Failure type and failure level detection of insulators according to monitored leakage current.

Due to the ever-increasing growth of electric energy consumption, the production of high-quality, reliable and high-reliability electricity is very important. Therefore, it is essential to have distribution and sub-transmission networks with a good reliability factor. In power distribution and sub-transmission lines, it is necessary to somehow isolate the conductors under voltage from the towers, and insulators are used for this purpose. These insulators have two main tasks. One of the main tasks of insulators is to isolate (insulate) the line conductor from the body of the tower. The insulators must be able to isolate the high voltages of the lines from the body of the tower without having a leakage current, and on the other hand, the insulators must be able to withstand the mechanical forces resulting from the weight of the conductors and the applied forces caused by wind and ice. Also, leakage current is one of the important parameters for condition monitoring of insulators in power grid lines. Failure to inspect the insulation for contamination and health conditions will lead to insulator failure and will cause faults in the electrical system. Therefore, it is very important to monitor the condition of the insulator. Based on this, in this paper, according to the data related to leakage current and also according to the introduced wear out function, a procedure for measuring the condition of insulators has been obtained. Finally, the condition of each insulator will be determined according to the defined indicators. Also, the failure level of each monitored data will be obtained using sensitivity analysis.

Exploring the Processing Paradigm of Input Data for End-to-End Deep Learning in Tool Condition Monitoring.

Tool condition monitoring technology is an indispensable part of intelligent manufacturing. Most current research focuses on complex signal processing techniques or advanced deep learning algorithms to improve prediction performance without fully leveraging the end-to-end advantages of deep learning. The challenge lies in transforming multi-sensor raw data into input data suitable for direct model feeding, all while minimizing data scale and preserving sufficient temporal interpretation of tool wear. However, there is no clear reference standard for this so far. In light of this, this paper innovatively explores the processing methods that transform raw data into input data for deep learning models, a process known as an input paradigm. This paper introduces three new input paradigms: the downsampling paradigm, the periodic paradigm, and the subsequence paradigm. Then an improved hybrid model that combines a convolutional neural network (CNN) and bidirectional long short-term memory (BiLSTM) was employed to validate the model's performance. The subsequence paradigm demonstrated considerable superiority in prediction results based on the PHM2010 dataset, as the newly generated time series maintained the integrity of the raw data. Further investigation revealed that, with 120 subsequences and the temporal indicator being the maximum value, the model's mean absolute error (MAE) and root mean square error (RMSE) were the lowest after threefold cross-validation, outperforming several classical and contemporary methods. The methods explored in this paper provide references for designing input data for deep learning models, helping to enhance the end-to-end potential of deep learning models, and promoting the industrial deployment and practical application of tool condition monitoring systems.

Evaluation of Hand-Crafted Feature Extraction for Fault Diagnosis in Rotating Machinery: A Survey.

This article presents a comprehensive collection of formulas and calculations for hand-crafted feature extraction of condition monitoring signals. The documented features include 123 for the time domain and 46 for the frequency domain. Furthermore, a machine learning-based methodology is presented to evaluate the performance of features in fault classification tasks using seven data sets of different rotating machines. The evaluation methodology involves using seven ranking methods to select the best ten hand-crafted features per method for each database, to be subsequently evaluated by three types of classifiers. This process is applied exhaustively by evaluation groups, combining our databases with an external benchmark. A summary table of the performance results of the classifiers is also presented, including the percentage of classification and the number of features required to achieve that value. Through graphic resources, it has been possible to show the prevalence of certain features over others, how they are associated with the database, and the order of importance assigned by the ranking methods. In the same way, finding which features have the highest appearance percentages for each database in all experiments has been possible. The results suggest that hand-crafted feature extraction is an effective technique with low computational cost and high interpretability for fault identification and diagnosis.

A Multidimensional Health Indicator Based on Autoregressive Power Spectral Density for Machine Condition Monitoring.

Condition monitoring (CM) is the basis of prognostics and health management (PHM), which is gaining more and more importance in the industrial world. CM, which refers to the tracking of industrial equipment's state of health during operations, plays, in fact, a significant role in the reliability, safety, and efficiency of industrial operations. This paper proposes a data-driven CM approach based on the autoregressive (AR) modeling of the acquired sensor data and their analysis within frequency subbands. The number and size of the bands are determined with negligible human intervention, analyzing only the time-frequency representation of the signal of interest under normal system operating conditions. In particular, the approach exploits the synchrosqueezing transform to improve the signal energy distribution in the time-frequency plane, defining a multidimensional health indicator built on the basis of the AR power spectral density and the symmetric Itakura-Saito spectral distance. The described health indicator proved capable of detecting changes in the signal spectrum due to the occurrence of faults. After the initial definition of the bands and the calculation of the characteristics of the nominal AR spectrum, the procedure requires no further intervention and can be used for online condition monitoring and fault diagnosis. Since it is based on the comparison of spectra under different operating conditions, its applicability depends neither on the nature of the acquired signal nor on a specific system to be monitored. As an example, the effectiveness of the proposed method was favorably tested using real data available in the Case Western Reserve University (CWRU) Bearing Data Center, a widely known and used benchmark.

