Strict-Feedback Backstepping Digital Twin and Machine Learning Solution in AE Signals for Bearing Crack Identification

Bearings are nonlinear systems that can be used in several industrial applications. In this study, the combination of a strict-feedback backstepping digital twin and machine learning algorithm was developed for bearing crack type/size diagnosis. Acoustic emission sensors were used to collect normal and abnormal data for various crack sizes and motor speeds. The proposed method has three main steps. In the first step, the strict-feedback backstepping digital twin is designed for acoustic emission signal modeling and estimation. After that, the acoustic emission residual signal is generated. Finally, a support vector machine is recommended for crack type/size classification. The proposed digital twin is presented in two steps, (a) AE signal modeling and (b) AE signal estimation. The AE signal in normal conditions is modeled using an autoregressive technique, the Laguerre algorithm, a support vector regression technique and a Gaussian process regression procedure. To design the proposed digital twin, a strict-feedback backstepping observer, an integral term, a support vector regression and a fuzzy logic algorithm are suggested for AE signal estimation. The Ulsan Industrial Artificial Intelligence (UIAI) Lab’s bearing dataset was used to test the efficiency of the combined strict-feedback backstepping digital twin and machine learning technique for bearing crack type/size diagnosis. The average accuracies of the crack type diagnosis and crack size diagnosis of acoustic emission signals for the bearings used in the proposed algorithm were 97.13% and 96.9%, respectively.

Propagation Characteristics of Acoustic Emission Signals in Stiffened Cylindrical Shells Based on the Multipath Propagation Model

A stiffener attached to a cylindrical shell strongly interferes with the propagation of the acoustic emission (AE) signal from the fault source and reduces the fault detection accuracy. The interaction of AE signals with the stiffener on the cylindrical shell is thoroughly investigated in this paper. Based on the proposed model of the AE signal propagating inside the cylindrical shell with a stiffener, the installation constraints for the sensor are derived, resulting in the separation of the direct signal, the stiffener scattering signal, and other signals in the time domain. On this basis, combinations of the excitation frequency and the stiffener height are simulated, and the reflection and transmission of the AE signal in each case are quantitatively characterized by the scattering coefficients. The results indicate that there is a “T-shaped” transformation of the signal at the stiffener, which evolves into a variety of other modes. Moreover, the reflection and transmission coefficients of the incident AE signal are displayed as a function of the excitation frequency and the height of the stiffener. In addition, the accuracy of the scattering coefficients obtained from the numerical simulations is verified by experiments, and a good consistency between simulation results and experiment results is presented. This work illustrates the propagation characteristics of AE signals in a cylindrical shell with a stiffener, which can be used as guidance for optimizing the spatial arrangement of sensors in AE monitoring.

Long-Range Correlations and Natural Time Series Analyses from Acoustic Emission Signals

This work focuses on analyzing acoustic emission (AE) signals as a means to predict failure in structures. Two main approaches are considered: (i) long-range correlation analysis using both the Hurst (H) and the Detrended Fluctuation Analysis (DFA) exponents, and (ii) natural time domain (NT) analysis. These methodologies are applied to the data collected from two application examples: a glass fiber reinforced polymeric plate and a spaghetti bridge model, where both structures were subjected to increasing loads until collapse. A traditional (AE) signal analysis is also performed to reference the study of the other methods. Results indicate that the proposed methods yield a reliable indication of failure in the studied structures.

The Use of Wavelet Transform To Evaluate The Sensitivity of Acoustic Emission Attributes To Variation of Cutting Parameters in Milling Aluminium Alloys

Appropriate acquisition and assessment of the dominant acoustic emission (AE) signal attributes generated under various experimental cutting conditions may provide significant knowledge. Consequently, it enhances the efficiency in manufacturing process monitoring and control. However, according to the literature, a lack of information was noticed on the behavior of AE signal attributes under various cutting conditions. Considering that milling is among the most widely used machining operations, the aim of this investigation is to acquire adequate knowledge about interactions between cutting parameters and their direct and indirect effects on the obtained AE signals attributes from the milling process. In the course of this work, the effects of cutting conditions on the attributes calculated from wavelet transform (WT) of AE signals will be presented. WT signal processing was conducted with five models of mother wavelets, and appropriate decomposition numbers were deployed. The approximated signal attributes obtained from each decomposition were assessed. According to signal processing and statistical calculations, cutting speed, feed rate, and coating significantly impacted the variation of AE signal attributes. Also, the most sensitive AE signal attributes and decompositions were rms, std, entropy and energy, and 2nd and 6th decompositions, respectively. The outcome of this work can be integrated into advanced artificial intelligence (AI) approaches to implement real-time monitoring of manufacturing processes.

Classification of acoustic emission signals in wood damage and fracture process based on empirical mode decomposition, discrete wavelet transform methods, and selected features

The nondestructive testing technology of generated acoustic emission (AE) signals for wood is of great significance for the evaluation of internal damages of wood. To achieve more accurate and adaptive evaluation, an AE signals classification method combining the empirical mode decomposition (EMD), discrete wavelet transform (DWT), and linear discriminant analysis (LDA) classifier is proposed. Five features (entropy, crest factor, pulse factor, margin factor, waveform factor) are selected for classification because they are more sensitive to the uncertainty, complexity, and non-linearity of AE signals generated during wood fracture. The three-point bending load damage experiment was implemented on sample wood of beech and Pinus sylvestris to generate original AE signals. Evaluation indexes (precision, accuracy, recall, F1-score) were adopted to assess the classification model. The results show that the ensemble classification accuracies of two tree species reach 94.58% and 90.58%, respectively. Moreover, compared with the results of the original AE signal, the accuracy of the AE signal processed by the methods proposed is increased by 27.68%. It indicates that the EMD and DWT signal processing methods and selected features improve the classification accuracy, and this automatic classification model has good AE signal recognition performance.

