Vibroacoustic Signals Research Articles

Application of machine learning methods for the classification of vibroacoustic signals measured at the moment of transformer energization with varying degrees of core and winding defect complexity

Application of machine learning methods for the classification of vibroacoustic signals measured at the moment of transformer energization with varying degrees of core and winding defect complexity

Maximum Fourier spectrum cyclostationarity blind deconvolution and its application in structural health monitoring of power transformers

Abstract Power transformer is the most important equipment that affects whether the electric power system can be operated safely and normally, whose condition assessment problem has attracted considerable attention. Background noise frequently affects the effectiveness of nonintrusive techniques based on vibro-acoustic signals for structural health monitoring in power transformers. It is challenging to effectively eliminate background noise interference from time-frequency domain transformer defect detection techniques such as wavelet transformations, modal decomposition, etc. In this scenario, the Fourier spectrum cyclostationarity index (FSC) is designed based on the cyclostationarity index used for rotating machinery fault diagnosis to construct a maximum Fourier spectrum cyclostationarity blind deconvolution method (MFSCBD) for transformer fault detection in this paper. Firstly, the limitations of the traditional blind deconvolution (BD) in transformer fault detection are discussed in the mathematical principle. Then, a new BD framework based on Kepler optimization algorithm is proposed according to the principle of convex optimization to address the problems of difficulty in solving the complex blind objective function in the traditional differential BD framework and the ill-condition problem in the Rayleigh quotient BD framework. Subsequently, a synthetic nonstationary and nonlinear simulation signal is constructed for numerical verification, and a six-microphone array is designed to obtain the practical signals from the operating transformer to verify the performance of MFSCBD. Finally, the applications on the simulated and experimental signals of power transformers demonstrate that MFSCBD outperforms complete ensemble empirical mode decomposition with adaptive noise and successive variational mode decomposition to some extent for structural health monitoring.

Detection of Early Browning in Pears Based on Time–Frequency Images of Vibro-Acoustic Signals and Improved MobileNetV3

Detection of Early Browning in Pears Based on Time–Frequency Images of Vibro-Acoustic Signals and Improved MobileNetV3

Жизненный цикл изделия: мониторинг процессов механической обработки и фильтрация виброакустических сигналов

Introduction. In modern manufacturing, the product life cycle comprises various stages, from conception to disposal. Among these stages, machining plays a significant role, as it directly influences the durability and functionality of the finished product. With increasing competition and the need to reduce production costs, optimizing machining processes has become a crucial task. Traditionally, conservative technological approaches have been used to ensure processing quality. However, this often leads to decreased productivity and higher costs. Modern monitoring and diagnostic techniques can significantly improve process control, particularly through tool condition monitoring. The subject. This paper discusses the stages of the product life cycle and emphasizes the importance of monitoring machining processes. It explores the potential of using vibroacoustic signals to continuously monitor equipment and product conditions. Special attention is paid to the use of vibroacoustic signals for diagnostics and quality control. Modern approaches to filtering these signals, including the use of the fast Fourier transform and various window functions, are analyzed in order to improve the accuracy of the analysis and identify potential defects. The purpose of this work is to develop an algorithm for an online monitoring system that will monitor the condition of cutting tools based on the creation of a digital shadow using a vibroacoustic complex. The main tasks to be solved are to establish the ranges of applicability of frequency response of acoustic signals and optimal window functions, as well as to establish the relationship between the degree of wear on the cutting tool and the results of vibration diagnostics and surface roughness. The methods and technologies for filtering vibroacoustic signals and their application in real–world production settings are discussed. Special attention is given to the role of digital twins in integrating monitoring and filtering data, allowing for the creation of a virtual model of a product to predict its behavior and optimize processes throughout the life cycle. A comparison of various monitoring methods and technologies is conducted, as well as an analysis of practical examples of digital twin implementation in production processes and its impact on improved control. Results and discussion are presented, identifying current research and practical advancements, while also proposing existing challenges and promising areas for future research in the fields of monitoring, signal filtering, and the use of digital twins in mechanical manufacturing.

