Data-driven condition monitoring reduces downtime of wind turbines and increases reliability. Wind turbine operation and maintenance (O&M) cost is a significant factor that calls for automated fault ...detection systems in wind turbines. In this manuscript, the anomaly detection problem for wind turbine gearbox is formulated based on adaptive threshold and twin support vector machine (TWSVM). In this work, SCADA data from wind farms located in the U.K. is considered with samples from twelve months before failure, and from one month before failure. Gearbox oil and bearing temperatures are used as two univariate time-series for analyzing adaptive threshold. The effectiveness of the proposed method is compared with standard classifiers like support vector machines (SVM), k-nearest neighbors (KNN), multi-layer perceptron neural network (MLPNN), and decision tree (DT). Anomaly detection of wind turbine gearbox using TWSVM and adaptive threshold results in an accurate performance, thus increasing the reliability. The missed failure and false positive rate that indicate the proposed methodology's ability is also investigated to discriminate between false alarms, and comparison with previous studies shows superior performance.
Machine condition monitoring (MCM) uses signal processing and machine learning methods to analyze monitoring data and perform timely condition-based maintenance. Since monitoring data usually have a ...sparsity property, sparsity measures (SMs) are naturally considered to quantify the sparsity of signals and they serve as the objective functions of many signal processing and machine learning methods. Although Gini index, kurtosis, smoothness index, negative entropy, and Lp / Lq norm have been considerably investigated for MCM, the design of new SMs for enhancing MCM is rarely reported. In this article, based on the ratio of different quasi-arithmetic means (RQAM), two new SMs, coined as Gini index Ⅱ (GI2) and Gini index Ⅲ (GI3), are designed. New proofs show that the GI2 and GI3 satisfy all six sparsity attributes. Subsequently, the GI2 and GI3 of the square envelope of Gaussian white noise are theoretically investigated and their theoretical values are, respectively, equal to 2/3 and 1/3, which can be used as baselines for machine abnormality detection. Once GI2 and GI3 exceed the baselines, abnormal health conditions can be detected without needing historical data and prior fault knowledge. Finally, simulated and experimental case studies showed that the proposed GI2 and GI3 have competitive performance with Gini index and that they are better than kurtosis, negative entropy, and smoothness index, in characterizing the sparsity of signals. This article demonstrates that the RQAM is a potential framework to design new SMs.
The complex structure of turning aggravates obtaining the desired results in terms of tool wear and surface roughness. The existence of high temperature and pressure make difficult to reach and ...observe the cutting area. In-direct tool condition, monitoring systems provide tracking the condition of cutting tool via several released or converted energy types, namely, heat, acoustic emission, vibration, cutting forces and motor current. Tool wear inevitably progresses during metal cutting and has a relationship with these energy types. Indirect tool condition monitoring systems use sensors situated around the cutting area to state the wear condition of the cutting tool without intervention to cutting zone. In this study, sensors mostly used in indirect tool condition monitoring systems and their correlations between tool wear are reviewed to summarize the literature survey in this field for the last two decades. The reviews about tool condition monitoring systems in turning are very limited, and relationship between measured variables such as tool wear and vibration require a detailed analysis. In this work, the main aim is to discuss the effect of sensorial data on tool wear by considering previous published papers. As a computer aided electronic and mechanical support system, tool condition monitoring paves the way for machining industry and the future and development of Industry 4.0.
Proposed fully interpretable neural network architecture for machine condition monitoring.
Display omitted
•A fully interpretable neural network for machine health monitoring is proposed.•A fully ...interpretable neural network architecture consists of four physical hidden layers.•The first to third hidden layers exhibit the cyclo-stationarity of repetitive transients.•Sparsity measures are used to characterize the cyclo-stationarity of repetitive transients.•An iterative optimization strategy is developed to optimize the proposed neural network.
In recent years, various neural networks have been developed to process vibration signals for machine condition monitoring. Nevertheless, the physical interpretation of neural networks is still on-going and not fully explored. This paper aims to design a fully interpretable neural network for machine condition monitoring from the aspects of signal processing and physical feature extraction. The main idea of the fully interpretable neural network is to extend the uninterpretable structure of extreme learning machine (ELM) to an interpretable structure for machine condition monitoring. From the aspect of signal processing, wavelet transform, square envelope and Fourier transform are incorporated into the input layer of the original ELM to extract repetitive transients, localize informative frequency bands for an enhancement of a signal-to-noise ratio, and realize square envelope spectra for exhibiting cyclo-stationarity of repetitive transients. Hence, the first to four layers of the proposed network are physically interpretable. From the aspect of physical feature extraction, popular sparsity measures are innovatively incorporated into all random nodes in the single-hidden layer of the original ELM to interpret the use of all hidden nodes in the fifth layer of the proposed network to characterize cyclo-stationarity of repetitive transients. The significance of this paper is to show that signal processing algorithms and physical feature extraction can be reformulated as the architecture of an interpretable neural network to automatically localize informative frequency bands for machine condition monitoring. This paper attempts to inspire researchers in the field of signal processing and machine learning to think about the design of more advanced interpretable neural networks for machine condition monitoring.
Recent growth of the insulated gate bipolar transistor (IGBT) module market has been driven largely by the increasing demand for an efficient way to control and distribute power in the field of ...renewable energy, hybrid/electric vehicles, and industrial equipment. For safety-critical and mission-critical applications, the reliability of IGBT modules is still a concern. Understanding the physics-of-failure of IGBT modules has been critical to the development of effective condition monitoring (CM) techniques as well as reliable prognostic methods. This review paper attempts to summarize past developments and recent advances in the area of CM and prognostics for IGBT modules. The improvement in material, fabrication, and structure is described. The CM techniques and prognostic methods proposed in the literature are presented. This paper concludes with recommendations for future research topics in the CM and prognostics areas.
