Article
Engineering, Mechanical
Yanyan Nie, Fangyi Li, Liming Wang, Jianfeng Li, Mengyao Wang, Mingshuai Sun, Guoyan Li, Yanle Li
Summary: Planetary gearboxes are prone to gear local faults due to their difficult working environment, which can lead to significant losses and catastrophic events. This study proposes an improved phenomenological vibration model and calculates the relative phases between gear pairs to establish vibration models with/without local faults in order to diagnose these faults. Spectral structures and LFCFs are derived for diagnosis purposes, and simulation and experimental studies demonstrate the effectiveness of the proposed models.
MECHANISM AND MACHINE THEORY
(2022)
Article
Engineering, Electrical & Electronic
Anurag Choudhary, Deepam Goyal, Shimi Sudha Letha
Summary: The study proposed a method based on 2D-DWT infrared thermography for diagnosing bearing faults in induction motors. By using PCA and MD methods to extract and rank the most relevant features, SVM was employed for fault classification. The results showed that this method performed well in identifying bearing faults.
IEEE SENSORS JOURNAL
(2021)
Article
Computer Science, Artificial Intelligence
Xianhua Chen, Zhigang Tian, Meng Rao
Summary: A hybrid model with multiple sub-classifiers is proposed for intelligent fault diagnosis of gearboxes. Each sub-classifier is optimized to deal with different fault types, and the selected features are then combined to form the optimal features for all types of gearbox failures in the hybrid model. The proposed method, known as block feature selection (BFS), utilizes NSGA-II and a new sorting algorithm to optimize the feature selection process for each sub-classifier. The effectiveness of BFS is demonstrated through experiments on bearing and gear faults in a planetary gearbox rig, and its robustness is verified by incorporating white noise at different levels.
ADVANCED ENGINEERING INFORMATICS
(2023)
Article
Engineering, Electrical & Electronic
Gianluca Salata, Linga Reddy Cenkeramaddi, Van Khang Huynh, Kjell Gunnar Robbersmyr, Ajit Jha
Summary: In this article, the use of millimeter-wave (mmW) radar for noncontact, nondestructive detection, characterization, and prediction of motor failure probability is proposed and experimentally demonstrated. The features extracted from time-frequency representations distinguish faulty bearings from healthy bearings. The probability of motor failure is quantified using a probabilistic approach and model parameters.
IEEE SENSORS JOURNAL
(2023)
Article
Chemistry, Analytical
Juan Jose Saucedo-Dorantes, Francisco Arellano-Espitia, Miguel Delgado-Prieto, Roque Alfredo Osornio-Rios
Summary: The scientific and technological advances in the field of rotating electrical machinery are leading to increased efficiency in processes and systems involving them. The proposed methodology based on deep feature learning is effective in diagnosing and identifying bearing faults for different bearing technologies, such as metallic, hybrid, and ceramic bearings, in electromechanical systems. The methodology consists of three main stages: design of a deep learning-based model for feature extraction, feature fusion for increased discrimination capabilities, and final assessment using a softmax layer for classification results.
Article
Engineering, Multidisciplinary
Shubhasish Sarkar, Prithwiraj Purkait, Santanu Das
Summary: A fault diagnosis technique based on NI-CompactRIO for stator winding has been proposed, which can effectively discriminate and detect partial degradation and complete breakdown of the insulation in 3-phase induction motors. Principal Component Analysis method is utilized to extract significant features, and the variances of the first two principal components play a key role in fault detection and severity assessment.
Article
Engineering, Mechanical
Fabian Perez-Sanjines, Cedric Peeters, Timothy Verstraeten, Jerome Antoni, Ann Nowe, Jan Helsen
Summary: This study proposes an automated fault detection method using deep learning to analyze vibration signals and detect persistent changes in modulation characteristics by examining cyclic coherence maps. By combining cyclostationary signal processing and deep learning, the proposed methodology can effectively detect and track mechanical faults in non-stationary vibration data of rotating machinery.