Информационные свойства частотных характеристик динамической системы резания при диагностике износа инструментов

Introduction. One of the directions for increasing the efficiency of cutting is related to the creation of tool wear diagnostic systems. Algorithms and devices have been developed that evaluate wear based on analysis of the vibroacoustic emission signal. These algorithms, as a rule, do not reveal the nature of its formation and the reasons for the change as wear develops. Subject. The paper is devoted to the analysis of the reasons for changes in vibration properties with the development of tool wear. The aim of the work is to study the changes in the frequency characteristics of a dynamic cutting system caused by the development of wear, and to build diagnostic information models on this basis, as well as its use in industry. Method and methodology. The results of mathematical simulation of a perturbed dynamic cutting system are presented, in which the observed vibration sequences are a consequence of disturbances transformed by a dynamic system, the parameters of which depend on wear. Two frequency ranges are considered. Results and discussions. The first range includes frequencies that lie within the bandwidth of the instrument subsystem. The second is outside of it. In the first frequency range, it has been analytically and experimentally proven that the development of wear leads to fundamental changes in the frequency properties of the cutting system as a converter of disturbances into tool vibrations. There is a shift in the natural frequencies of the oscillatory circuits formed by the cutting system, a decrease in its quality factor, and as wear develops, some identified features of the vibration spectra appear, including the ratio of the low-frequency and high-frequency parts of the spectrum, etc. In the second frequency range, a model of force emission in the form of a random pulse sequence is considered and wear is displayed in it. The results of studying the coherence function between the forces acting on the tool and vibrational displacement are presented. Information models of wear are proposed, an example of an information model of wear and the results of its use in industry are given.

Variational Autoencoder feature clustering for tissue classification in robotic palpation

Abstract Minimally Invasive Robotic Surgery (MIRS) has emerged as a transformative approach in surgical practice, offering reduced patient trauma and enhanced precision. However, challenges persist, including the loss of tactile feedback for surgeons. This study explores the application of machine learning algorithms, specifically variational autoencoders, in vibro-acoustic (VA) signal analysis to address this issue. Our comparative analysis evaluates the potential of supervised learning in surgical data analysis, contributing to advancements in surgical technology. Despite achieving an accuracy of 81%, our results indicate opportunities for further refinement, considering the superior accuracies reported in previous studies. This research underscores the importance of innovative approaches in medical data analysis for optimizing patient care in minimally invasive surgery.

Instrument Interactions as Source of Information in Robot-Assisted Surgery

Abstract The interaction between the surgical instrument and tissue is a yet little-exploited source of information. Vibroacoustic signals resulting from an interaction can be analyzed to extract relevant information related to the tissue. This work presents a summary of feasibility studies with synthetic-, animal-, and ex-vivo tissue specimen. A standard palpation probe is equipped with a vibration measurement system. A manual and robot-assisted setup is used for data acquisition. Simple features derived from the signals are used in a subsequent classification step using kNN and SVM. The obtained results show the capability to differentiate various tissues, such as subchondral bone with an accuracy of 98.44%. This demonstrates the potential of the approach to provide tissue related information based on instrument interactions.

Compressor Surge Identification in Innovative Heat-Pump Systems Equipped by Vaned Diffuser Centrifugal Compressor

