designed and experimentally investigated (tested). A block diagram and the operation algorithm of the system are presented. The relationship between the results of low-frequency, high-frequency, and statistical processing of recorded data and the physical features of input signals is shown using as an example simulated acoustic signals in the various units of the designed system. The system was tested on real test objects in the case of three-point bending of nitrided steel specimens with various thicknesses of the hardened surface layer. It is shown that crack formation leads to emission of signals with an amplitudes of up to 4 V, whereas during deformation in the absence of pronounced cracking, the amplitude of the amplified signal does not exceed 0.5 V.     

Analysis of experimentally generated friction related acoustic emission signals

Acoustic emission signals generated by sliding friction between two flat steel surfaces are characterized. A test fixture to simulate the reciprocating motion between the two surfaces under controlled conditions is developed. Sliding friction under several combinations of surface roughness, relative velocity, and normal pressure was examined. Wideband AE sensors and instrumentation were used for acquiring and analyzing the acoustic emission signals. Acoustic emission events occurred primarily during the slip portion of the stick-slip cycles. AE waveform features obtained during these experiments were indicative of the tribological conditions. Frequency components in excess of 700 kHz were seen during these experiments. The characteristics of the experimentally observed acoustic emission signals were in general agreement with earlier numerical predictions. Friction related acoustic emission signals were distinguishable from those from other sources such as fatigue crack growth. The characterization of friction related acoustic emission signals is likely to be of value in many tribological and structural health monitoring applications.     

Method of coupled dipoles for localization of the sources of acoustic emission

A technique for minimizing errors in location of the sources of acoustic emission in bedrock via the use of the method of coupled dipoles is considered. Specific calculations are made to show that, if the system of equations is selected improperly, the method does not have any practical value.     

Linear discriminant function analysis of acoustic emission signals for cutting tool monitoring

A principal setback to automation of the machining process is the inability to completely monitor the condition of the cutting tool in real time. Whereas several of the techniques developed to date are useful in specific applications, no universally applicable sensor is yet available.Acoustic emission is one of the most promising techniques to be recently developed for on-line cutting tool monitoring. However, signal analysis is still an area that requires further investigation to enhance the potential of acoustic emission. For this purpose, frequency-based pattern recognition concepts using linear discriminant functions have been used in analysing acoustic emission signals generated during machining to distinguish between different signal sources, specifically chip formation, tool fracture, and chip noise. Five features were used for classification in the frequency range of 100 kHz to 1 MHz, with each feature consisting of a 20 kHz bandwidth, and were selected using the class mean scatter criterion. The coefficients of the discriminant functions were obtained by training the system using signals generated by each of the sources of interest. An AISI 1018 steel was machined using a titanium carbide-coated cutting tool. Cutting speeds ranged from 200 to 800 ft/min (1 to 4 m/sec) with feed rats of 0·0005 to 0·0075 in/rev (0·0133 mm/rev to 0·191 mm/rev) and depth of cut 0·17 in (4·32 mm). The results show a successful classification rate of 90% for tool breakage, while those for chip formation and chip noise were 97 and 86% respectively.     

A new acoustic emission source location technique based on wavelet transform and mode analysis

For wave propagation in dispersive media, the arrival time of the acoustic emission signal to the sensor is dependent on the setting of the threshold voltage, which results in the inaccuracy of the acoustic emission location. Based on the wavelet transform and the theory of modal acoustic emission, a new method is proposed to improve the accuracy of acoustic emission source location. It is believed that the acoustic emission signal propagation in the structure has the characteristics of multi-mode and dispersion, and the acoustic emission source location should use the arrival time to sensors obtained from the output signals not only at the same mode but also at the same frequency. The wavelet transform is used to resolve the problem. By utilizing the time-frequency data of the wavelet, the frequency-dependent arrival time traveling is easily obtained; by numerical computation of the wave’s propagation in structure, the group velocity of the guided mode is also obtained, therefore the accuracy source location is realized. The acoustic emission source location experiments were conducted in a thin steel plate and results show that the technique is an effective tool for acoustic emission source location. __________ Translated from Chinese Journal of Scientific Instrument, 2005, 26(3) (in Chinese)     

An application of the compressive sampling method for compressing and processing acoustic signals

The compressive sampling method for signals and its application is considered. The simulation results in a Matlab package of the compressive sampling algorithm are given based on the example of an optical image and echo signal.     

Design of a piezoelectric transducer cylindrical phase modulator for simulating acoustic emission signals

To conveniently carry out the pipeline leak experiment in a laboratory, leak acoustic signals are simulated by using the converse piezoelectric effect of a piezoelectric transducer (PZT) cylindrical phase modulator. On the basis of the piezoelectric equations and electromechanical equivalence principle, the transfer function of a PZT cylindrical phase modulator is delivered. A PZT cylindrical phase modulator is designed, and the numerical simulation is conducted. Results prove that the PZT cylindrical phase modulator can effectively simulate leak acoustic emission signals when the frequency is lower than 25 KHz. __________ Translated from Journal of Beijing University of Technology, 2006, 32(8): 683–687 [译自: 北京工业大学学报]     

声发射信号处理在轮轴故障检测中的应用研究

声发射（Acoustic Emission）可以定义为物体或材料内部迅速释放能量而产生瞬态弹性波的一种物理现象,是材料或结构受内力或外力作用产生变形或断裂,以弹性波的形式释放出应变能的结果。在基于虚拟仪器的轮轴故障检测分析平台中,可实现对声发射信号的时域分析、频谱分析、参数分析、小波去噪等主要功能。通过该平台,对具有典型状态的滚动轴承的声发射信号进行了试验研究。试验研究结果表明,对于不同状态的轴承,所测取的声发射信号包络频率特征有着明显的不同,并且和理论特征频率有着一定的关系。     

