Condition Monitoring Techniques Research Articles

Revolutionizing condition monitoring techniques with integration of artificial intelligence and machine learning

The field of condition monitoring has made great strides in the development of innovative techniques that facilitate early detection and diagnosis of faults in equipment. Present article provides a comprehensive overview of condition-monitoring technologies, namely, ultrasound technique and radiography. By providing detailed insight into the strengths and weaknesses of mentioned techniques, present work highlights their suitability for various industrial applications and how artificial intelligence and machine learning can be integrated to form multimodal systems that improve overall performance. Additionally, it evaluates developments in artificial intelligence and machine learning that enable accurate and reliable condition monitoring systems.

Detecting train driveshaft damages using accelerometer signals and Differential Convolutional Neural Networks

Maintaining railway axles is crucial to prevent catastrophic failures and enormous human and economic costs. In recent years, there has been a growing interest in the railway industry to adopt condition monitoring techniques to enhance the safety and efficiency of the rail transport system, which maintenance is currently based on periodic inspections. In this context, this work presents a technique for real-time crack diagnosis on railway axles, based on advanced 2D-Convolutional Neural Network (CNN) architectures applied to time–frequency representations of vibration signals. One of the critical novelties is introducing a differential CNN structure that captures the system’s statistical properties, enabling generalisation between different mechanical sets and conditions. The proposed system has been trained with data corresponding to a unique wheelset assembly, showing that the model is able to diagnose cracks on the three different wheelset tested in operation under 32 different combinations of conditions, such as load, speed, sense of rotation and vibration direction. Four different crack levels have been introduced, representing the maximum one a 0.08% of the axle diameter, and the method proposed achieved Area Under the Curve (AUC) score of 0.85, significantly outperforming results obtained with other architectures proposed in the state-of-the-art, the score of the next below is 0.76. The results demonstrate the effectiveness and practicality of this approach to accurately classify the four crack levels tested within a condition monitoring system for non-stationary conditions, that would enable reliable real-time diagnosis, thus paving the way towards a more robust and efficient railway axle maintenance system.

Acoustic Analysis Using Sound Level Meter to Determine the Period of Usage of the Spindle Bearing of a Radial Drilling Machine

The condition monitoring techniques like acoustic emission, vibration analysis, and infrared thermography, used for the failure diagnosis of bearings, require longer processing time, as they have to perform acoustical measurement followed by signal processing and further analysis using special software. However, for any bearing, its period of usage can be easily determined within an hour, by measuring the bearing sound, using sound level meter (SLM). In this paper the acoustical analysis of the spindle bearing of a radial drilling machine was performed using SLM, by measuring the sound pressure level of the bearing in decibels, for different frequencies, while muting all the other noises. Then using an experimental set up, two database readings were taken, one for new bearing and the other for completely damaged bearing, both are SKF6207, which itself is the spindle bearing. From these three sets of sound pressure level readings, the period of usage of the spindle bearing, was calculated using an interpolation equation, by substituting the life of the bearing from the manufacturer’s catalogue. Therefore, for any machine with a SKF6207 bearing, its usage time can be estimated using the database readings and one measurement on that machine, all with the same speed.

Inverter-Fed Motor Drive System: A Systematic Analysis of Condition Monitoring and Practical Diagnostic Techniques

Due to their efficiency and control capabilities, induction motors fed with inverters have become prevalent in various industrial applications. However, ensuring the reliable operation of the motor and diagnosing faults on time are crucial for preventing unexpected failures and minimizing downtime. This paper systematically analyzes condition monitoring and practical diagnostic techniques for inverter-fed motor drive systems. This study encompasses a thorough evaluation of different methods used for condition monitoring and diagnostics of induction motors, with the most crucial faults in their stator, rotor, bearings, eccentricity, shaft currents, and partial discharges. It also includes an assessment of their applicability. The presented analysis includes a focus on the challenges associated with inverter-fed systems, such as high-frequency harmonics, common-mode voltages causing the bearing currents, and high voltage gradients (dv/dt) due to fast switching frequency, which can impact the motor operation, as well as its faults analysis. Furthermore, this research explores the usefulness and efficiency of various available diagnostic methods, such as motor current signature analysis and other useful analyses using advanced signal processing techniques. This study aims to present findings that provide valuable insights for developing comprehensive condition monitoring strategies, and practical diagnostic techniques that enable proactive maintenance, enhanced system performance, and improved operational reliability of inverter-fed motor drive systems.