Abstract In the present energy scenario, heat-pumps play an important role to improve energy efficiency in reversible cooling systems. In case of industrial size plants, centrifugal compressors are preferred with respect to other solutions; nonetheless, they work in variable operation and therefore they must withstand off-design conditions. Carrier provided the University of Genoa with a small size chiller rig equipped with an innovative high speed centrifugal compressor driven by a variable speed motor to study unstable operation of refrigerant closed loop systems. To this aim, vibro-acoustic signature analysis of such closed loop rig is performed mainly focusing on compressor mechanical response. This experimental characterization is firstly conducted in system stable operation. Afterwards, surge transients are obtained by acting on plant feeding lines valves; meanwhile, experimental signals are acquired at relevant plant locations to investigate system response just before instability onset. Indeed, system dynamics is affected by interposed volumes, and therefore the effect of heat exchangers in system response may be relevant. Therefore, system dynamics is analysed both in sub-synchronous frequency range and in high frequency region to assess how its vibro-acoustic response varies when moving from stable conditions towards surge. By doing so, suitable surge precursors can be defined by relying only on vibro-acoustic signals both in low and high frequency ranges to perform early surge detection in such chiller systems. The proposed approach main advantage is to exploit non-intrusive probes, which allow to define diagnostic indicators without interacting directly with the working fluid.

A smart look at monitoring while drilling (MWD) and optimizing using acoustic emission technique (AET)

Monitoring while drilling (MWD) is a crucial task in mining operations. Accurately measuring drill and rock-related operating parameters can significantly reduce the cost of drilling operations. This study explores the potential of monitoring drilling specific energy (SE) and optimizing drilling operations by processing vibroacoustic signals generated while drilling. For this purpose, 30 samples of different rocks, are used for drilling tests. During the drilling process, the acoustic and vibration signals are recorded and analyzed in the time, frequency, and time–frequency domains., and parameters related to the resulting spectra are extracted. After obtaining the vibroacoustic parameters for drilling, the relationship between them and the drilling SE was investigated. There is evidence that the progression of SE contributes to the magnitude of rock drilling vibroacoustic features, which could be employed to indicate energy conditions during drilling. Results obtained in this study have the potential to be used as the basis for an industrial monitoring system that can detect excessive energy consumption and advise the user of the end of the bit's useful life. This method can be an intelligent technique for measuring the behavior of real-time drilling operations based on the SE simply by installing vibroacoustic sensors on the drilling machines.

Enhancing Veress Needle Entry with Proximal Vibroacoustic Sensing for Automatic Identification of Peritoneum Puncture.

Laparoscopic access, a critical yet challenging step in surgical procedures, often leads to complications. Existing systems, such as improved Veress needles and optical trocars, offer limited safety benefits but come with elevated costs. In this study, a prototype of a novel technology for guiding needle interventions based on vibroacoustic signals is evaluated in porcine cadavers. The prototype consistently detected successful abdominal cavity entry in 100% of cases during 193 insertions across eight porcine cadavers. The high signal quality allowed for the precise identification of all Veress needle insertion phases, including peritoneum puncture. The findings suggest that this vibroacoustic-based guidance technology could enhance surgeons' situational awareness and provide valuable support during laparoscopic access. Unlike existing solutions, this technology does not require sensing elements in the instrument's tip and remains compatible with medical instruments from various manufacturers.

Transfer learning with cross-domain adaptation of vibro-acoustic signals for electric motor mechanical fault detection and diagnosis.

Contact-based vibration measurement techniques involving accelerometers are typically utilized for machine health monitoring and manufacturing end-of-line quality control. Acoustic signals, encompassing sound pressure and particle velocity, carry crucial information about the operational status of underlying equipment and can be utilized for detecting various machine faults. Utilizing one-dimensional Convolution Neural Networks (1D-CNN) based on deep learning for processing raw acoustic time signals has proven to be a valuable approach in detecting machinery faults, offering an effective alternative to using traditional accelerometer signals. However, the success of deep learning methods is heavily dependent on the quantity and availability of data for robust network training. When transitioning to new non-contact type sensor technology, manufacturers may face challenges due to insufficient data for training deep learning models. In such cases, leveraging existing historical accelerometer-based data becomes crucial. Transfer learning emerges as a promising solution, allowing deep learning models trained on a source domain (SD) to be applied to a separate but related target domain (TD). This paper hence introduces a domain adaptation methodology by implementing transfer learning, where the SD 1D-CNN network is trained on accelerometer signals, then the trained model weights are transferred to the TD 1D-CNN network that can process acoustic signals.

