Trouble diagnosis of the grinding process by using acoustic emission signals

The focus of this study is the development of a credible diagnosis system for the grinding process. The acoustic emission signals generated during machining were analyzed to determine the relationship between grinding-related troubles and characteristics of changes in signals. Furthermore, a neural network, which has excellent ability in pattern classification, was applied to the diagnosis system. The neural network was optimized with a momentum coefficient (m), a learning rate (a), and a structure of the hidden layer in the iterative learning process. The success rates of trouble recognition were verified.     

Characterization of the grinding process by acoustic emission

The properties of a ground surface can be estimated on-line during manufacturing based on the analysis of acoustic signals emitted by the grinding process. This possibility is demonstrated using an experimental system comprising an external grinding machine, a data acquisition unit and an artificial neural network. In the initial phase of system application, an empirical model of the grinding process is formed in the memory of the neural network by self-organized learning driven by empirical data consisting of the acoustic emission spectrum and a surface roughness correlation function. After learning, the system applies the model to estimate the correlation function of the surface profile from the input acoustic emission spectrum. For this purpose, non-parametric regression, based on the conditional average estimator, is utilized. Experiments were done on the grinding of hardened steel workpieces by a corundum wheel. During formation of the model, the surface profile and its correlation function were determined off-line, while in testing system performance the surface correlation function was estimated on-line from the acoustic emission spectrum. With respect to the estimation error, three characteristic periods of the process were observed corresponding to grinding with a newly dressed, slightly worn, and worn out wheel. The best estimation is obtained during grinding by a slightly worn wheel.     

Investigation of acoustic emission signals under a simulative environment of grinding burn

Grinding burn is a common phenomenon of thermal damage that has been one of the main constraints in grinding in respect of high efficiency and quality. An acoustic emission (AE) technique was tried in an attempt to identify grinding burn on-line. However, the AE features of grinding burn are relatively weak and are easily obscured by other AE sources. This paper presents an investigation of the AE features of the thermal expansion induced by laser irradiation, which was designed to simulate grinding thermal behaviour. By using wavelet packet transforms, AE features at the grinding burn temperature can successfully be extracted without other mechanical interferential factors. Such thermal AE features provide a firm foundation for analysing and monitoring the AE features of grinding burn.     

Fuzzy pattern recognition of AE signals for grinding burn

Grinding burn is a common phenomenon of thermal damage that has been one of the main constraints in grinding difficult-to-machine materials. Grinding burn damages materials and degrades properties, by causing tensile residual stresses or microfractures in the workpiece surface. Numerous methods have been proposed to identify grinding burn. However, the main problems of current methods are their sensitivity and robustness. This paper describes a new method of grinding burn identification with highly sensitive acoustic emission (AE) techniques. The wavelet packet transform is used to extract features from AE signals and fuzzy pattern recognition is employed for optimising features and identifying the grinding status. Experimental results show that the accuracy of grinding burn recognition is satisfactory.     

A study on ELID ultra precision grinding of optical glass with acoustic emission

ELID grinding of BK7 glass and Zerodur was investigated using acoustic emission. Experiments showed that the contacting area between the wheel and workpiece in a grinding process was critical to influence wheel loading for a fine grit size resin-bonded cup wheel. ELID can be used for efficient material removal when the wheel/workpiece contacting area is large. Correlations were observed between the dressing intensity on the ELID wheel and the detected AE signals. Aggressive ELID dressing parameters for grinding with finer grit size wheels corresponded to a lower AE level. With an increase in the processing time of an ELID wheel, low and stable AE amplitudes became large with fluctuations due to the deterioration of the grinding wheel. Results indicate that the AE sensing technique has the potential to be adopted as an effective method for monitoring an ultra precision grinding process, identifying the condition of the grinding wheel and investigating the mechanism of ELID grinding.     

Determining the stress intensity factor of a material with an artificial neural network from acoustic emission measurements

An artificial neural (ANN) network was trained to recognize the stress intensity factor in the interval from microcrack to fracture from acoustic emission (AE) measurements on compact tension specimens. The specimens were made from structural steel SWS490B whilst the ANN had a 5-14-1 structure. The number of neurons in the input layers was five inputs of the AE parameters such as ring-down counts, rise time, energy, event duration and peak amplitude. The performance of the ANN was tested using a specific set of the AE data. The ANN is a promising tool for predicting the stress intensity factor of material using AE data.     

Application of acoustic emission to seeded gear fault detection

Acoustic emission (AE) is gaining ground as a non-destructive technique for health diagnosis on rotating machinery. There are vast opportunities for development of the AE technique on various forms of rotating machinery, including gearboxes. This paper reviews some recent developments in application of AE to gear defect diagnosis. Furthermore, an experimental investigation that examines the effectiveness of AE for gear defect identification is presented. It is concluded that application of the AE technique to seeded gear defect detection is fraught with difficulties. In addition, the viability of the AE technique for gear defect detection from non-rotating components of a machine is called into question.     