Data-driven condition monitoring of two-stroke marine diesel engine piston rings with machine learning

ABSTRACT Maintaining the condition of a vessel and its equipment guarantees the scheduled completion of voyages and the safety of the crew. This paper presents condition monitoring techniques for early detection of faults related to piston rings in remote cylinders of two-stroke marine diesel engines. Operational sensor data from the main engine of a container ship are provided by a shipping company. A graphical approach complimented by correlation heatmaps and feature importance from gradient boosting trees are used for feature selection. Support Vector Machine, Random Forest and Extreme Gradient Boosting Trees are tested for residual generation from the nominal behavior. The residual time series gives a good indication of the degradation of the system and can be used for alarm raising under strict rules. It is proven that the proposed method could alert the engine crew of a change in the condition of the piston rings much earlier than existing methods.

Analysis of key techniques of condition monitoring and fault diagnosis of mechanical system

Analysis of key techniques of condition monitoring and fault diagnosis of mechanical system

Railway Axle Early Fatigue Crack Detection through Condition Monitoring Techniques

The detection of cracks in rotating machinery is an unresolved issue today. In this work, a methodology for condition monitoring of railway axles is presented, based on crack detection by means of the automatic selection of patterns from the vibration signal measurement. The time waveforms were processed using the Wavelet Packet Transform, and appropriate alarm values for diagnosis were calculated automatically using non-supervised learning techniques based on Change Point Analysis algorithms. The validation was performed using vibration signals obtained during fatigue tests of two identical railway axle specimens, one of which cracked during the test while the other did not. During the test in which the axle cracked, the results show trend changes in the energy of the vibration signal associated with theoretical defect frequencies, which were particularly evident in the direction of vibration that was parallel to the track. These results are contrasted with those obtained during the test in which the fatigue limit was not exceeded, and the test therefore ended with the axle intact, verifying that the effects that were related to the crack did not appear in this case. With the results obtained, an adjusted alarm value for a condition monitoring process was established.

Principal component analysis approach for detecting faults in rotary machines based on vibrational and electrical fused data

Rotating machines are frequently used in industrial applications. However, due to their severity, mechanical failures such as rotor imbalance, shaft imbalance, pulley imbalance, structural breakage, and bearing imbalance can lead to unplanned shutdowns. While vibration analysis-based condition monitoring techniques can detect and diagnose many early errors, certain mechanical faults have associated vibration characteristics that make it difficult to identify and distinguish these faults. To address this issue, this paper proposes a method based on data fusion for vibrational and electrical signatures to achieve new fused signatures for healthy and different faulty cases. The weighted decision fusion method generates the fused decision by weighting and combining the output of multiple sensors. Conventional vibration evaluation parameters diagnose unbalance, pulley misalignment, belt damage, and combined faults. However, these parameters have more dimensions and correlated features for some faulty cases, such as unbalance and misalignment. Therefore, the Principal Component Analysis (PCA) was applied to reduce the dimensionality of evaluating parameters and preserve almost all data variation. The PCA produces uncorrelated Principal Components (PCs) for each case. A backpropagation neural network (BPNN) was constructed to construct an integrated fault diagnosis framework. The first and second PC was inserted as input parameters in the training set of BPNN. It was observed that BPNN achieves 2.1762×10-10 Mean Squared Error (MSE) demonstrates superior data fusion solutions and PCA in the condition monitoring of rotating machines. Overall, this study proposes an effective method for diagnosing mechanical faults in rotating machines, which can improve reliability and reduce downtime in industrial applications.

