Department of Mechanical engineering

Permanent URI for this collectionhttp://localhost:4000/handle/123456789/1921

Browse

Now showing 1 - 18 of 18

An adaptive source-free unsupervised domain adaptation method for mechanical fault detection
(Elsevier, 2025-04) Choudhury, Madhurjya Dev
Cross-machine fault detection is crucial due to the challenges of data labeling. While domain adaptation methods facilitate diagnosis across rotating machines, they often require data sharing, which is impractical due to privacy concerns and large data transmission. Although domain generalization and source-free unsupervised domain adaptation (SFUDA) methods address privacy issues, most fail to consider dynamic distribution shifts within and between domains, limiting their effectiveness. To overcome this challenge, an adaptive SFUDA method named AI3M is proposed. The AI3M pre-trains a source model with intra- and inter-domain information maximization loss to reduce distribution shifts within and between domains, and then adapts the model with a target-guided adaptation strategy to minimize the dynamic gap between different machines. Experiments on datasets from 11 wind turbines across 8 wind farms show that the proposed method outperforms state-of-the-art DG and SFUDA approaches, achieving superior cross-machine fault detection performance.
Bearing signal classification using dynamic time warping
(Springer, 2025-03) Choudhury, Madhurjya Dev
Bearing is a critical machine element whose fault-free health status is crucial to the reliable operation of the machine. In recent times, machine learning-based approaches of training models on huge amount of measured data are adopted to monitor the bearing health status. However, collecting a large number of complete fault samples in an industrial set-up is very expensive thereby fostering the need to introduce methods that have excellent classification and clustering performance without depending on an extensive amount of training data. To address this, a dynamic time warping (DTW)-based signal similarity measuring approach is proposed. In this paper, a reference signal corresponding to a particular bearing health class is generated, which is then warped by DTW on any test bearing signals to reveal the inherent similarity by calculating their cumulative distances. The distance measure is used to classify the bearing health. It is observed that DTW alone cannot obtain acceptable results when employed to handle complex bearing signals because of the presence of measurement noise and unwanted interfering components. An enhancement is proposed by integrating DTW-based similarity search with spectral kurtosis (SK)-based demodulation for enhanced classification. The proposed method is validated using the Case Western Reserve University (CWRU) bearing dataset.
Bridge Configured Wounded Switched Reluctance Motor
(Elsevier, 2016) Choudhury, Madhurjya Dev
Conventional switched reluctance motors suffer from high undesired acoustic noises and vibrations caused by the production of a considerable amount of radial force due to non-uniformity in air-gap which can be controlled in order to make it more efficient. A feasible solution to pacify this problem is the introduction of a special winding scheme called bridge configured winding (BCW) in switched reluctance motor (SRM). Various winding configurations (generally dual set of windings) have been developed till date in order to produce radial force in SRM. This paper presents the incorporation of bridge configured winding capable of producing both the torque and a controllable radial force using a single set of winding, thus reducing the use of additional winding for radial force production.
A Comparative Analysis Between EMD- and VMD-Based Tacho-Less Order Tracking Techniques for Fault Detection in Gears
(Springer, 2021-04) Choudhury, Madhurjya Dev
Vibration signals collected from gearboxes in an industrial environment are characterized by inherent machine speed fluctuation and complex multicomponents, which make it difficult to extract reliable fault information. The presence of speed fluctuation smears the vibration spectrum of the gearbox. Under such conditions, order tracking (OT) technique is adopted, which performs signal resampling using additional speed information of the operating shaft. Thus, OT is dependent on the availability of shaft speed information, which is generally measured using a tachometer. This study proposes a tacho-less method to accurately estimate the shaft speed information in the presence of speed fluctuations by the use of adaptive mode decomposition (AMD) methods, like empirical mode decomposition (EMD) and variational mode decomposition (VMD). The extracted shaft speed information is then used to perform OT for gear fault detection in the presence of speed fluctuation. A comparison has also been conducted to evaluate the effectiveness of extracting shaft speed information by using both VMD and EMD. The analysis results show that speed information can be better extracted using VMD as compared to EMD. The effectiveness of the proposed algorithm is demonstrated using a MATLAB-based simulation analysis.