Evaluation of the Diagnostic Sensitivity of Digital Vibration Sensors Based on Capacitive MEMS Accelerometers.

In recent years, there has been an increasing use of digital vibration sensors that are based on capacitive MEMS accelerometers for machine vibration monitoring and diagnostics. These sensors simplify the design of monitoring and diagnostic systems, thus reducing implementation costs. However, it is important to understand how effective these digital sensors are in detecting rolling bearing faults. This article describes a method for determining the diagnostic sensitivity of diagnostic parameters provided by commercially available vibration sensors based on MEMS accelerometers. Experimental tests were conducted in laboratory conditions, during which vibrations from 11 healthy and faulty rolling bearings were measured using two commercial vibration sensors based on MEMS accelerometers and a piezoelectric accelerometer as a reference sensor. The results showed that the diagnostic sensitivity of the parameters depends on the upper-frequency band limit of the sensors, and the parameters most sensitive to the typical fatigue faults of rolling bearings are the peak and peak-to-peak amplitudes of vibration acceleration. Despite having a lower upper-frequency range compared to the piezoelectric accelerometer, the commercial vibration sensors were found to be sensitive to rolling bearing faults and can be successfully used in continuous monitoring and diagnostics systems for machines.

Natural Frequency Transmissibility for Detection of Cracks in Horizontal Axis Wind Turbine Blades.

Defects on horizontal axis wind turbine blades are difficult to identify and monitor with conventional forms of non-destructive examination due to the blade's large size and limited accessibility during continuous operation. This article examines both strain and acceleration transmissibility as methods of continuous damage detection on wind turbine blades. A scaled 117 cm offshore wind turbine blade was first designed, 3D printed, and modelled numerically in ANSYS. Transverse cracks were deliberately introduced to the blade at 10 cm intervals along its leading edge. Subsequent changes in the transmissibility, relative to an undamaged baseline model, were measured using different variable combinations at the blade's first three natural frequencies. Experimental results indicated that strain transmissibility was able to locate a 1.0 cm defect at a range of 70-110 cm from the blade hub using the amplitudes of the first natural frequency of vibration. The numerical model was able to simulate the strain experimental results and was determined to be valid for future defect characterization. Acceleration transmissibility was unable to experimentally identify defects sized at 1.0 cm and below but was able to identify 1.0 cm sized defects numerically. It was concluded that transmissibility is viable for continuous damage detection on blades but that further research into other defect types and locations is required prior to conducting full-scale testing.

Performance-Degradation Analysis of the Planetary Roller Screw Mechanism under Multi-Factor Coupling Effects.

The performance-degradation pattern of the planetary roller screw mechanism (PRSM) is difficult to predict and evaluate due to a variety of factors. Load-carrying capacity, transmission accuracy, and efficiency are the main indicators for evaluating the performance of the PRSM. In this paper, a testing device for the comprehensive performance of the PRSM is designed by taking into account the coupling relationships among temperature rise, vibration, speed, and load. First, the functional design and error calibration of the testing device were conducted. Secondly, the PRSM designed in the supported project was taken as the research object to conduct degradation tests on its load-bearing capacity and transmission accuracy and analyze the changes in transmission efficiency. Third, the thread profile and wear condition were scanned and inspected using a universal tool microscope and an optical microscope. Finally, based on the monitoring module of the testing device, the vibration status during the PRSM testing process was collected in real time, laying a foundation for the subsequent assessment of the changes in the performance state of the PRSM. The test results reveal the law of performance degradation of the PRSM under the coupled effects of temperature, vibration, speed, and load.

Novel Reference Method for the Characterization of PD Measuring Systems Using HFCT Sensors.