CDTFAFN: A novel coarse-to-fine dual-scale time-frequency attention fusion network for machinery vibro-acoustic fault diagnosis

When the machinery device operates abnormally, it is not sufficient for fault detection only via extracting fault features from a single sensor due to the latent fault information may be scattered across multiple sensors. Multi-sensory fusion techniques with deep learning framework have attracted increasing attention from researchers due to the exploiting and integration of fault information between multiple sensors. Nevertheless, there are two remaining shortcomings in most existing multi-sensory fusion technologies. (1) Most existing fusion methods merely concentrate on conducting multi-sensory information fusion from time-domain or frequency-domain to achieve fault diagnosis, which are often unsatisfactory in the face of strong noise environments. (2) The collaborative fusion between several vibration sensors is generally considered in the past works, whereas the complementary information fusion between multi-sensory vibro-acoustic heterogeneous data are rarely studied. To address these deficiencies, this paper proposes a novel coarse-to-fine dual-scale time-frequency attention fusion network (CDTFAFN) for machinery fault diagnosis, which not only adequately considers the complementary information fusion of vibro-acoustic signal, but also has robust feature learning capabilities in a noisy scenario. Firstly, the signal-to-image encoding unit (SIEU) containing the improved constant-Q non-stationary Gabor transform (ICQ-NSGT) is introduced to convert the collected raw vibro-acoustic heterogeneous signal into time-frequency representation (TFR) and achieve the coarse-grained feature fusion. Secondly, the time-frequency attention feature fusion unit (TFA-FFU) is designed to concurrently learn the fine-grained features at two scales from the low-level fused features which are meaningful for fault diagnosis. Finally, the coarse-to-fine features are sequentially concatenated and fed into softmax classifier to preferably promote the network learning performance and automatically implement fault classification. The performance of the proposed approach is validated against those state-of-the-art results on two groups of multi-sensory vibro-acoustic data in different experimental platforms. Experiment results show that the proposed method with the diagnosis accuracy of 99 % above outperforms other several representative fusion technologies (i.e., 2MNet, MFF-GBFD, MSCNN-BiLSTM, MFAN-VAF, 1D-CNN-VAF, MI-CNN-TFT and TFFN-VAF) in the raw noise-free addition scenario. Moreover, the average testing accuracy of the proposed method can still reach 97 % above in the noisy scenarios with Gaussian white noises, which shows its competitive superiority and strong robustness against noises in machinery fault diagnosis. According to the five ensemble macro-average performance evaluation metrics (i.e., accuracy, precision, sensitivity, specificity and F1-score) and the receiver operator characteristic (ROC) analysis, our findings also emphasize the superiority of applying our method for machinery fault diagnosis under the colored noises compared with other fusion technologies (i.e., 2MNet, MFF-GBFD, MSCNN-BiLSTM, MFAN-VAF, 1D-CNN-VAF, MI-CNN-TFT and TFFN-VAF) reported in this paper.

Assessment of the effectiveness of mounting the aircraft piston combustion engine on a test stand

The subject of the considerations described in the paper is the problem of efficiency of mounting an aircraft engine on a dyno test bench. The paper presents the empirical research results of vibroacoustic signal application to the evaluation of correctness of positioning of the Rotax 912 piston aircraft engine. The variability of selected parameters and vibroacoustic characteristics for real engine operating conditions was assessed, and functional relations and their discrete representations were developed, reflecting the efficiency of mounting the object on the frame. Thanks to this process, quantitative measures of the diagnostic assessment of the object were obtained, which can be used in periodic or continuous operational control of the object, taking into account the impact of tribological processes taking place in the test object.

Hybrid Vibration Sensor for Equipment Monitoring and Diagnostics.