MECHANICAL SYSTEMS AND SIGNAL PROCESSING
(2023)
Article
Engineering, Electrical & Electronic
Madhurjya D. Choudhury, Liu Hong, Jaspreet S. Dhupia
Summary: An adaptive tacho-less order-tracking method that combines variational mode decomposition and fast dynamic time warping has been proposed for fault detection in gearboxes under speed fluctuations. By decomposing the measured gearbox vibration signal and resampling it based on an optimal warping path, the method successfully isolates shaft speed information and detects gear faults effectively. The proposed algorithm's effectiveness has been demonstrated through simulation analysis and experimental validation using measurements from different wind-turbine gearboxes.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2021)
Article
Green & Sustainable Science & Technology
Boualem Merainani, Sofiane Laddada, Eric Bechhoefer, Mohamed Abdessamed Ait Chikh, Djamel Benazzouz
Summary: The wind power industry faces high failure rates in wind turbine high-speed shaft bearings, and this paper proposes a practical and effective data-driven methodology for predicting RUL of HSSBs. The methodology involves constructing a new health indicator and using an Elman neural network for estimation, as well as computing prediction intervals to quantify errors associated with the predictions.
Article
Chemistry, Multidisciplinary
Pinyang Zhang, Changzheng Chen
Summary: This paper proposes an analysis and diagnosis two-stage framework based on time-frequency information analysis. It uses a U-net model for the semantic segmentation of vibration time-frequency spectrum and shape features to extract useful information for health state classification of planetary gearboxes.
APPLIED SCIENCES-BASEL
(2023)
Article
Chemistry, Multidisciplinary
Xiaolu Wang, Aohan Li, Guangjie Han
Summary: This paper proposes an industrial motor bearing fault diagnosis algorithm based on multi-local model decision conflict resolution (MLMF-CR) to fully integrate multi-source heterogeneous information and reasonably resolve multi-source information conflicts. Experiments evaluate the effectiveness of the proposed method, which can effectively resolve the high degree of conflict in the decision-making fusion process.
APPLIED SCIENCES-BASEL
(2023)
Article
Chemistry, Multidisciplinary
Jin Guo, Yefeng Liu, Kangju Li, Qiang Liu
Summary: This paper proposes an early fault detection method based on Improved Deep Principal Component Analysis (ID-PCA), which establishes a model using the time-series characteristic information of the vibration signal. By utilizing the deep decomposition theorem, a multi-layer data processing model is created to fully extract weak fault features in the vibration signal.
APPLIED SCIENCES-BASEL
(2022)
Article
Automation & Control Systems
Mohsen Zafarani, Bashir H. Jafari, Bilal Akin
Summary: In this article, the lateral and torsional vibrations in multistack rotor induction motors are modeled in the same framework to analyze the precursors in the current waveform. A lumped mass-spring model is used to calculate natural frequencies and torsional-mode shapes of the mechanical system, while winding function theory is employed to mimic the effects of lateral vibrations on motor inductances. Experimental results are presented and compared with analytical results to validate the modeling approach.
IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS
(2021)
Article
Automation & Control Systems
Qinkai Han, Ziyuan Jiang, Yun Kong, Tianyang Wang, Fulei Chu
Summary: This study proposes a motor current model for diagnosing localized defects in planet rolling bearings (PRBs). A dynamic analysis of the motor-driven planetary gearbox system is conducted and a translational-torsional vibration model is developed. The model takes into account the time-varying impact caused by defects, as well as the influence of Hertzian contact and radial clearance on the support forces of PRBs. The model is validated through electromagnetic finite element calculations and dynamic response tests on a typical motor-driven planetary gearbox system. The results provide a theoretical basis for the quantitative diagnosis of PRB-localized defects.
IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS
(2023)
Article
Engineering, Multidisciplinary
Jianqun Zhang, Baoming Xu, Zhenya Wang, Jun Zhang
Summary: The paper introduces a novel compound fault diagnosis method named FSK-MBCNN for wind turbine gearboxes, which can diagnose faults with over 97% accuracy. The robustness of the method is validated, showing the potential to accurately identify different statuses of wind turbine gearboxes and reduce maintenance costs and downtime.