During their lifespan, high-voltage (HV) electrical systems are subjected to operating conditions in which electrical, mechanical, thermal and environmental-related stresses occur. These conditions over time lead to unforeseen failures caused by various types of defects. For this reason, there are several technologies for measuring and monitoring the electrical systems, with the aim of minimizing the number of faults. The early detection of defects, preferably in their incipient state, will enable the necessary corrective actions to be taken in order to avoid unforeseen failures. These failures generally lead to human risks and material damage, lack of power supply and significant economic losses. An efficient maintenance technique for the early detection of defects consists of the supervision of the dielectrics status in the installations by means of on-line partial discharge (PD) measurement. Nowadays, there are numerous systems in the market for the measurement of PD in HV installations. The most efficient with a reasonable cost will be those that offer greater security guarantees and the best positioned in the market. Currently, technology developers and users of PD measuring systems face difficulties related to the lack of reference procedures for their complete characterization and to the technical and economic drawback of performing the characterization tests on site or in laboratory installations. To deal with the previous difficulties, in this paper a novel method for the complete and standardized characterization of PD measuring systems is presented. The applicability of this method is mainly adapted for the characterization of systems operating in on-line applications using high-frequency current transformer (HFCT) sensors. For the appropriate application of the method, an associated and necessary scale modular test platform is used. In the test platform, the real on-site measuring conditions of an HV insulated distribution line are simulated in a controlled way. Practical characterizations, showing the convenience and advantages of applying the method using the modular test platform, are also presented.

A Low-Frequency Vibration Sensor Based on Ball Triboelectric Nanogenerator for Marine Pipeline Condition Monitoring.

Marine pipeline vibration condition monitoring is a critical and challenging issue, on account of the complex marine environment, while powering the required monitoring sensors remains problematic. This study introduces a vibration sensor based on a ball triboelectric nanogenerator (B-TENG) for marine pipelines condition monitoring. The B-TENG consists of an acrylic cube, polyester rope, aluminum electrodes, and PTFE ball, which converts vibration signals into electrical signals without the need for an external energy supply. The experimental results show that B-TENG can accurately monitor the frequency, amplitude, and direction of vibration in the range of 1-5 Hz with a small error of 0.67%, 4.4%, and 5%, and an accuracy of 0.1 Hz, 0.97 V/mm, and 1.5°, respectively. The hermetically sealed B-TENG can monitor vibration in underwater environments. Therefore, the B-TENG can be used as a cost-effective, self-powered, highly accurate vibration sensor for marine pipeline monitoring.

Strain Virtual Sensing Applied to Industrial Presses for Fatigue Monitoring.

The techniques that allow one to estimate measurements at the unsensed points of a system are known as virtual sensing. These techniques are useful for the implementation of condition monitoring systems in industrial equipment subjected to high cyclic loads that can cause fatigue damage, such as industrial presses. In this article, three different virtual sensing algorithms for strain estimation are tested using real measurement data obtained from a scaled bed press prototype: two deterministic algorithms (Direct Strain Observer and Least-Squares Strain Estimation) and one stochastic algorithm (Static Strain Kalman Filter). The prototype is subjected to cyclic loads using a hydraulic fatigue testing machine and is sensorized with strain gauges. Results show that sufficiently accurate strain estimations can be obtained using virtual sensing algorithms and a reduced number of strain gauges as input sensors when the monitored structure is subjected to static and quasi-static loads. Results also show that is possible to estimate the initiation of fatigue cracks at critical points of a structural component using virtual strain sensors.

Phase-Resolved Partial Discharge (PRPD) Pattern Recognition Using Image Processing Template Matching.

This paper proposes a new method for recognizing, extracting, and processing Phase-Resolved Partial Discharge (PRPD) patterns from two-dimensional plots to identify specific defect types affecting electrical equipment without human intervention while retaining the principals that make PRPD analysis an effective diagnostic technique. The proposed method does not rely on training complex deep learning algorithms which demand substantial computational resources and extensive datasets that can pose significant hurdles for the application of on-line partial discharge monitoring. Instead, the developed Cosine Cluster Net (CCNet) model, which is an image processing pipeline, can extract and process patterns from any two-dimensional PRPD plot before employing the cosine similarity function to measure the likeness of the patterns to predefined templates of known defect types. The PRPD pattern recognition capabilities of the model were tested using several manually classified PRPD images available in the existing literature. The model consistently produced similarity scores that identified the same defect type as the one from the manual classification. The successful defect type reporting from the initial trials of the CCNet model together with the speed of the identification, which typically does not exceed four seconds, indicates potential for real-time applications.

In-situ tool wear condition monitoring during the end milling process based on dynamic mode and abnormal evaluation.

Rapid tool wear conditions during the manufacturing process are crucial for the enhancement of product quality. As an extension of our recent works, in this research, a generic in-situ tool wear condition monitoring during the end milling process based on dynamic mode and abnormal evaluation is proposed. With the engagement of dynamic mode decomposition, the real-time response of the sensing physical quantity during the end milling process can be predicted. Besides, by constructing the graph structure of the time series and calculating the difference between the predicted signal and the real-time signal, the anomaly can be acquired. Meanwhile, the tool wear state during the end milling process can be successfully evaluated. The proposed method is validated in milling tool wear experiments and received positive results (the mean relative error is recorded as 0.0507). The research, therefore, paves a new way to realize the in-situ tool wear condition monitoring.

