IIT Madras is a public technical and research university located in Chennai, Tamil Nadu. Founded in 1959, it is recognised as an Institute of National Importance.

IIT Madras has been ranked as the top engineering institute in India for four years in a row by the National Institutional Ranking Framework of the MHRD

It currently offers undergraduate, postgraduate and research degrees across 16 disciplines in Engineering, Sciences, Humanities and Management. About 596 faculty belonging to science and engineering departments and centres of the Institute are engaged in teaching, research and industrial consultancy.

IIT Madras

The defect frequency amplitude of the High-Frequency Resonance Technique spectrum is a useful feature for bearing diagnosis. However, it is dependent on the accuracy of the resonance frequency band selection. The vibration data are filtered, with the resonance frequency band being the passband. An alternative approach based on Empirical Mode Decomposition and Instantaneous Energy is proposed to obtain the filtered signal without the need of identifying the resonance frequency band. The objective is to maximize the defect frequency amplitude of the High-Frequency Resonance Technique spectrum. The proposed approach is compared with various optimization techniques and the frequently used Fast Kurtogram approach. These approaches are compared for the simulated and experimental data (seeded and natural defects). The comparison reveals that the proposed approach performs superior to the Fast Kurtogram approach and close to optimization algorithms (the best possible) in a fraction of the time taken by these techniques. © 2021 Elsevier Ltd

World University root node

Department of Mechanical Engineering

Measurement: Journal of the International Measurement Confederation

An approach to improve high-frequency resonance technique for bearing fault diagnosis

Grinding is a finishing operation performed to obtain the desired finish on the component. Wheel wear is one of the primary constraints in achieving the desired productivity in grinding. A new methodology is proposed for accurate and timely identification of wheel wear in cylindrical grinding using Hilbert Huang transform and support vector machine. During the grinding of EN31 carbon steel, the condition of the wheel and its wear was monitored with an accelerometer and power cell. Both vibration and power signals captured were used to identify the condition of the wheel and its wear. An exhaustive feature set is generated in the frequency and the time-frequency domain. Hilbert Huang transform, an adaptive time-frequency analysis technique, was used to extract the features of tool wear in the time-frequency domain. The first three IMF constituents were further chosen for feature extraction of statistical parameters based on their mean energy. Random forests algorithm was used to identify the relevant features. The methodology was validated with several grinding experiments and, is found to give an accuracy of 100% with both low and high cutting depths. The results indicated the robust and reliable wheel wear detection in cylindrical grinding with the use of relatively cheap sensors like accelerometers. The proposed method can be widely used in many applications in the industry where grinding is predominantly used as the finishing operation. © 2021 Elsevier Inc.

Precision Engineering

A robust condition monitoring methodology for grinding wheel wear identification using Hilbert Huang transform

Rolling element bearing is a crucial element of rotating machinery. A sudden failure of the bearing may result in a catastrophic failure. Therefore, the identification of bearing failure in the incipient stage is essential. The vibration signal generated by bearing fault is used for condition monitoring and fault diagnosis. The parameters extracted from vibration data provide the fault indication. However, a single parameter may miss useful information, resulting in less accurate fault diagnosis. The fused parameters after data fusion of individual parameters are more informative and efficient than a single parameter. Mahalanobis–Taguchi–Gram–Schmidt method, principal component analysis, and independent component analysis are the most popular data fusion techniques, and these techniques are applied to three most occurring fault data such as the outer race, the inner race, and rolling elements. The bearing vibration data, available on the NASA Web site, are used for the data analysis and extracting parameters. A comparison is made on the effectiveness of various data fused parameters obtained from different data fusion techniques. The best data fusion technique with respect to each type of defect (inner race, outer race, and ball) is identified. © 2021, Springer Nature Singapore Pte Ltd.

Lecture Notes in Mechanical Engineering

Rolling Element Bearing Fault Diagnosis by Different Data Fusion Techniques

Vibration signal analysis is a widely used condition monitoring technique. Though the extraction of information from the raw vibration signals is difficult because of its non-stationary and non-linear nature, the Empirical mode decomposition has proven to be an effective method. In this method, the raw vibration signal is decomposed into the various intrinsic mode functions. Further, the energy content of these intrinsic modes and the spectral entropy’s associated with each intrinsic mode is analyzed. The method is validated with the available data sets. Based on the results obtained, this method has proven to work better for early fault detection.