Article
Energy & Fuels
Artvin-Darien Gonzalez-Abreu, Roque-Alfredo Osornio-Rios, Arturo-Yosimar Jaen-Cuellar, Miguel Delgado-Prieto, Jose-Alfonso Antonino-Daviu, Athanasios Karlis
Summary: This review provides an overview of the reported works concerning the efficiency of motors and drives and the power quality of the electric grid. It discusses the relationship between motors and drives that cause electric disturbances, affecting power quality, and how these disturbances in turn affect the efficiency of motors and drives. It also reviews techniques for detecting, classifying, and mitigating power quality disturbances. The article presents trends and future work in power quality analysis from the perspective of motor and drive efficiency.
Article
Chemistry, Analytical
Alvaro Ivan Alvarado-Hernandez, Israel Zamudio-Ramirez, Arturo Yosimar Jaen-Cuellar, Roque Alfredo Osornio-Rios, Vicente Donderis-Quiles, Jose Alfonso Antonino-Daviu
Summary: The monitoring of machine conditions is crucial for productivity, economic benefits, and maintenance. Infrared thermography, along with smart sensor technology, has become increasingly important in nonintrusive fault diagnosis.
Article
Computer Science, Information Systems
Eduardo Perez-Anaya, David A. Elvira-Ortiz, Roque A. Osornio-Rios, Jose A. Antonino-Daviu
Summary: This paper proposes a method to estimate different levels of dust accumulation in photovoltaic panels using a machine learning approach, which is implemented with real signals to validate its effectiveness in identifying dust accumulation conditions and determining maintenance actions.
Article
Energy & Fuels
Arturo Yosimar Jaen-Cuellar, Miguel Trejo-Hernandez, Roque Alfredo Osornio-Rios, Jose Alfonso Antonino-Daviu
Summary: Kinematic chains integrate various components including induction motors, mechanical couplings, and loads for industrial processes that require motion interchange. The induction motor plays a crucial role in providing power and generating motion, but faults in other elements can affect its operation. To detect gradual wear in the gearbox, a proposed methodology uses statistical features and genetic algorithms along with current and vibration sensors at different frequencies, resulting in effective fault detection.
Article
Chemistry, Analytical
Vicente Biot-Monterde, Angela Navarro-Navarro, Israel Zamudio-Ramirez, Jose A. Antonino-Daviu, Roque A. A. Osornio-Rios
Summary: Induction motors (IMs) are widely used in industrial applications due to their robustness, versatility, and performance. However, rotor bar breakage (BRB) is a common fault that occurs due to high currents during start-up. The use of soft starters is common to reduce stresses, but it complicates fault diagnosis. This paper proposes a method to automatically classify rotor health state in IMs driven by soft starters using a Convolutional Neural Network (CNN) based on Persistence Spectrum (PS) of start-up stray-flux signals.
Article
Energy & Fuels
Joaquin Soldado-Guaman, Victor Herrera-Perez, Mayra Pacheco-Cunduri, Alejandro Paredes-Camacho, Miguel Delgado-Prieto, Jorge Hernandez-Ambato
Summary: This paper compares the Isolated (Flyback) and non-Isolated (Buck) multiple input-single output (MISO) DC-DC converters using pulsed voltage source cells (PVSC). The modeling of both converter types is detailed through mathematical models and electrical simulations. The comparison focuses on sizing parameters, non-ideal output characteristics, and efficiency. Results show that the MISO Buck converter has a linear dependence on the duty cycle control signal and slightly higher efficiency than the Flyback converter.
Article
Physics, Multidisciplinary
David A. Elvira-Ortiz, Juan J. Saucedo-Dorantes, Roque A. Osornio-Rios, Rene de J. Romero-Troncoso
Summary: Gears are prone to wear due to constant contact forces, making it important to develop condition monitoring strategies for proper functioning of power transmission systems. This study proposes the use of entropy features for a high-performance characterization of vibration signals and fusion of different techniques to improve the detection of wear severities in gears compared to conventional approaches.