Intelligent condition monitoring with CNN and signal enhancement for undersampled signals.

High-frequency signals like vibration and acoustic emission are crucial for condition monitoring, but their high sampling rates challenge data acquisition, especially for online monitoring. Our research developed a novel method for condition identification in undersampled signals using a modified convolutional neural network integrated with a signal enhancement approach. A frequency-domain filtering is applied to suppress similar sidebands and obtain more discriminative features of different conditions, followed by an interpolation-based upsampling in the time domain to restore the signal length and strengthen the low-frequency harmonic information. Enhanced signals are converted into two-dimensional grayscale images for neural network analysis. Tested on bearing datasets and real-world data from regenerative thermal oxidizer lift valve leakage, our method effectively extracts features from low-frequency signals, achieving over 95% fault identification accuracy.

Image-Based Approach Applied to Load Torque Estimation in Three-Phase Induction Motors.

This paper presents a novel method for load torque estimation in three-phase induction motors using air gap flux measurement and the conversion of this type of time-domain signal into grayscale images for further processing as inputs for an inception-type convolutional neural network. The magnetic flux was measured employing a Hall effect sensor installed inside the machine, near the stator slots, and above the stator windings. In this case, the sensor was able to measure a resultant magnetic flux density, having both rotor and stator magnetic flux contributions. The present methodology does not require motor parameters for torque prediction. The proposed approach successfully estimated load torque using three optimizers across almost the entire motor load operational range, spanning from 1.5% to 93.9% of the rated load. Four model configurations achieved a mean absolute percentage error (MAPE) less than or equal to 3.7%. Specifically, two models for a 40 × 50 pixel image achieved MAPE of 3.7% and 3%, one model for a 40 × 25 pixel image achieved a MAPE of 3.5%, and one model for a 50 × 80 pixel image achieved a MAPE of 3.3%. This research has been experimentally validated with a 7.5 kW squirrel cage induction machine.

Automatic Detection of the Running Surface of Railway Tracks Based on Laser Profilometer Data and Supervised Machine Learning.

The measurement of the longitudinal rail profile is relevant to the condition monitoring of the rail infrastructure. The running surface is recognizable as a shiny metallic area on top of the rail head. The detection of the running surface is crucial for vehicle-based rail profile measurements, as well as for defect detection. This paper presents a methodology for the automatic detection of the running surface based on a laser profilometer. The detection of the running surface is performed based on the light reflected from the rail surface. Three rail surfaces with different surface conditions are considered. Supervised machine learning is applied to classify individual surface elements as part of the running surface. Detection by a linear support vector machine is performed with accuracy of >90%. The lateral position of the running surface and its width are calculated. The average deviation from the labeled widths varies between -1.2mm and 5.6mm. The proposed measurement approach could be installed on a train for the future onboard monitoring of the rail network.

Vibration, and temperature run-to-failure dataset of ball bearing for prognostics.

Condition based maintenance (CBM) has become a very important issue in the industry because it can decrease the inventory as the need of parts can be planned by the identification of a potential failure. However, in order to predict the life span of the ball bearing, it is necessary to acquire data according to the all life span of the bearing. This article presents the time-series dataset, including vibration, and temperature, of the ball bearing under run-to-failure. Through the accelerated life test, the ball bearing was failed at 128 working hours, and the vibration and temperature data for the all running section were included. The type of fault was identified through microscopic analysis of the damaged ball bearing. The established dataset can be used to verify newly developed state-of-the-art methods for prognosis the remaining useful life (RUL) of the ball bearing. Mendeley Data. DOI: 10.17632/5hcdd3tdvb.6.

Evaluation of a Condition Monitoring Algorithm for Early Bearing Fault Detection.