Development of a machine learning model to predict the friction stir welding tool condition

Tool damage results in surface polishing and dimensional accuracy losses for machined parts, as well as potential damage to a workpiece or machine. In the industrial business, tool condition monitoring (TCM) is significant. As a result, it is essential for production efficiency and profitability. In order to prevent damages from occurring, also known as predictive maintenance, we may use the TCMS to identify tool wear and damage that eventually impacts the workpiece. Machine learning models play a significant part in TCMS by way of a GUI that identifies the tool's failure state. This method can be applied in any kind of system which is used in the industry to predict the damage. This TCMS is applied in a friction stir welding machine (FSW) to predict its tool wear. Friction stir welding, an eco-friendly solid state welding method, is used to join two metals that are difficult to fuse together using conventional fusion welding. FSW is largely recognized as the most important innovation in metal joining technology in the previous decade. This FSW has been widely employed in numerous sectors such as shipbuilding, computer covers, vehicles, and so on. Developing a GUI for the prediction of tool condition in FSW with AZ31B(Mg) as a workpiece is a novel approach. In this condition monitoring technique, the real time vibrational data is collected with the help of an accelerometer. These data’s are being processed and features are extracted and a proper classifier is used to predict the faults. This paper aims to develop a part of TCMS through vibrational data. GUI and ML models were developed to predict the condition of FSW tools. The ML model includes Decision Tree, Random Forest, Light Gradient Boosted Machine Classifiers. Among all these Light Gradient Boosted Machine Classifiers perform better. Using this Classifier a GUI is built to predict the condition.

Power Distribution System Faults and Wildfires: Mechanisms and Prevention

Wildfires are one of the most hazardous natural disasters in Australia in terms of fatalities, property damage and financial losses. Events of catastrophic wildfires are recorded across the world including in the United States and Canada. Failures along power distribution infrastructure and network faults have been identified as some of the causes for the initiation of wildfires. Thus, it is critical to better understand the mechanisms and the potential prevention strategies for wildfires caused by power distribution system faults. In this light, this paper presents how the power distribution network faults cause wildfires highlighting the main mechanisms. Further, this paper reviewed studies on recent advancements for the prediction, detection and prevention of wildfires. Condition assessment of power distribution infrastructure including poles, crossarms, overhead cables and other attachments are paramount to detect potential defects and to carry out timely replacements which can subsequently mitigate the possibility of wildfire initiation. Therefore, this paper summarized the studies on condition monitoring and surveillance techniques for power distribution infrastructure. Altogether, this paper aimed to enhance the awareness about the prevention strategies for wildfires caused by power distribution system faults.

Online condition monitoring system for rotating machine elements using edge computing

ABSTRACT Misalignment, imbalance, induced vibrations, and noise in rotating machines must be identified early on using condition monitoring and signal processing techniques. If it is not detected early, the machine’s reliability will suffer, potentially resulting in a catastrophic failure of the machine components. In this study, a web application for real-time fault detection is designed and built using a novel approach of edge computing and IoT. Vibration signature analysis are used to determine the severity of faults in machine rotating components and to provide an early warning even when the maintenance crew is located in a remote location. The vibration spectrum analysis results are successfully obtained and verified using the vibration-metre tool VibXpert-II. The purpose of this research is to improve real-time condition monitoring of rotating systems like bearings and induction motors and make it available on an online platform for predictive maintenance. Vibration signatures provide more accurate information regarding the type and location of rotor faults than current signatures.

Transmission-error- and vibration-based condition monitoring of gear wear with contaminated lubricant

The presence of contaminant particles in lubricants could cause severe gear wear and damage. Conventional condition monitoring techniques, i.e., gear inspection and oil/wear debris analysis, are often utilised to assess the presence of wear due to contaminants and its severity such as wear depth. However, the traditional approaches require experience or off-line analysis. Although vibration is a non-invasive technique for fault detection and diagnostics, it is less well developed for wear analysis. Recently, an alternative sensor technology, i.e., transmission error (TE), combining angular measurements obtained from the input and output shafts has shown strong potential for wear monitoring. This work aims to investigate the usage of vibration and TE to identify the existence of contaminants and measure the gear wear severity caused by the oil contamination. This is achieved by conducting gear wear tests on a spur gearbox rig, using clean and contaminated lubricants with silica sand. The test in a normal lubrication condition provides a healthy reference of the spur gearbox under a nominal operating condition. In addition to the nominal operating condition, TE and vibration measurements are collected throughout the tests at different operating conditions, together with moulds of the gear tooth surfaces for wear analysis. The results show that the absolute TE can quantitively measure wear depth, and the RMS of vibration can qualitatively correlate with the trend of the average wear depths. The proposed technique extends the field of using TE to diagnose and predict gear wear in contaminated lubricant conditions.