A Critical Investigation of Hilbert-Huang Transform Based Envelope Analysis for Fault Diagnosis of Gears
(IEEE, 2018) Choudhury, Madhurjya Dev
Hilbert-Huang Transform (HHT) has been extensively used for fault diagnosis due to its capability in handling amplitude-frequency modulated (AM-FM) and multicomponent signals. However, its performance in handling the signals having complexity such as arising from practical gearbox measurements, like speed and load variation is debatable. This study is conducted to understand the use of traditional HHT in gear fault diagnosis by carrying out a literature survey followed by an envelope spectrum (ES) analysis for gear fault detection. The investigation demonstrates the capability of HHT in decomposing a multicomponent fault signal into its different meaningful modes, which can then be exploited for diagnosis. HHT demodulated envelope signal spectrum is found to be effective in revealing fault induced peaks during constant speed operation but its performance deteriorates under speed variation. Various gear fault simulation models are investigated to validate the effectiveness of HHT and a discussion is provided to conclude the paper.
A Deep Learning Based Fault Diagnosis Method Combining Domain Knowledge and Transfer Learning
(IEEE, 2023-11) Choudhury, Madhurjya Dev
Deep learning (DL) based fault diagnosis methods have gained considerable attention in the field of machine health monitoring due to their powerful feature learning capabilities. However, embedding domain diagnosis knowledge into the DL framework to obtain enhanced features having better correlation with the exact health conditions of machine elements for improved fault predictions is still an open challenge. In this paper, a fault diagnosis method combining two-dimensional (2D) image representations of squared envelop spectrum (SES) of vibration signals of bearings, a critical machine element, and a pretrained convolutional neural network (CNN) is proposed. SES is one of the most efficient indicator for the assessment of second order cyclostationary symptoms of damages, which are typically observed in bearings. In this paper, we integrate this knowledge in designing a DL framework for efficient fault diagnosis in bearings. The proposed method is tested and evaluated on an experimental bearing vibration dataset collected under different operating and fault conditions. Experimental results demonstrate that the proposed method can achieve a high diagnosis accuracy and present a better generalization ability both in balanced and imbalanced data scenarios
Deep Learning based Time-Frequency Image Enhancement Method for Machinery Health Monitoring
(IEEE, 2023) Choudhury, Madhurjya Dev
Reliable machinery health monitoring using measured vibration signals requires a good readability of time-frequency (TF) images. However, conventional TF methods suffer from a limited time–frequency resolution and cross-term interferences, which limit their practical applicability in health monitoring. To address this issue, a TF image improvement method using deep learning is proposed in this paper. The proposed method employs a deep learning technique known as conditional generative adversarial network (cGAN) to convert a noisy low-resolution TF image of a bearing vibration signal into a noise-free high-resolution image such that the true frequency characteristics of measured signals may be revealed. In this paper, the cGAN model is trained using a simulation-based dataset generated from a bearing analytical model. The trained cGAN model is then utilized to improve TF images generated from real bearings under different fault and operating conditions. The results reveal that the proposed image improvement method generates high-resolution TF representations which are better than both the traditional TF images and those generated using TF reassignment methods
Design and Analysis of a Radial Active Magnetic Bearing for Vibration Control
(Elsevier, 2016-05) Choudhury, Madhurjya Dev
Vibration caused by rotor unbalance is one of the most pertinent problems facing the rotating machines, including electrical motors and turbo machinery among others. Thus vibration attenuation has become very essential in improving the overall performance of such machines. In this paper, a 12-pole radial Active Magnetic Bearing (AMB), using AC excitation has been proposed to counteract the unbalance. Here a switching variation of AMB teeth excitation currents is implemented to generate a rotating force, synchronous with the rotor unbalance but in opposite direction.
Design methodology for a special single winding based bearingless switched reluctance motor
(IET, 2017-07) Choudhury, Madhurjya Dev
Bearingless switched reluctance motors (BSRMs) have both magnetic bearing as well as conventional motor characteristics which make them suitable for diverse industrial applications. This study proposes a design methodology for a BSRM in order to calculate the appropriate geometrical dimensions essential for realising a minimum levitation force at every orientation of rotor. It is based on the stator–rotor overlap angle and helps in reducing the complexities associated with the self-bearing operation of a switched reluctance motor (SRM). Different from a conventional SRM, the motor under study deploys a special single set parallel winding scheme for simultaneous production of torque as well as radial force. An analytical model incorporating this single set winding is developed for calculating the torque and the radial force. The proposed bearingless design is verified by developing a two-dimensional finite-element model of a 12/8 SRM in ANSYS Maxwell.