Fulltext

Procedia CIRP

Early fault detection based on empirical mode decomposition method

The Year 2019 is historical for the Indian Institute of Technology Madras, India as it celebrates its Diamond Jubilee, and also received National Rank No.1 for the best engineering institution in the country for the fourth consecutive year. The faculty of the Department of Mechanical Engineering, IIT Madras, ranked in the group 51 – 75 in the Shanghai World Ranking, carries out cutting edge research work in the broad areas of Design, Manufacturing and Thermal streams. Founded in 1968, the Machine Design Section, formerly Machine Elements and Mechanical Handling Laboratory, is well known in India and abroad for decades for its contribution in the advancement of Machine Element Design, Design Optimization, Advanced Materials and Design, Health Monitoring, Tribology, and Sound, Machine Dynamic and vibration.The research on design and development of mechanical power transmission elements in the modern era has moved to a new dimension with the availability of modern materials such as customizable composites, light weight metals, lubricating polymers, nano-materials, etc, design tools and methods for efficient product and system design, and integrated design and manufacturing methods such as additive manufacturing, surface modification, etc., The increasing demand of the society for reliable and comfortable modern mobility engineering involving hybrid and electrical drive systems where mechanical power transmission elements play a significant role directs the researchers and designers to look for newer materials, manufacturing routes, design tools and testing for better performance.

IOP Conference Series: Materials Science and Engineering

Preface

Grinding being a finishing process, the quality of the ground surface is one of the most important performance evaluation parameters. Grinding process being highly stochastic in nature, surface finish is affected by many factors and experimental evaluation of each factor is a tedious task. In this study, the in-process signals collected using various sensors attached to a cylindrical grinding machine such as Accelerometer and Power are processed, and their features are correlated with a surface finish parameter. This correlation is modelled using gradient boosting algorithm and surface finish obtained is predicted and validated on an industrial application. © 2020 The Authors.

In-process characterization of surface finish in cylindrical grinding process using vibration and power signals

The Artificial Neural Network (ANN) and curve fitting are traditionally prevalent for prognosis of bearing due to the simplicity of application. The fusion techniques are known to increase the diagnostic capabilities of the damage identification parameters (such as RMS, Peak etc.). However, the role of the fusion techniques in the prognosis of a bearing is still to be explored (attempted in this work). The Mahalanobis-Taguchi-Gram-Schmidt (MTGS) technique is used to fuse various damage identification parameters into a single fused parameter, Mahalanobis Distance (MD). The MD is used as a damage identification parameter for the ANN and the curve fitting. Three methods (i) ANN with damage identification parameters (ii) ANN with MD and (iii) Curve fitting with MD are compared for the effectiveness of the prognosis. The polynomial fit, the Gaussian fit, the sum of sinusoidal fits, and the sigmoid shape fit are attempted. The vibration data for these methods is acquired from a naturally induced and progressed defect through an accelerated life test. © Proceedings of ISMA 2018 - International Conference on Noise and Vibration Engineering and USD 2018 - International Conference on Uncertainty in Structural Dynamics. All rights reserved.

Proceedings of ISMA 2018 - International Conference on Noise and Vibration Engineering and USD 2018 - International Conference on Uncertainty in Structural Dynamics

A comparison of the prognostic approaches for rolling element bearings

A methodology is proposed for classification of type of defect in a bearing using vibration data. Hilbert-Huang Transform (HHT) is used to obtain Instantaneous energy density (IE) values corresponding to interaction of rolling elements and the defect. IE values are treated as a time series and autocorrelation coefficients (ACs) with varying lags are calculated. Based on the ACs, a Defect Occurrence Index (DOI) is proposed to rank the possibility of a type of bearing damage. Based on the DOI, an adaptive filtering process is used to filter the maximum IE corresponding to the peaks generated during the defect interaction. The main feature of the present approach involves capturing the phenomenon of amplitude modulation (using the approach of autocorrelation) for an inner race defect and uses its absence in the case of outer race defect to distinguish between the inner and outer race defects. Statistical techniques (Chi-square test and Coefficient of variation (CV)) on the filtered IE are used for damage type identification. A new parameter, Defect Severity Value (DSV), is proposed for assessment of the severity of the bearing defect. The proposed methodology is validated on simulated outer/inner race defect, and on two different vibration datasets obtained from seeded defect experiments. The proposed methodology helps uniquely identify the bearing damage. © SAGE Publications.

JVC/Journal of Vibration and Control

Bearing damage classification using instantaneous energy density

Employing multiple sensors that generate different physical parameters from the measured system to monitor its health increases the diagnosis reliability. In the present work, bearing diagnosis capabilities of proximity probes are explored by exploiting its advantages and alleviating its shortcomings using appropriate signal processing of the raw time domain data. A Time Synchronous Averaging based method is proposed for processing of the data acquired by proximity probes and its benefit is illustrated on test bearings. Simultaneous synchronous data is acquired with the help of proximity probes and accelerometer during a life test as the defect is naturally induced and progressed with time. The proximity probe is shown to perform better diagnosis for inner race defect compared to accelerometer due to a direct transmission path for this defect. The use of proximity probe can effectively supplement the information from accelerometer and improve the accuracy of bearing diagnosis. © 2015 Elsevier Ltd.