Self-adaptive fusion of local-temporal features for tool condition monitoring: A human experience free model

Cutting tool condition monitoring (TCM) techniques optimize production cost and machining quality, which is recognized as an important enabling technology for smart manufacturing. Unlike the existing monitoring algorithms, which depend on manual operation and domain knowledge in implementation, this paper proposes an autonomous TCM model without human experience based on the automatic selection and fusion of local-temporal features. In the proposed self-adaptive fusion model, the local features are automatically selected and fused by the adaptive fusion of multi-domain features according to their performance. Temporal features can be extracted autonomously by a heterogeneous feature fusion model with the help of a convolutional block attention module. To address the empirical dependence in feature fusion, a method for pre-evaluation of fused features is proposed by calculating the fusion performance indexes, which provides a reliable basis for the selection of dimension reduction methods. Compared with existing fusion methods, the proposed model not only enables the classification of the extracted features but also facilitates the regression and monitoring of specific values of tool wear. The performance of the fused features by self-adaptive fusion is improved by 9.48% on average compared to the best-performing conventional fusion method. Milling experiments show that the proposed model can be robustly adapted to different cutting conditions and achieve optimal performance with minimal computational resources, which provides a theoretical reference for the development of standardized and autonomous TCM.

A Deep Study on Machine Learning Techniques for Tool Condition Monitoring in Turning of Titanium-based Superalloys.

The current state-of-the-art review on tool condition monitoring for turning of titanium-based superalloys is presented in this paper. Titanium (Ti) superalloys are widely utilised in aerospace industry, automobile industry, petrochemical applications. Ti superalloys are also used in fabrication of biomedical components due to their outstanding combination of mechanical properties and strong corrosion resistance at extreme temperatures. But these superalloys are difficult-to-cut because to their low heat conductivity, low elastic modulus, high strength, and strong chemical resistance. Literature review highlights the drastic reduction in tool life of titanium superalloys at highspeed and feed rates throughout the machining process. The review paper focuses on (i) various reasons to deploy tool condition monitoring; and (ii) study of tool condition monitoring methods based on machine learning techniques to identify the ideal parameters for the prevention of catastrophic tool failure.

Piston Slap Condition Monitoring and Fault Diagnosis Using Machine Learning Approach

<div>Various internal combustion (IC) engine condition monitoring techniques exist for early fault detection and diagnosis to ensure smooth operation, increased durability, low emissions, and prevent breakdowns. A fault, such as piston slap, can damage critical components like the piston, piston rings, and cylinder liner and is among those faults that may lead to such consequences. This research has been conducted to monitor piston slap conditions by analyzing the engine vibration and acoustic emission (AE) signals. An experimental setup has been established for acquiring vibration and AE sensor signatures for various piston slap severity conditions. Time-domain features are extracted from vibration and AE sensor signatures, and among them, the best features are selected using one-way analysis of variance (ANOVA) to create machine learning (ML) models. Apart from individual sensor feature classification, the feature fusion method increases the prediction accuracy. ML algorithms used in this study for building the prediction models are classification and regression trees (CART), random forest, and support vector machine (SVM). Performance comparisons of these trained models are made using different performance measures. It is observed that about 94.95% of maximum classification accuracy is obtained in predicting the piston slap severity at different speeds and load conditions.</div>