A Digital Twin Based Framework for Real-Time Machine Condition Monitoring
(IEEE, 2023) Choudhury, Madhurjya Dev
Condition Monitoring (CM) is an important approach to extending the life of complex equipment by forecasting the outcome of an event before catastrophic failure occurs. Recent advancements in digital twins (DT) offer additional benefits to machine condition monitoring. In this study, a framework based on DT for real-time condition monitoring of industrial machines is proposed. The multi-layer DT framework consists of a physical entity (PE), virtual equipment (VE), edge device, fidelity service and digital twin services. The virtual equipment is a replica of the physical entity or the monitored machine. It also contains a cloud platform to store data online and an application to interface with the cloud enabling users to check the data remotely. The fidelity service ensures conformity between the PE and the VE. The digital service provides optimal operation and maintenance schedules based on the data from both physical and virtual spaces. The integration of the edge layer enables real-time handling of high-frequency machine data for effective health monitoring. The validity of the proposed framework is demonstrated with a case study based on monitoring a critical component of an industrial drivetrain test rig. The features of the framework allow end-users to visualize the component's real-time health status remotely.
Domain generalization using pseudo triplet network learning for vibration signal-based fault diagnosis
(IEEE, 2025-02) Choudhury, Madhurjya Dev
Domain generalization (DG) based intelligent fault diagnosis has developed rapidly in recent years owing to the need for applying trained neural networks to unseen domains. However, models trained using DG often suffer from performance degradation when in presence of nonstationary working conditions. To address this challenge, this work proposes a DG based intelligent fault diagnosis approach based on a vibration response mechanism guided pseudo triplet network, which extracts suitable features that correlate well with the health conditions. Firstly, the proposed approach estimates the cyclic spectral correlation maps of vibration signals to provide vibration response mechanism of different health conditions. Then, a pseudo triplet neural network is designed to calculate the distance between the representations of the prior input, the negative input from the representation of the main input. The prior input is the specific part of the cyclic spectral correlation map with the selected carrier band and it guides the network focus on the fault-related features. Finally, the proposed approach is evaluated through experiments conducted on data collected from nonstationary working conditions.
Internet-of-Things Enabled Smart Condition Monitoring System for Maintenance of Industrial Equipment, in International Symposium on Flexible Automation
(JSTAGE, 2022-07) Choudhury, Madhurjya Dev
Equipment failures lead to unexpected downtimes that often result in a significant capital and productivity loss. A condition monitoring system (CMS) is critical in alleviating such equipment downtime and to ensure reliability by providing real-time analytics on condition monitoring (CM) data. However, many small and medium enterprises find CM data inaccessible due to the requirement of expensive hard-wired data acquisition system and the difficulty to install the additional hardware and sensors within the existing plant set-up. In this regard, Internet of Things (IoT) brings a new opportunity to CMS, increasing the system's ability to collect and analyze data with more flexibility. This paper presents a proof-of-concept prototype of an IoT-enabled CMS that utilizes edge-devices to remotely collect various measurements from industrial equipment and carry out real-time CM. Moreover, the system leverages a cloud, where history of machinery health state is stored. The cloud also enables real-time access of machinery health status either on demand or push notifications. This cloud service can be further expanded to facilitate the integration of advanced data analytic tools for predictive maintenance. The effectiveness of the implemented system is presented by demonstrating remote data acquisition and spectral analysis of CM signals from a rotating machinery test-rig
A Methodology to Handle Spectral Smearing in Gearboxes Using Adaptive Mode Decomposition and Dynamic Time Warping
(IEEE, 2021-02) Choudhury, Madhurjya Dev
Tacho-less order tracking is an effective tool for fault detection in gearboxes operating under speed fluctuations. This technique's performance depends on extracting instantaneous shaft speed information from a complex multicomponent gearbox signal, which is first preprocessed by a bandpass filter. However, spectral overlap resulting from the time-varying frequency components makes it challenging to isolate the shaft speed harmonic mono-component using this approach. In order to overcome such issues, an adaptive tacho-less order-tracking (OT) method combining the variational mode decomposition (VMD) and the fast dynamic time warping (FDTW) is proposed in this article. The proposed method first decomposes the measured gearbox vibration signal using VMD to estimate the instantaneous shaft speed profile to construct the shaft vibration signal. The gearbox vibration signal is then resampled based on the optimal warping path obtained by FDTW, which performs an “elastic” stretching and compression along the time axis of the extracted shaft vibration signal with respect to a sinusoidal reference signal with a constant shaft rotational frequency. Finally, the presence of gear fault is detected by constructing the order spectrum of the resampled vibration signal. The effectiveness of the proposed algorithm is demonstrated using both simulation analysis and experimental validation using measurements from two different wind-turbine gearboxes collected under test and field conditions, respectively.