Vibration diagnostics based on vibroacoustic signal data belong to the most common ways to monitor the technical condition of equipment and technical structures. The paper considers the general issues of vibration-based diagnostics and shows that in general, it is required to monitor both axial and torsional oscillations, as well as the inclination angle, occurring during the operation of various technical objects. To comprehensively monitor these parameters, a hybrid vibration sensor is proposed, simultaneously implementing three operating modes: recording linear displacements of the vibrating object; recording the rotation angle of the object at its torsional oscillations; recording the object angular deviation from the vertical component of the natural local geomagnetic field, i.e., the inclinometer mode. The proposed hybrid sensor design is described, and a theoretical analysis of the sensor's operation in each of the aforementioned operating modes is performed. The authors show that in the inclinometer mode the sensor actually operates as a fluxgate meter. Generalizing the results of the sensor's operation simultaneously in all three operating modes, an equation for the total output data signal has been obtained, which allows for obtaining the required information on the current values of linear displacements and rotation and inclination angles by selectively filtering it with respective three filters tuned to specific frequencies. The experimental studies of the proposed hybrid vibration sensor confirmed its ability to record various vibrational disturbances and changes in the inclination angle of the monitored object.

Multimodel Neural Network for Live Classification of Water Pipe Leaks From Vibro-Acoustic Signals

Multimodel Neural Network for Live Classification of Water Pipe Leaks From Vibro-Acoustic Signals

Micro-crack localization for steel strands using nonlinear vibro-acoustic modulation based on modified signal bispectrum analysis

Under continuous environmental erosion and long-time high tension, fatigue micro-cracks of bridge cables appear frequently to challenge the reliability of cables. The vibro-acoustic modulation analysis is adopted because of its sensitivity to contact-type defects, especially fatigue cracks. In this paper, using multiple sets of low-frequency vibration and high-frequency ultrasonic excitation, the propagation of the ultrasonic wave in the steel strands is studied. Then, the modulated signal bispectrum is proposed to improve the detection reliability of nonlinear modulation components in the nonlinear vibro-acoustic modulation signal of the contact-type defect. The numerical simulations and laboratory tests of straight wire, helical wire, and multi-wire strands are carried out for micro-crack detection to verify the robustness of the proposed method. Compared with the traditional method, the modulated signal bispectrum can still produce clearer peak detection under high noise conditions. The results show that it provides a promising, dynamic and rapid cable defect detection technique.

Application of Statistical Features in Vibro-acoustic Signals to Detect Early Browning Disorder in Pears Compared with Food Chemistry Method

- Browning in pears is one of the most serious diseases in pear fruit, which is caused by Alternaria alternata. The browning process is accompanied by changes in the chemical properties of the fruit, which affect its taste and nutritional composition. Additionally, as a typical postharvest disease, internal browning in pears can cause fruit tissue decay during storage that can reduce the shelf stability of fruit, and bring serious losses to sellers. Because it is difficult to identify the browning pears by appearance, a non-destructive detection technology is highly desirable to correctly discriminate a pear at the early stage of browning for increasing the market value. Firstly, 11 and 7 statistical features were calculated from the time-domain and frequency-domain, respectively. Then, sensitive features in time-domain set, frequency-domain set and combined feature set were selected by the distance evaluation technology, respectively. These selected features were used to train classifier based on K-nearest neighbor algorithm under different K-values. With the selected combined features adopted, the constructed KNN classifier performed the best classification performance. It allowed a high overall accuracy of 91.8 % to classify the healthy and browning pears. Also, the F1 value of 92.6 % indicated that the classifier can be successfully generalized. Therefore, the classification model established in this study is effective for identifying the early browning disorder in pear fruit.