Prediction of the Stability of the Loaded Rock Based on the Acoustic Emission Characteristics of the Loaded Rock Based on Data Mining

The rock masses that occur in nature are damaged and unstable due to the impact of rock burst, coal and gas outbursts, and other human mining activities, posing a major threat to human life and safety. In the light of the early warning of the danger of the loaded rock mass, this paper adopts acoustic emission (AE) device to analyze the AE signal characteristics and damage laws of the loaded rock under different stress levels. Then, based on the AE signal characteristics of the loaded rock, data mining technology is used to construct a model to predict the failure and instability of the loaded rock mass and, finally, verify the reliability of the prediction model based on data mining. The results show that the AE signal characteristics of red sandstone under uniaxial load are related to the magnitude of the bearing stress. Before the plastic deformation stage, the AE energy and the cumulative count per second are both small. After the loaded rock enters the plastic deformation stage, the AE energy and the cumulative count per second both increase sharply. After the AE energy is greater than 500 mV ∗ ms and the cumulative count per second is greater than 150, the loaded rock mass will issue an early warning signal. The research results can provide a reference value for the safe production of the project site and the dangerous early warning of the loaded rock mass.

Acoustic Emission Wave Velocity Measurement of Asphalt Mixture by Arbitrary Wave Method

The wave velocity of acoustic emission (AE) can reflect the properties of materials, the types of AE sources and the propagation characteristics of AE in materials. At the same time, the wave velocity of AE is also an important parameter in source location calculation by the time-difference method. In this paper, a new AE wave velocity measurement method, the arbitrary wave (AW) method, is proposed and designed to measure the AE wave velocity of an asphalt mixture. This method is compared with the pencil lead break (PLB) method and the automatic sensor test (AST) method. Through comparison and analysis, as a new wave velocity measurement method of AE, the AW method shows the following advantages: A continuous AE signal with small attenuation, no crosstalk and a fixed waveform can be obtained by the AW method, which is more advantageous to distinguish the first arrival time of the acoustic wave and calculate the wave velocity of AE more accurately; the AE signal measured by the AW method has the characteristics of a high frequency and large amplitude, which is easy to distinguish from the noise signal with the characteristics of a low frequency and small amplitude; and the dispersion of the AE wave velocity measured by the AW method is smaller, which is more suitable for the measurement of the AE wave velocity of an asphalt mixture.

Acoustic Emission-Based Condition Monitoring and Remaining Useful Life Prediction of Hydraulic Cylinder Rod Seals

The foremost reason for unscheduled maintenance of hydraulic cylinders in industry is caused by wear of the hydraulic seals. Therefore, condition monitoring and subsequent estimation of remaining useful life (RUL) methods are highly sought after by the maintenance professionals. This study aimed at investigating the use of acoustic emission (AE) sensors to identify the early stages of external leakage initiation in hydraulic cylinders through run to failure studies (RTF) in a test rig. In this study, the impact of sensor location and rod speeds on the AE signal were investigated using both time- and frequency-based features. Furthermore, a frequency domain analysis was conducted to investigate the power spectral density (PSD) of the AE signal. An accelerated leakage initiation process was performed by creating longitudinal scratches on the piston rod. In addition, the effect on the AE signal from pausing the test rig for a prolonged duration during the RTF tests was investigated. From the extracted features of the AE signal, the root mean square (RMS) feature was observed to be a potent condition indicator (CI) to understand the leakage initiation. In this study, the AE signal showed a large drop in the RMS value caused by the pause in the RTF test operations. However, the RMS value at leakage initiation is seen to be a promising CI because it appears to be linearly scalable to operational conditions such as pressure and speed, with good accuracy, for predicting the leakage threshold.

A Novel Machine Learning-Based Methodology for Tool Wear Prediction Using Acoustic Emission Signals

There is an increasing trend in the industry of knowing in real-time the condition of their assets. In particular, tool wear is a critical aspect, which requires real-time monitoring to reduce costs and scrap in machining processes. Traditionally, for the purpose of predicting tool wear conditions in machining, mathematical models have been developed to extract the information from the signal of sensors attached to the machines. To reduce the complexity of developing physical models, where an in-depth knowledge of the system being modelled is required, the current trend is to use machine-learning (ML) models based on data from the tool wear. The acoustic emission (AE) technique has been widely used to capture data from and understand the real-time condition of industrial assets such as cutting tools. However, AE signal interpretation and processing is rather complex. One of the most common features extracted from AE signals to predict the tool wear is the counts parameter, defined as the number of times that the amplitude of the signal exceeds a predefined threshold. A recurrent problem of this feature is to define the adequate threshold to obtain consistent wear prediction. Additionally, AE signal bandwidth is rather wide, and the selection of the optimum frequencies band for feature extraction has been pointed out as critical and complex by many authors. To overcome these problems, this paper proposes a methodology that applies multi-threshold count feature extraction at multiresolution level using wavelet packet transform, which extracts a redundant and non-optimal feature map from the AE signal. Next, recursive feature elimination is performed to reduce and optimize the vast number of predicting features generated in the previous step, and random forests regression provides the estimated tool wear. The methodology presented was tested using data captured when turning 19NiMoCr6 steel under pre-established cutting conditions. The results obtained were compared with several ML algorithms such as k-nearest neighbors, support vector machines, artificial neural networks and decision trees. Experimental results show that the proposed method can reduce the predicted root mean squared error by 36.53%.