Modeling and analysis of bearingless switched reluctance motor equipped with specialized stator winding
(IEEE, 2016) Choudhury, Madhurjya Dev
Conventional switched reluctance motor undergoes extensive vibration and undesirable acoustic noise caused by the production of significant amount of radial force. The reason behind this is the non-uniformity in air-gap of the motor which can be controlled in order to make it more efficient. A viable solution to pacify this problem is the introduction of bearingless technology with a special winding scheme called bridge configured winding in switched reluctance motor. Various designs of winding have been introduced by researchers to produce radial force which can be utilized in order to make the motor bearingless. This paper presents the incorporation of a single set winding called bridge configured winding in a 12/8 switched reluctance motor capable of producing both torque and controllable radial force, which can be used for bearingless operation of the motor. An analytical model of a 12/8 switched reluctance motor equipped with bridge configured winding is developed considering the effects of magnetic saturation of the motor. A mathematical model is obtained by solving the magnetic equivalent circuit of the proposed design. Also a finite element model of the design is developed in ANSOFT Maxwell 2D in order to verify the developed analytical model.
A novel tacholess order analysis method for bearings operating under time-varying speed conditions
(Elsevier, 2021-12) Choudhury, Madhurjya Dev
Order tracking (OT) of vibration envelope signal based on instantaneous frequency (IF) estimated from a time-frequency representation (TFR) is generally applied to detect bearing defects under time-varying speed conditions. However, these TFR based OT approaches require a high resolution for accurate estimation of the IF. To overcome these issues, a tacholess OT technique based on the fast dynamic time warping (FDTW) algorithm is applied to detect bearing defects for the first time in this paper. FDTW based signal transformation is initially applied to align a filtered shaft signal to a reference signal of constant frequency, followed by reconstructing the bearing envelope signal. The resulting envelope signal is able to provide a richer fault indication than the original signal. The efficacy of the developed OT method is established using a simulation analysis followed by experimental validation using measurements obtained from a laboratory test-rig and a wind turbine bearing.
Order Tracking Using Variational Mode Decomposition to Detect Gear Faults Under Speed Fluctuations, in 11th Annual Conference of the PHM Society 2019
(Prognostics and Health Management Society, 2019-09) Choudhury, Madhurjya Dev
Fault detection in gearboxes plays a significant role in ensuring their reliability. Vibration signals collected during gearbox operation contain a wealth of valuable condition information that can be exploited for fault detection. However, in an industrial environment machine operating speed always fluctuates around its nominal value, which causes smearing of the gearbox vibration spectrum. Considering operating speed fluctuation and multi-component nature of measured gearbox vibration signals, an order-tracking method combining the variational mode decomposition (VMD) and the fast dynamic time warping (FDTW) is proposed in this paper. Firstly, the multi-component vibration signal is decomposed into several intrinsic mode functions (IMFs) using VMD in order to extract a signal component with higher signal-to-noise ratio (SNR). Then, the sensitive fault information carrying IMF is exploited to estimate the instantaneous speed profile in order to construct the shaft rotational vibration signal. The measured vibration signal is then resampled based on the optimal warping path obtained by FDTW, which performs an “elastic” stretching and compression along the time axis of the extracted shaft vibration signal with respect to a sinusoidal reference signal of constant shaft rotational frequency. Finally, the gear fault is detected by constructing the order spectrum of the resampled vibration signal. The effectiveness of the proposed algorithm is demonstrated using simulation results.
An Overview of Fault Diagnosis of Industrial Machines Operating Under Variable Speeds
(Springer, 2021-04) Choudhury, Madhurjya Dev
This paper provides an overview of the recent advances made in the field of fault diagnosis of industrial machines operating under variable speed conditions. First, the shortcomings of the traditional techniques in extracting reliable fault information are laid down, followed by a discussion on the different approaches adopted to overcome these issues. Next, these approaches are discussed by categorizing them as resampling based and resampling free methods. The principle and implementation procedures of these methods are discussed by summarizing the key literature in this area. Finally, the paper is concluded by highlighting the future challenges to address in this area.
Vibration-based Condition Monitoring of Industrial Drivetrains Operating under Non-stationary Conditions
(The Prognostics and Health Management Society, 2019-09-22) Choudhury, Madhurjya Dev
This research is focused on advancing the state-of-the art in fault diagnostics of industrial drivetrains. It is proposed to develop a fault detection algorithm and fault severity prediction model for critical drivetrain components using measured vibration signals. The fault detection algorithm will be developed to overcome the challenge of extracting reliable fault information for drivetrains operating under speed and load fluctuations, whereas the severity prediction model will aid in predicting the degradation level of a faulty component. The results of this research can, therefore, help in improving the availability of industrial drivetrains, by providing a dependable platform to the maintenance personnel for proper decision-making.