Bearing diagnosis using proximity probe and accelerometer

A methodology is developed for defect type identification in rolling element bearings using the integrated Mahalanobis–Taguchi–Gram–Schmidt (MTGS) method. Vibration data recorded from bearings with seeded defects on outer race, inner race and balls are processed in time, frequency, and time–frequency domains. Eleven damage identification parameters (RMS, Peak, <a href="https://www.sciencedirect.com/topics/engineering/crest-factor">Crest Factor</a>, and Kurtosis in time domain, amplitude of outer race, inner race, and ball defect frequencies in FFT spectrum and HFRT spectrum in frequency domain and peak of HHT spectrum in time–frequency domain) are computed. Using MTGS, these damage identification parameters (DIPs) are fused into a single DIP, <a href="https://www.sciencedirect.com/topics/engineering/mahalanobis-distance">Mahalanobis distance</a> (MD), and gain values for the presence of all DIPs are calculated. The gain value is used to identify the usefulness of DIP and the DIPs with positive gain are again fused into MD by using <a href="https://www.sciencedirect.com/topics/engineering/gram-schmidt-orthogonalization">Gram–Schmidt Orthogonalization</a> process (GSP) in order to calculate Gram–Schmidt Vectors (GSVs). Among the remaining DIPs, sign of GSVs of frequency domain DIPs is checked to classify the probable defect. The approach uses MTGS method for combining the damage parameters and in conjunction with the GSV classifies the defect.A Defect Occurrence Index (DOI) is proposed to rank the probability of existence of a type of bearing damage (ball defect/inner race defect/outer race defect/other anomalies). The methodology is successfully validated on vibration data from a different machine, bearing type and shape/configuration of the defect. The proposed methodology is also applied on the vibration data acquired from the accelerated life test on the bearings, which established the applicability of the method on naturally induced and naturally progressed defect. It is observed that the methodology successfully identifies the correct type of bearing defect. The proposed methodology is also useful in identifying the time of initiation of a defect and has potential for implementation in a real time environment.

Journal of Sound and Vibration

Bearing diagnosis based on Mahalanobis–Taguchi–Gram–Schmidt method

A methodology is proposed for the online detection of health status of rolling element bearing into various damage stages for naturally progressing defect. Various damage identification parameters are derived from processing vibration data in time domain, frequency domain and time–frequency domain. The parameters are fused into a single parameter, <a href="https://www.sciencedirect.com/topics/engineering/mahalanobis-distance">Mahalanobis distance</a>, by application of <a href="https://www.sciencedirect.com/topics/engineering/gram-schmidt-orthogonalization">Gram–Schmidt Orthogonalization</a> process. Chebyshev׳s inequality is applied to the Mahalanobis distance for online monitoring and damage stage detection. A simulation study is first carried out to show working of the proposed methodology in presence of varying trends of damage identification parameters. The proposed methodology is then validated on experimental data. The first validation is on the vibration data acquired from a bearing having seeded defect. Later, two accelerated life tests are conducted on a specially designed <a href="https://www.sciencedirect.com/topics/engineering/test-rig">test rig</a> at different load and speed combinations on the bearings for ensuring naturally induced and progressed defects. The methodology is successfully verified on the vibration data acquired from the naturally induced and progressed defect experiments.

A novel methodology for online detection of bearing health status for naturally progressing defect

An exhaustive study is undertaken to explore various time, frequency and time frequency domain parameters based on vibration data taken from a test rig to identify their effectiveness in rolling element bearing diagnostics. The damage identification parameters extracted after signal processing of vibration data are ranked in order of robustness, sensitivity and early detectivity. In order to better comprehend the defect propagation status in rolling element bearings, a holistic effect of all damage identification parameters need to be considered. Mahalanobis Taguchi System (MTS) is demonstrated as an effective tool to combine these parameters into a single damage identification parameter to track health of rolling element bearing. The Mahalanobis distance is found to be very sensitive and effective parameter that can be used for bearing fault diagnosis. © (2012) by the Katholieke Universiteit Leuven Department of Mechanical Engineering All rights reserved.

International Conference on Noise and Vibration Engineering 2012, ISMA 2012, including USD 2012: International Conference on Uncertainty in Structure Dynamics