Modal analysis of acoustic emission signals from CFRP laminates

As a result of its continuous and in situ detection capabilities, the acoustic emission (AE) technique is the prime candidate for damage monitoring in loaded composite structures. None of the AE analysis techniques used in laboratory studies has, however, proven to be capable of consistently dealing with the difficulties encountered in larger structures: large amount of data, the elimination of noise sources and the influence of wave propagation effects (attenuation, dispersion). This work will use the modal acoustic emission (MAE) technique as a more intelligent and efficient way of analysing AE results. AE waveforms obtained during tensile and bending testing of CFRP laminates will be presented. It will be demonstrated how taking into account the modal nature of AE waves can in future lead to more quantitative and accurate results.     

Precision manufacturing process monitoring with acoustic emission

Current demands in high-technology industries such as semiconductor, optics, MEMS, etc. have predicated the need for manufacturing processes that can fabricate increasingly smaller features reliably at very high tolerances. In situ monitoring systems that can be used to characterize, control, and improve the fabrication of these smaller features are therefore needed to meet increasing demands in precision and quality. This paper discusses the unique requirements of monitoring of precision manufacturing processes, and the suitability of acoustic emission (AE) as a monitoring technique at the precision scale. Details are then given on the use of AE sensor technology in the monitoring of precision manufacturing processes; grinding, chemical–mechanical planarization (CMP) and ultraprecision diamond turning in particular.     

改进的神经网络技术在声发射定位中的应用

针对时差定位法受很多因素影响的弊端，将神经网络技术应用到声发射源定位中。提取最能揭示声发射源的特征参数和运用主元分析技术来降低输入样本的数量；采用增加隐含层神经元个数探讨它们的误差变化来确定隐含层；运用附加动量法和优化选取初始阈值等措施进行网络设计。将设计好的网络运用到实例中，通过与实际缺陷位置的比较，结果表明，选择合理的网络结构和输入参数可准确定出结构损伤位置，且精度有较大的提高，计算更简单有效。     

基于声发射和神经网络的数控机床刀具故障诊断

分析了数控刀具的切削状态,介绍了声发射检测系统和神经网络技术,对刀具切削状态信息声发射检测的可行性和神经网络技术智能诊断方法进行了分析,并通过数控机床刀具故障诊断实例,验证了通过声发射提取刀具切削状态方法的有效性和通过神经网络智能诊断技术检测刀具切削状态方法的正确性。     

Use of acoustic emission technique for the early detection of aluminum alloys exfoliation corrosion

Exfoliation corrosion of aluminum alloys is a form of localized corrosion which affects many industries, specially aeronautics. The study of this corrosion mode using only electrochemical techniques is not fully efficient for the detection and control on line of this phenomenon. Therefore, we developed a non-destructive testing technique based on the acoustic emission recordings in order to follow-up this form of corrosion on aluminum alloys. Indeed, recent works have shown the interest of the acoustic emission for the detection, the monitoring and the localization of pitting corrosion on aluminum alloys. This pitting corrosion phenomenon is currently well understood and the experimental methodology acquired during that study is transposed to the study of exfoliation corrosion of aluminum alloys.The present study is conducted on two aluminum alloys: (Al 2024 T3, and Al 7449 T6 and T7). Samples are immersed 4 days in the modified ASTM STP 1134 saline solution. Observations of the structures after tests show that the exfoliation corrosion sensitivity of alloy 7449 T6 is more important than for alloy 7449 T7 which exhibits only the presence of small and non-occluded pits. Very severe exfoliation corrosion was also observed on Al 2024 T3, but after a longer immersion time or in a more acid solution.The recording of the acoustic emission activity shows evident links between this activity and the exfoliation corrosion rate. The analysis of the signal's characteristics reveals a population corresponding to the release of hydrogen bubbles. A few more energetic signals have also been observed. Their source can be either, the cracking resulting from the separation of sheets of metal, or the development and evolution of hydrogen bubbles formed inside blisters during exfoliation corrosion.     

基于FNN智能型磨削参数决策系统

本文结合对磨削机理,模糊逻辑,神经网络的研究提出了一种基于ＦＮＮ的智能型磨削参数决策系统。在这个系统中,首先由上层专家系统决策出初始磨削参数,然后由两个ＦＮＮ单元对磨削参数进行修正。其中ＦＮＮ１利用磨削结果对磨削参数进行修正,ＦＮＮ２利用当前砂轮磨损状况对磨削参数进行修正。．     

Influence of diurnal and seasonal temperature variations on the detection of corrosion in reinforced concrete by acoustic emission

Chloride rich reinforced concrete prisms were coupled to chloride-free prisms and exposed to diurnal and seasonal temperature cycles typical of those found in the UK. Acoustic emissions (AE) and galvanic currents were continuously monitored and correlated with ambient temperature. AE and galvanic currents were found to emulate the evolution of temperature in the diurnal cycles, although no specific relationship between AE and galvanic current could be obtained. The influence of seasonal variations in galvanic current had no obvious influence on AE Energy per second over the range of corrosion rates studied. The findings suggest that AE is more sensitive to short term (diurnal) changes in corrosion rates than the longer (seasonal) effects. It was hypothesised that this is due to transitory changes in the internal microclimate of the concrete.     