Evaluation of grinding wheel loading phenomena by using acoustic emission signals

In the industrial manufacturing field, machining is a major process. Machining operations involve grinding, drilling, milling, turning, pressing, molding, and so on. Among these operations, grinding is the most precise and complicated process. The surface condition of the grinding wheel plays an important role in grinding performance, and the identification of grinding wheel loading phenomena during the grinding process is critical. Accordingly, this present study describes a measurement method based on the acoustic emission (AE) technique to characterize the loading phenomena of a Si₂O₃ grinding wheel for the grinding mass production process. The proposed measurement method combines the process-integrated measurement of AE signals, offline digital image processing, and surface roughness measurement of the ground workpieces for the evaluation of grinding wheel loading phenomena. The experimental results show that the proposed measurement method provides a quantitative index from the AE signals to evaluate the grinding wheel loading phenomena online for the grinding mass production process, and this quantitative index is determined via some experiments in advance in the same grinding environment to help the monitoring and controlling of the grinding process.     

火车轮裂纹声发射源定位技术研究

针对火车轮运行过程中受变载荷作用产生的裂纹具有的特征,对现有的裂纹声发射定位技术进行了优化.在硬件上采用更适用于车轮运行环境的安装方式,并确定了前向通道的各项参数.根据模态声发射理论,即AE信号在传播过程中具有的频散现象和多模态特性,通过对同一模式波到达各传感器的时差信号的精确分析,采用时差法和面域定位相结合的方法进行源位置的计算.由于采用多种技术进行分析,结果将更为精确.     

基于小波变换的管道泄漏声发射检测信号处理

针对管道泄漏声发射检测信号中含有大量噪声导致定位精度低这一实际,研究了基于小波变换的声发射检测信号降噪技术,利用小波函数阀值降噪法对泄漏系统所采集的信号进行滤波处理,以保证采集到的瞬态信号有效性,并采用互相关分析法对泄漏点定位.10种工况下实验结果表明,采用小波变换对声发射检测信号处理处理后,管道的定位精度在5％之内,满足工程实际需要.     

Best wavelet basis for wavelet transforms in acoustic emission signals of concrete damage process

It is critical to select the best wavelet basis for implementing wavelet algorithms. Firstly, the mathematical characteristics of the eight kinds of wavelet basis commonly used in the field of engineering are summarized and compared based on the analysis of the features of acoustic emission signals from concrete damage process. Secondly, wavelet basis suitable for acoustic emission signals in concrete damage are given by the theoretical analysis. Finally, the best wavelet basis for wavelet transforms in acoustic emission signals of concrete damage process are obtained by calculating the energy entropy. The results show that the energy entropy can be used to select the wavelet basis for acoustic emission signals during the process of concrete damage. Among them, the wavelets of Symlets can be used as the optimum wavelet basis.     

A hardware and software system for measuring the angular structure of the acoustic fields in acoustic tomography

A hardware and software system for measuring the angular structure of acoustic fields using the vector-phase methods for processing the complex M-sequence probing signals is described. The results of testing the system are given. A unique relationship between the change in the temperature condition on the stationary acoustic route and the times and angles of arrival of the acoustic energy is determined under controlled hydrological conditions. It is shown that, using vector receivers in tomographic schemes, it is possible to obtain an independent additional parameter of the pulse response of the diagnosed waveguide and increase thereby the efficiency of reconstruction of hydrophysical fields from acoustic probing data.     

The relationship between acoustic emission signals and cutting phenomena in turning process

The development of intelligent manufacturing by using machine tools is advancing in leaps and bounds. To maintain accuracy in machining and in the interests of fail-safe operation, monitoring of the cutting state or the final machining is very important. Acoustic emissions (AE) comprise elastic stress waves produced as a result of the deformation and fracture of materials. By measuring the AE generated during a turning process, it is possible to estimate the state of the machining operation. The correlation between cutting phenomena and AE in a turning process was examined experimentally by using a steel workpiece and a cermet tool in a numerically controlled turning process. The process of formation of chips, the types of chip, and the shear angle all markedly affected the AE signals. There was a strong negative correlation between the shear angle and the AE signal level. Similar results were obtained for various feed rates and for workpieces of various degrees of hardness. Correlations related to surface roughness and to tool wear are also described that permit the evaluation of the state of the turning process.     

基于声发射信号衰减特性的材料成分分数检测

针对传统的材料成分分数检测方法需损坏被测材料这一问题,提出一种基于声发射信号衰减特性的材料成分分数检测新方法。选取微晶石墨/聚乙烯醇(PVA)复合材料为对象,从声发射原信号和声发射信号特征小波包两个方面,建立其衰减特性与材料成分分数间的关系,并对该关系进行验证和两方面的验证结果的对比。结果表明:本研究选取信号的特征小波包为125~171.85 k Hz频段时,对于原矿微晶石墨/PVA复合材料以此特征小波包计算原矿微晶石墨质量分数误差为2.2%,比用原信号计算误差低1.6%;对于提纯微晶石墨/PVA复合材料以此特征小波包计算提纯微晶石墨质量分数误差为2.4%,比用原信号计算误差低2.0%,检测结果与材料实际成分分数吻合度较高。因此,该方法为材料成分分数无损检测提供了一种新的研究手段。