1,721,094 research outputs found
A CNN-based Explainable Fault Diagnosis Model for Gearboxes in Rotating Machinery
No full textConvolutional neural networks(CNN) as a class of deep neural networks are attracting remarkable attention due to their powerful feature extraction capability in various areas such as gearbox fault diagnosis in rotary machinery. Although the identification performance of the CNN has demonstrated a superiority over traditional approaches, it is difficult to explain which parts of the inputs to the CNN are learned by this black box model. Hence, understanding the relationship between the inputs and the deep learning models will help to establish connection to the physical meaning of the fault diagnosis, contributing to the broad acceptance of deep learning models as a trustworthy complement to physical-based reasoning by human experts. In this paper, using Gradient-weighted Class Activation Mapping++ (Grad-CAM++) as the interpreter, the CNNs trained by two-dimensional time-frequency domain signal are interpreted by Grad-CAM++ to show the attention part in these signals by the CNN model
Fatigue crack diagnostics: A comparison of the use of the complex bicoherence and its magnitude
No full textThis paper considers the performance of a novel extension of previous methods for use in the diagnosis of cracks in structures. It concentrates on the comparison of two ways of exploiting the bicoherence as tool to detect cracks: the first method just using the magnitude of the bicoherence and the second using its real and imaginary components. This comparison is conducted using measured data from cracked and un-cracked compressor blades from an aircraft engine
Machine Prognosis with Full Utilization of Truncated Lifetime Data
No full textIntelligent machine fault prognostics estimates how soon and likely a failure will occur with little human expert judgement. It minimizes production downtime, spares inventory and maintenance labour costs. Prognostic models, especially probabilistic methods, require numerous historical failure instances. In practice however, industrial and military communities would rarely allow their engineering assets to run to failure. It is only known that the machine component survived up to the time of repair or replacement but there is no information as to when the component would have failed if left undisturbed. Data of this sort are called truncated data. This paper proposes a novel model, the Intelligent Product Limit Estimator (iPLE), which utilizes truncated data to perform adaptive long-range prediction of a machine component's remaining lifetime. It takes advantage of statistical models' ability to provide useful representation of survival probabilities, and of neural networks ability to recognise nonlinear relationships between a machine component's future survival condition and a given series of prognostic data features. Progressive bearing degradation data were simulated and used to train and validate the proposed model. The results support our hypothesis that the iPLE can perform better than similar prognostics models that neglect truncated data
The Significance of Assets Condition Monitoring in the Development of Engineering Asset Management
No full textFrom its inception, about two decades ago, the basic concept of asset management required government agencies and local government authorities to capitalize and depreciate infrastructure assets according to their residual value, rather than expense them against earnings. Asset management thus embraced the approach of accounting for the condition of physical assets, including concepts such as asset renewal and rehabilitation. Asset management was thus formulated as a systematic, structured process covering the whole life of an asset, from initial design and construction through to application, maintenance, and eventual renewal or disposal. The need therefore arose for a holistic view of the impact of assets condition on the effectiveness of an asset's whole-of-life return on investment. Such a holistic view required integration of condition monitoring techniques and condition monitoring standards with various maintenance regimes. Factors such as appropriate condition monitoring, effective preventive maintenance and a responsive logistic supply system could help keep assets degradation to a minimum; however, it is the asset's inherent maintainability that determines this minimum. Testability, an important subset of maintainability, is a design characteristic that allows the condition status of a physical asset to be determined, and faults to be isolated in a timely and efficient manner. Attention must therefore be paid to ensuring that all asset designs incorporate features that allow for non-destructive testing, and trade-offs must be made on the use of built-in-tests (BIT) versus other means of assets fault detection. However, asset flaws and defects and the need to identify them will never disappear, and continual development of fault detection and characterisation techniques is necessary. There is thus an urgent need for new and applicable approaches to extend the scope of condition monitoring research to meet requirements of Engineering Asset Management. The Cooperative Research Centre for Integrated Engineering Asset Management (CIEAM), an Australian Federal Government funded research organization, incorporates research programs that specifically study the needs of managing engineering assets in both the public and private sectors. This paper presents discussion on the significance of condition monitoring in the development of know-how to manage such assets.No Full Tex
Going Beyond Counting First Authors in Author Co-citation Analysis
Get PDFThe present study examines one of the fundamental aspects of author co-citation analysis (ACA) - the way co-citation
counts are defined. Co-citation counting provides the data on which all subsequent statistical analyses and mappings
are based, and we compare ACA results based on two different types of co-citation counting - the traditional type that
only counts the first one among a cited work's authors on the one hand and a non-traditional type that takes into
account the first 5 authors of a cited work on the other hand. Results indicate that the picture produced through this non-traditional author co-citation counting contains more coherent author groups and is therefore considerably clearer. However, this picture represents fewer specialties in the research field being studied than that produced through the traditional first-author co-citation counting when the same number of top-ranked authors is selected and analyzed. Reasons for these effects are discussed
Variations on the Author
Get PDF“Variations on the Author” discusses two of Eduardo Coutinho’s recent films (Um Dia na Vida, from 2010, and Últimas Conversas, posthumously released in 2015) and their contribution to the general question of documentary authorship. The director’s filmography is characterized by a consistent yet self-effacing form of authorial self-inscription: Coutinho often features as an interviewer that rather than express opinions propels discourses; an interviewer that is good at listening. This mode of self-inscription characterizes him as an author who is not expressive but who is nonetheless markedly present on the screen. In Um Dia na Vida, however, Coutinho is completely absent form the image, while Últimas Conversas, on the contrary, includes a confessional prologue that moves the director from the margins to the center of his films. This article examines the ways in which these works stand out in the filmography of a director who offers new insights into the notion of cinematic authorship
Appropriate Similarity Measures for Author Cocitation Analysis
Get PDFWe provide a number of new insights into the methodological discussion about author cocitation analysis. We first argue that the use of the Pearson correlation for measuring the similarity between authors’ cocitation profiles is not very satisfactory. We then discuss what kind of similarity measures may be used as an alternative to the Pearson correlation. We consider three similarity measures in particular. One is the well-known cosine. The other two similarity measures have not been used before in the bibliometric literature. Finally, we show by means of an example that our findings have a high practical relevance.information science;Pearson correlation;cosine;similarity measure;author cocitation analysis
Dispelling the Myths Behind First-author Citation Counts
Get PDFWe conducted a full-scale evaluative citation analysis study of scholars in the XML research field to explore just how different from each other author rankings resulting from different citation counting methods actually are, and to demonstrate the capability of emerging data and tools on the Web in supporting more realistic citation counting methods. Our results contest some common arguments for the continued
use of first-author citation counts in the evaluation of scholars, such as high correlations between author rankings by first-author citation counts and other citation
counting methods, and high costs of using more realistic citation counting methods that are not well-supported by the ISI databases. It is argued that increasingly available digital full text research papers make it possible for citation analysis studies to go beyond what the ISI databases have directly supported and to employ more
sophisticated methods
- …