Article
Chemistry, Analytical
Carlos Gustavo Manriquez-Padilla, Isaias Cueva-Perez, Aurelio Dominguez-Gonzalez, David Alejandro Elvira-Ortiz, Angel Perez-Cruz, Juan Jose Saucedo-Dorantes
Summary: Nowadays, the use of renewable, green/eco-friendly technologies is gaining attention from researchers to ensure the availability of Electric Vehicles (EVs). This work proposes a methodology based on Genetic Algorithms (GA) and multivariate regression to estimate and model the State of Charge (SOC) in Electric Vehicles. The proposed approach achieves a maximum accuracy of approximately 95.5% when validated with real data from a self-assembly Electric Vehicle, making it a reliable diagnostic tool for the automotive industry.
Article
Chemistry, Analytical
Artvin Darien Gonzalez-Abreu, Roque Alfredo Osornio-Rios, David Alejandro Elvira-Ortiz, Arturo Yosimar Jaen-Cuellar, Miguel Delgado-Prieto, Jose Alfonso Antonino-Daviu
Summary: This article introduces a novelty detection method for detecting power disturbances, using six different techniques to detect power disturbances from solar photovoltaic and wind power generation systems. The contribution of this method is the development of a set of techniques that allows for optimal performance under different conditions in the power quality assessment of renewable energy systems.
Article
Engineering, Electrical & Electronic
Leonardo Esteban Moreno-Suarez, Luis Morales-Velazquez, Arturo Yosimar Jaen-Cuellar, Roque Alfredo Osornio-Rios
Summary: This study developed an electro-mechanical model based on the Hardware-in-the-Loop (HIL) structure for a two-wheeler electric scooter, using the BLDC motor to explore its response and to test linear controllers for speed and torque management under variable operating conditions. The proposed model can be used to improve the design of the controller and estimate mechanical and electrical loads.
Article
Engineering, Electrical & Electronic
Geovanni Diaz-Saldana, Roque Alfredo Osornio-Rios, Israel Zamudio-Ramirez, Irving Armando Cruz-Albarran, Miguel Trejo-Hernandez, Jose Alfonso Antonino-Daviu
Summary: This paper proposes a method for tool wear detection using magnetic stray flux and motor current signals from a CNC lathe, as well as analysis of machined parts images. The results show that signal fusion significantly improves detection efficiency, achieving a system that allows online/offline wear detection according to different machining parameters.
Article
Engineering, Electrical & Electronic
Juan Jose Saucedo-Dorantes, Arturo Yosimar Jaen-Cuellar, Angel Perez-Cruz, David Alejandro Elvira-Ortiz
Summary: The importance of induction motors in industrial processes cannot be underestimated. This study proposes a method using empirical wavelet transform and self-organizing map structure for the detection and classification of induction motor abnormalities. A genetic algorithm is utilized to reduce the error in the map. Experimental results demonstrate that this technique can detect and classify faults of different severities regardless of the operating frequency.
Article
Engineering, Electrical & Electronic
Roque Alfredo Osornio-Rios, Israel Zamudio-Ramirez, Arturo Yosimar Jaen-Cuellar, Jose Antonino-Daviu, Larisa Dunai
Summary: Electric motors are essential drivers in various processes, and their importance lies in their high efficiency and robustness. However, operating conditions expose electric motors to different stresses that can cause electromechanical damages and lead to irreversible failures and high repair costs if not detected correctly. Damages in motor components also reduce motor efficiency. This article presents recent advances in electric motor condition monitoring, including the development of a proprietary data fusion system (DFS) for automatic fault diagnosis. The DFS combines the analysis of currents, stray magnetic fluxes, and infrared data, measured noninvasively using simple and low-cost sensors. The article demonstrates the capabilities of the developed methodology through results obtained from applying the system to actual machines.
IEEE INDUSTRIAL ELECTRONICS MAGAZINE
(2023)