•The automatic monitoring of road conditions for multiple countries is addressed.•Deep Learning models are trained for detecting road damages in India, Japan, and Czech.•Recommendations are provided ...for reusing the data and models released by any country.•A large-scale road damage dataset comprising 26,620 annotated road images is proposed.•The Global Road Damage Detection Challenge’2020 utilizes a part of the proposed data.
Many municipalities and road authorities seek to implement automated evaluation of road damage. However, they often lack technology, know-how, and funds to afford state-of-the-art equipment for data collection and analysis of road damages. Although some countries, like Japan, have developed less expensive and readily available Smartphone-based methods for automatic road condition monitoring, other countries still struggle to find efficient solutions. This work makes the following contributions in this context. Firstly, it assesses usability of Japanese model for other countries. Secondly, it proposes a large-scale heterogeneous road damage dataset comprising 26,620 images collected from multiple countries (India, Japan, and the Czech Republic) using smartphones. Thirdly, it proposes models capable of detecting and classifying road damages in more than one country. Lastly, the study provides recommendations for readers, local agencies, and municipalities of other countries when one other country publishes its data and model for automatic road damage detection and classification. A part of the proposed dataset was utilized for Global Road Damage Detection Challenge’2020 and can be accessed at (https://github.com/sekilab/RoadDamageDetector/).
Analysis of lubricating oil is an effective approach in judging machine's health condition and providing early warning of machine's failure progression. Many studies from both academia and industry ...have been conducted. This paper presents a comprehensive review of the state-of-the-art online sensors for measuring lubricant properties (e.g. wear debris, water, viscosity, aeration, soot, corrosion, and sulfur content). These online sensors include single oil property sensors based on capacitive, inductive, acoustic, and optical sensing and integrated sensors for measuring multiple oil properties. Advantages and disadvantages of each sensing method, as well as the challenges for future developments, are discussed. Research priorities are defined to address the industry needs of machine health monitoring.
Rotor bearing systems (RBSs) play a very valuable role for wind turbine gearboxes, aero−engines, high speed spindles, and other rotational machinery. An in−depth understanding of vibrations of the ...RBSs is very useful for condition monitoring and diagnosis applications of these machines. A new twelve−degree−of−freedom dynamic model for rigid RBSs with a localized defect (LOD) is proposed. This model can formulate the housing support stiffness, interfacial frictional moments including load dependent and load independent components, time−varying displacement excitation caused by a LOD, additional deformations at the sharp edges of the LOD, and lubricating oil film. The time−varying displacement model is determined by a half−sine function. A new method for calculating the additional deformations at the sharp edges of the LOD is analytical derived based on an elastic quarter−space method presented in the literature. The proposed dynamic model is utilized to analyze the influences of the housing support stiffness and LOD sizes on the vibration characteristics of the rigid RBS, which cannot be predicted by the previous dynamic models in the literature. The results show that the presented method can give a new dynamic modeling method for vibration formulation for a rigid RBS with and without the LOD on the races.
•A new 12DOF dynamic model for a rigid rotor bearing system is proposed.•Time-varying excitations caused by a localized defect are formulated.•A new method is presented to calculate additional deformations at defect sharp edges.•Effects of housing support stiffness and defect sizes on vibrations of the system is studied.
A novel scheme is proposed for online condition monitoring of bond wires present in insulated gate bipolar transistor (IGBT) package. The proposed method detects bond wire degradation using on-state ...collector emitter voltage at the inflection point. Previously reported condition monitoring methods based on on-state collector-emitter voltage as a precursor of aging require an accurate knowledge of junction temperature which is difficult to measure online during an inverter operation. The key advantage of the proposed scheme is that it monitors the bond wire degradation irrespective of the junction temperature. Therefore, this technique is not affected by increase in junction temperature due to die attach degradation or change in ambient temperature. The proposed scheme is verified experimentally under realistic operating conditions.
This paper proposes the concept of how machinery condition monitoring can be taken to the next level, through micro-sensing of tribological phenomena occurring between contacting surfaces. By ...considering wear transitions and wear rates it is possible to distinguish between benign and potentially harmful wear scenarios. By measuring the tribological phenomena associated with these conditions, it should then be possible to determine with greater accuracy the health of a machine at any point in its life. For this approach to succeed, it is necessary to develop a comprehensive and holistic monitoring strategy and target sensing technologies for the key wear factors. The paper has two main sections. Firstly, tribological phenomena and the onset of wear which sets out why and what needs to be monitored. The factors influencing the wear process are grouped into three key areas: lubricant condition, tribo-pair condition and operating condition. Through a critical and comprehensive review of developing and state-of the-art tribo-sensing, the second section identifies the potential technologies for monitoring or measuring the physical parameters within these three groupings and thus sets out how the next generation of machine condition monitoring will need to evolve in order to achieve early wear detection and the related benefits.
•The factors influencing wear are grouped into three areas: lubricant condition, tribo-pair condition and operating condition.•In-situ measurements of tribological phenomena provide an early indication of wear, ahead of impending failure.•We use the derivative of instantaneous wear rate (dIWR) to distinguish between natural and induced wear transitions.•Wear transitions and rates distinguish benign and potentially harmful scenarios to determine the health of a machine.•Reviews monitoring tribo-phenomena, identifies existing capability and gaps and how current technologies address these gaps.