A review of aircraft environmental control system simulation and diagnostics

The aircraft Environmental Control System (ECS) enables the aircraft to maintain a comfortable and safe environment for its passengers throughout its operating envelope. The Pressurised Air Conditioner (PACK) is the heart of the ECS, and is composed of multiple sub-systems: heat exchangers, valves, compressor, turbine, and a water separator. The PACK’s principle function is to enable conditioning of the hot, high pressure bleed air from the engine or APU, for temperature, pressure and humidity against the cabin requirements. The operation of the PACK is governed by a control system which has the ability to mask degradation in its component during operation until severe degradation or failure results. The required maintenance is then both costly and disruptive. The PACK has been reported as major driver of unscheduled maintenance by the operators. The aviation industry is currently proactively exploring innovative health management solutions that aid the maintenance of aircraft key systems based on predictive based maintenance approaches using online condition monitoring techniques. This paper presents a comprehensive review of the simulation and diagnostic methodologies applicable to fault diagnostics of the ECS PACK. The existing literature suggests that model-based and data-driven methods are effective for conducting fault detection and isolation of the PACK system. The conceived findings indicate that the model-based diagnostic approach have been extensively employed to conduct PACK diagnostics at component level only. Their successful implementation requires robust experimental verification and validation against the actual data under healthy and faulty conditions. Although a substantial amount of work has been reported on developing first principles based simulation models and diagnostic strategies for the ECS, the acquired findings suggest that there is a compelling need for a verified and validated ECS simulation model to enable accurate PACK system-level diagnostics based on single and multiple component level degradation scenarios. It has also been identified that the existing literature lacks the evaluation of humidity regulation and the effect of the control system on the PACK performance characteristics. Finally, a taxonomy of diagnostic techniques and simulation models is compiled based on the available literature.

A Review of Predictive Techniques Used to Support Decision Making for Maintenance Operations of Wind Turbines

The analysis of reliable studies helps to identify the credibility, scope, and limitations of various techniques for condition monitoring of a wind turbine (WT) system’s design and development to reduce the operation and maintenance (O&M) costs of the WT. In this study, recent advancements in data-driven models for condition monitoring and predictive maintenance of wind turbines’ critical components (e.g., bearing, gearbox, generator, blade pitch) are reviewed. We categorize these models according to data-driven procedures, such as data descriptions, data pre-processing, feature extraction and selection, model selection (classification, regression), validation, and decision making. Our findings after reviewing extensive relevant articles suggest that (a) SCADA (supervisory control and data acquisition) data are widely used as they are available at low cost and are extremely practical (due to the 10 min averaging time), but their use is in some sense nonspecific. (b) Unstructured data and pre-processing remain a significant challenge and consume a significant time of whole machine learning model development. (c) The trade-off between the complexity of the vibration analysis and the applicability of the results deserves further development, especially with regards to drivetrain faults. (d) Most of the proposed techniques focus on gearbox and bearings, and there is a need to apply these models to other wind turbine components. We explain these findings in detail and conclude with a discussion of the main areas for future work in this domain.

Introduction to non-destructive testing and condition monitoring techniques in wind energy

Introduction to non-destructive testing and condition monitoring techniques in wind energy

Gate Oxide Degradation Condition Monitoring Technique for High-Frequency Applications of Silicon Carbide Power MOSFETs

Gate oxide degradation, which considerably affects turn- <sc xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">on</small> /- <sc xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">off</small> dynamics of the switch, embraces a large percentage of chip-related failure modes both in silicon and silicon carbide power MOSFETs. The gate oxide layer is thinner in silicon carbide power MOSFETs in comparison to their silicon-based counterparts. Consequently, the problem of gate oxide degradation has become a more crucial impediment in achieving reliable performance in silicon carbide power MOSFETs. This problem is even more severe in high-frequency applications due to higher EMI signature and complicated and costly measurement. In this article, a reliable fully analog cost-effective gate oxide degradation condition monitoring technique is proposed and validated. High-order harmonics magnitudes of drain–source voltage are used to produce a dc signal as the aging precursor of the gate oxide region. Using a dedicated degradation setup, the credibility of the developed condition monitoring technique was examined at different rates of gate oxide degradation for 650-V/22-A silicon carbide discrete MOSFET. In 200-kHz, 217-V switch operation, the proposed precursor showed 68% change in comparison to its initial value. This brings a high-resolution assessment on the reliability level of the switch gate oxide region.

Railway Track and Vehicle Onboard Monitoring: A Review

Railway condition monitoring is the compound of activities aimed at the detection and mitigation of faults in both tracks and vehicles. This minimizes damage and disruption, and increases the usability and benefits of railways. Trackside monitoring provides frequent data, but is costly and ineffective for large segments of track. Onboard monitoring using dedicated vehicles provides deep knowledge of the track, but is infrequent and requires allocated resources and track availability. The use of sensors onboard passenger trains overcomes these issues by acquiring signals more frequently and with relatively inexpensive devices, hence the increasing use of these devices. This review presents the current techniques applied for track and train condition monitoring techniques using onboard sensors, as well as the use of trackside sensors.