Проблемы диагностики современных тепловозных двигателей

Objective: to consider the issue of diagnostics of modern diesel engines, in particular vibration diagnostics. Consider modern strategies and methods based on vibroacoustic signals that allow you to track and diagnose diesel engine malfunctions, as well as an assessment of the working process in the engine. To determine which means of measuring vibration parameters and methods of processing vibration signals can be considered the most reliable and informative. Propose to make adjustments to the currently used methods of processing vibration signals. Methods: comparison of the effectiveness of vibration measurement tools and mathematical methods of vibration signal processing. Results: the necessity of choosing mathematical methods of vibration signal processing is indicated, both for individual components of a diesel locomotive engine and for evaluating the workflow. Modern mathematical methods, applied to vibration diagnostics, require updating, due to an increase in computing power, as well as due to the development of powerful signal processing methods. To increase the reliability of the comparison results, it is necessary to take into account the large variability of the components and processes of the diesel engine. The need for additional study of modern mathematical methods of vibration signal processing is revealed. Practical importance: the necessity of introducing more modern methods of vibration signal processing is shown, which will optimize the durability of the components’ design using long operating cycles, reduce maintenance costs, track the service life of the internal combustion engine during the operation of the locomotive, improve the monitoring and diagnostics systems of the locomotive engine. The presented methods of assessing the use of certain methods of vibration diagnostics for diesel locomotives can be recommended for practical use.

Using Deep Learning to Detect Anomalies in On-Load Tap Changer Based on Vibro-Acoustic Signal Features

An On-Load Tap Changer (OLTC) that regulates transformer voltage is one of the most important and strategic components of a transformer. Detecting faults in this component at early stages is, therefore, crucial to prevent transformer outages. In recent years, Hydro Quebec initiated a project to monitor the OLTC’s condition in power transformers using vibro-acoustic signals. A data acquisition system has been installed on real OLTCs, which has been continuously measuring their generated vibration signal envelopes over the past few years. In this work, the multivariate deep autoencoder, a reconstruction-based method for unsupervised anomaly detection, is employed to analyze the vibration signal envelopes generated by the OLTC and detect abnormal behaviors. The model is trained using a dataset obtained from the normal operating conditions of the transformer to learn patterns. Subsequently, kernel density estimation (KDE), a nonparametric method, is used to fit the reconstruction errors (regarding normal data) obtained from the trained model and to calculate the anomaly scores, along with the static threshold. Finally, anomalies are detected using a deep autoencoder, KDE, and a dynamic threshold. It should be noted that the input variables responsible for anomalies are also identified based on the value of the reconstruction error and standard deviation. The proposed method is applied to six different real datasets to detect anomalies using two distinct approaches: individually on each dataset and by comparing all six datasets. The results indicate that the proposed method can detect anomalies at an early stage. Also, three alarms, including ignorable anomalies, long-term changes, and significant alterations, were introduced to quantify the OLTC’s condition.

A two-level fusion model of vibro-acoustic signals for centrifugal fan blade crack detection

Blade crack detection is the key to ensuring the smooth and safe operation of centrifugal fans. However, a single vibration signal is difficult to fully reflect the health state of the blade and is susceptible to noise interference in the industrial field, which makes it difficult to detect blade cracks. Therefore, a two-level fusion model of vibro-acoustic signals is proposed for blade crack detection of centrifugal fans. Firstly, based on the designed correlated degree of cyclostationarity fusion rule, the data-level fusion of multi-source homogeneous vibro-acoustic signals is completed, and the fused signals with more obvious fault features are obtained. Then, a one-dimensional feature pyramid network is proposed to extract the vibro-acoustic features and generate initial decisions. Finally, a basic probability assignment acquisition method based on the precision of the initial classifiers and a weighted average method based on the Pignistic probability distance are proposed to minimize the conflict between multi-source evidence, and the Dempster-Shafer evidence theory method is used to obtain the blade crack detection results. The effectiveness of the proposed method is verified by the centrifugal fan blade crack detection experiments. Compared with other related methods, the proposed method has better detection performance and stronger robustness.