Roller bearings are critical components in various mechanical systems, and the timely detection of potential failures is essential for preventing costly downtimes and avoiding substantial machinery breakdown. This research focuses on finding and verifying a robust method that can detect failures early, without creating false positive failure states. Therefore, this paper introduces a novel algorithm for the early detection of roller bearing failures, particularly tailored to high-precision bearings and automotive test bed systems. The featured method (AFI-Advanced Failure Indicator) utilizes the Fast Fourier Transform (FFT) of wideband accelerometers to calculate the spectral content of vibration signals emitted by roller bearings. By calculating the frequency bands and tracking the movement of these bands within the spectra, the method provides an indicator of the machinery's health, mainly focusing on the early stages of bearing failure. The calculated channel can be used as a trend indicator, enabling the method to identify subtle deviations associated with impending failures. The AFI algorithm incorporates a non-static limit through moving average calculations and volatility analysis methods to determine critical changes in the signal. This thresholding mechanism ensures the algorithm's responsiveness to variations in operating conditions and environmental factors, contributing to its robustness in diverse industrial settings. Further refinement was achieved through an outlier detection filter, which reduces false positives and enhances the algorithm's accuracy in identifying genuine deviations from the normal operational state. To benchmark the developed algorithm, it was compared with three industry-standard algorithms: VRMS calculations per ISO 10813-3, Mean Absolute Value of Extremums (MAVE), and Envelope Frequency Band (EFB). This comparative analysis aimed to evaluate the efficacy of the novel algorithm against the established methods in the field, providing valuable insights into its potential advantages and limitations. In summary, this paper presents an innovative algorithm for the early detection of roller bearing failures, leveraging FFT-based spectral analysis, trend monitoring, adaptive thresholding, and outlier detection. Its ability to confirm the first failure state underscores the algorithm's effectiveness.

Centrifugal Pump Fault Detection with Convolutional Neural Network Transfer Learning.

The centrifugal pump is the workhorse of many industrial and domestic applications, such as water supply, wastewater treatment and heating. While modern pumps are reliable, their unexpected failures may jeopardise safety or lead to significant financial losses. Consequently, there is a strong demand for early fault diagnosis, detection and predictive monitoring systems. Most prior work on machine learning-based centrifugal pump fault detection is based on either synthetic data, simulations or data from test rigs in controlled laboratory conditions. In this research, we attempted to detect centrifugal pump faults using data collected from real operational pumps deployed in various places in collaboration with a specialist pump engineering company. The detection was done by the binary classification of visual features of DQ/Concordia patterns with residual networks. Besides using a real dataset, this study employed transfer learning from the image detection domain to systematically solve a real-life problem in the engineering domain. By feeding DQ image data into a popular and high-performance residual network (e.g., ResNet-34), the proposed approach achieved up to 85.51% classification accuracy.

Tool Condition Monitoring Using Machine Tool Spindle Current and Long Short-Term Memory Neural Network Model Analysis.

In cutting processes, tool condition affects the quality of the manufactured parts. As such, an essential component to prevent unplanned downtime and to assure machining quality is having information about the state of the cutting tool. The primary function of it is to alert the operator that the tool has reached or is reaching a level of wear beyond which behaviour is unreliable. In this paper, the tool condition is being monitored by analysing the electric current on the main spindle via an artificial intelligence model utilising an LSTM neural network. In the current study, the tool is monitored while working on a cylindrical raw piece made of AA6013 aluminium alloy with a custom polycrystalline diamond tool for the purposes of monitoring the wear of these tools. Spindle current characteristics were obtained using external measuring equipment to not influence the operation of the machine included in a larger production line. As a novel approach, an artificial intelligence model based on an LSTM neural network is utilised for the analysis of the spindle current obtained during a manufacturing cycle and assessing the tool wear range in real time. The neural network was designed and trained to notice significant characteristics of the captured current signal. The conducted research serves as a proof of concept for the use of an LSTM neural network-based model as a method of monitoring the condition of cutting tools.

A Flexible, Adaptive, and Self-Powered Triboelectric Vibration Sensor with Conductive Sponge-Silicone for Machinery Condition Monitoring.

Vibration sensors for continuous and reliable condition monitoring of mechanical equipment, especially detection points of curved surfaces, remain a great challenge and are highly desired. Herein, a highly flexible and adaptive triboelectric vibration sensor for high-fidelity and continuous monitoring of mechanical vibration conditions is proposed. The sensor is entirely composed of flexible materials. It consists of a conductive sponge-silicone layer and a fluorinated ethylene propylene film. It can detect vibration acceleration of 5 to 50 m s-2 and vibration frequency of 10 to 100 Hz. It has strong robustness and stability, and the output performance barely changes after the durability test of 168 000 working cycles. Additionally, the flexible sensor can work even when the detection point of the mechanical equipment is curved, and the linear fit of the output voltage and acceleration is very close to that when the detection point is flat. Finally, it can be applied to monitoring the working condition of blower and vehicle engine, and can transmit vibration signal to mobile phone application through Wi-Fi module for real-time monitoring. The flexible triboelectric vibration sensor is expected to provide a practical paradigm for smart, green, and sustainable wireless sensor system in the era of Internet of Things.