Kaiser effects in acoustic emission from composites during thermal cyclic-loading

Kaiser effects in acoustic emission (AE) behavior of composite laminates under repetitive thermal cyclic-loads are quantitatively analyzed to identify AE source mechanisms. The repetitive thermal loads brought about a large reduction, i.e. an exponential decrease, in AE total ring-down counts and AE amplitudes. It was thought that generation of most thermo-AE events during the first thermal cycle was not caused by crack propagation, but by secondary micro-fracturing due to abrasive contact between crack surfaces. For subsequent thermal cycles, on the other hand, a small number of weak thermo-AE events were generated due to frictional sliding contact. Such behavior of thermo-AE showed different characteristics according to specimen types and the maximum temperature in the thermal load cycles.     

A reference standard for the development of acoustic emission pipeline leak detection techniques

A reference standard was constructed for setting up and evaluating AE equipment to be used in pipeline leak detection. The reference standard comprises a short length of 2-inch diameter piping with facilities for introducing several kinds of controlled leaks. The reference standard proved very valuable not only for checking out equipment, but also for characterizing source mechanisms as part of an integrated approach to quantitative AE leak detection/location technology. The effects of pressure and air injection were measured for thread leaks on the order of 0.1 gal h⁻¹, a leakage rate that is important in the context of environmental protection regulations. Taking this knowledge to the field, a thread leak of only 0.014 gal h⁻¹ was successfully detected and located by injecting nitrogen into the line at 25 psi. This leak was located with 1 foot accuracy, using two different location techniques and 25-foot sensor spacing. It is envisioned that in the future, AE inspectors in the field will make systematic use of several two-phase flow processes and soil enhancement mechanisms that are being characterized by means of this new reference standard.     

Application of acoustic emission in drilling of composite laminates

Acoustic emission (AE) technique was used to characterise drilling of composite laminates. Uni-directional glass fibre reinforced plastic (GFRP) laminates consisting of 12-layers and 16-layers (0/90)_s were drilled using a twist drill and the generated AE was monitored. Results of the investigations reveal that the complexion of the acoustic emission root mean square (AE-RMS) signal response changes from the drill entry to the exit thus giving an overall understanding about the different events that take place during drilling. Also, AE-RMS signal level increases with an increase in the applied thrust and further reveals that it is possible to evaluate the drill induced damages in composites through AE signal characterisation.     

Dynamic modeling of cutting acoustic emission via piezo-electric actuator wave control

The fundamental understanding of the dynamic behavior of acoustic emission signals in relation to machining process parameters plays an important role in the automation monitoring and control of metal cutting operations. This paper presents an analytical model for acoustic emission dynamics in orthogonal cutting with chip thickness variation. An analytical expression for acoustic emission generated in turning is established as an explicit function of the cutting parameters and tool/workpiece geometry. Based on the theoretical static cutting acoustic emission model, the generation of the RMS acoustic emission is formulated as the function of three process parameters, namely tool displacement, cutting speed, and rake angle. The incremental change of the RMS acoustic emission is related to the chip formation process in an elemental cutting area and it is characterized by the dynamic variation of these process parameters. The analysis of the RMS acoustic emission is then extended into the dynamic transfer function between tool displacement and RMS acoustic emission. Experimental results via simultaneous tool displacement control by piezo-electric actuator and measurement of RMS AE generated during cutting confirm the validity of the analytical acoustic emission model in orthogonal cutting process.     

Neural network based defect detection and depth estimation in TNDE

For many years, applications of the TNDE (Thermographic NonDestructive Evaluation) technique has been limited due to the complex non-linearity nature of related inversion problems such as defect depth estimation. Artificial neural networks have recently obtained success in revealing and providing quantitative information concerning defects in TNDE. In this paper, a three dimensional thermal model for non-homogenous materials such as carbon fiber reinforced plastic (CFRP) is first given. The modeling results are compared with the analytical solution based on Duhamel's theorem. Two back propagation neural networks (NN) as defect detector and depth estimator are then presented. Finally, simulated and experimental results are presented and discussed.     

Observations of acoustic emission activity during gear defect diagnosis

It is widely recognised that acoustic emission (AE) is gaining ground as a non-destructive technique (NDT) for health diagnosis on rotating machinery. The source of AE is attributed to the release of stored elastic energy that manifests itself in the form of elastic waves that propagate in all directions on the surface of a material. These detectable AE waves can provide useful information about the health condition of a machine. This paper reports on part of an ongoing experimental investigation on the application of AE for gear defect diagnosis. Furthermore, the possibility of monitoring gear defects from the bearing casing is examined. It is concluded that AE offers a complimentary tool for health monitoring of gears.