On the interpretation of deep learning models in bearing vibration diagnostics

In recent years, data-driven techniques such as deep learning (DL), have been widely represented in the literature in the field of bearing vibration condition monitoring. While these approaches achieve excellent performance in detecting bearing faults on controlled laboratory datasets, there is little information available on their applicability to more realistic working conditions. One challenge of these data-driven approaches is that they can learn non-classical features unrelated to the physical defect, making their generalizability debatable. To overcome the challenge of generalizability in DL models, we aim to first understand the underlying representation that the network uses to classify different bearing defects. Having an interpretable DL model may give us hints on how to increase its applicability by, e.g., data augmentation, changing input representations or adapting model architectures. To benefit from advances in interpretability in DL methods from computer vision, we first transform the vibration signal into an image. We evaluate a common input transformation, namely the spectrogram. Subsequently, the representations that the network has learnt are evaluated. We use the Grad-CAM algorithm together with signal modifications to evaluate which parts of the input signal contribute to class attribution. Our results show that the network learns signal features related to the transfer path, the physical properties of the test setup, rather than picking up classical features having a physical relation with the defect. Given that a transfer path is very machine specific, this could be an explanation for the lack of scalability of DL methods. To improve the generalizability of DL methods on bearing vibration analysis, the competing dominant machine specific features should be eliminated from the input representation. These results highlight the importance of combining domain expertise with data-driven approaches.

Analytical simulation of influential parameters affecting grooved wet clutches performance underdisengagement condition

Innovating new approaches and effective methods to improve the efficiency of mechanical systems in terms of energy losses and environmental effects serve as an attractive domain for researchers and industries. Wet clutches are widely used in power transmission systems in automotive and other tribological mechanisms. The wet clutch has two functional modes; under the engaged state, in which two disks come into contact to each other, and under disengagement the plates are located at a very short distance from each other, and oil flows between them. In disengaged state, the differential speed of driving and driven units causes oil shearing within the clearance which leads to transmission of certain amount of drag torque from the driving to the output shaft. This transferred drag torque is distinguished as power loss in form of heat. The governing physical relation based on continuity equation and Navier–Stokes equations reveals that in a certain rotational velocity, the pressure gradient at the outer radius of the clutch becomes null, and, in this circumstance, aeration occurs that is known as critical rotation speed. Experimental findings provide evidence that geometry manipulation and considering grooves over the frictional disk, reduces the critical rotational speed. But there is a shortage of physical analytical relations to predict the pressure gradient in grooved wet clutches. Therefore, this article is aimed to introduce analytical model to evaluate grooved wet clutches performance in terms of drag torque and critical rotational speed under single-phase flow condition.

Physics of Space Weather Phenomena: A Review

In the last few decades, solar activity has been diminishing, and so space weather studies need to be revisited with more attention. The physical processes involved in dealing with various space weather parameters have presented a challenge to the scientific community, with a threat of having a serious impact on modern society and humankind. In the present paper, we have reviewed various aspects of space weather and its present understanding. The Sun and the Earth are the two major elements of space weather, so the solar and the terrestrial perspectives are discussed in detail. A variety of space weather effects and their societal as well as anthropogenic aspects are discussed. The impact of space weather on the terrestrial climate is discussed briefly. A few tools (models) to explain the dynamical space environment and its effects, incorporating real-time data for forecasting space weather, are also summarized. The physical relation of the Earth’s changing climate with various long-term changes in the space environment have provided clues to the short-term/long-term changes. A summary and some unanswered questions are presented in the final section.

Generalizing from Fission

In Chapter 4, the second step of the Divergence Argument is assessed. This second step consists in generalizing from fission, and, perhaps, other divergent cases, to the conclusion that identity never matters. Various positive rationales offered in support of this inference are examined. One such rationale involves positing cases of identity both without a certain relevant psychological/physical relation (Parfit’s M relation) and without what matters. But it is suggested that there no uncontroversial basis for positing the possibility of cases fitting this pattern. It is also argued that a rationale that appeals to simplicity does not take us to this general result. In addition, it is demonstrated that a further rationale that depends on the claim that the constitution relation between facts about identity and facts about psychological continuity is sufficient to ground this final inference is based on a principle that is subject to counterexamples. It is concluded that it is not plausible to think that the Divergence Argument for the claim that identity never matters is successful.

Head and body structure infants’ visual experiences during mobile, naturalistic play

Infants’ visual experiences are important for learning, and may depend on how information is structured in the visual field. This study examined how objects are distributed in 12-month-old infants’ field of view in a mobile play setting. Infants wore a mobile eye tracker that recorded their field of view and eye movements while they freely played with toys and a caregiver. We measured how centered and spread object locations were in infants’ field of view, and investigated how infant posture, object looking, and object distance affected the centering and spread. We found that far toys were less centered in infants’ field of view while infants were prone compared to when sitting or upright. Overall, toys became more centered in view and less spread in location when infants were looking at toys regardless of posture and toy distance. In sum, this study showed that infants’ visual experiences are shaped by the physical relation between infants’ bodies and the locations of objects in the world. However, infants are able to compensate for postural and environmental constraints by actively moving their head and eyes when choosing to look at an object.

Artificial Intelligence Approach for Tomato Detection and Mass Estimation in Precision Agriculture

Application of computer vision and robotics in agriculture requires sufficient knowledge and understanding of the physical properties of the object of interest. Yield monitoring is an example where these properties affect the quantified estimation of yield mass. In this study, we propose an image-processing and artificial intelligence-based system using multi-class detection with instance-wise segmentation of fruits in an image that can further estimate dimensions and mass. We analyze a tomato image dataset with mass and dimension values collected using a calibrated vision system and accurate measuring devices. After successful detection and instance-wise segmentation, we extract the real-world dimensions of the fruit. Our characterization results exhibited a significantly high correlation between dimensions and mass, indicating that artificial intelligence algorithms can effectively capture this complex physical relation to estimate the final mass. We also compare different artificial intelligence algorithms to show that the computed mass agrees well with the actual mass. Detection and segmentation results show an average mask intersection over union of 96.05%, mean average precision of 92.28%, detection accuracy of 99.02%, and precision of 99.7%. The mean absolute percentage error for mass estimation was 7.09 for 77 test samples using a bagged ensemble tree regressor. This approach could be applied to other computer vision and robotic applications such as sizing and packaging systems and automated harvesting or to other measuring instruments.

Timescale-dependent AMOC-AMO relationship

<p>Previous studies showed that both AMOC and AMO work in different ways in interdecadal and multidecadal timescales. Although their relationship has also been covered in many studies, the possibility that overlapping between multiple timescales may have diluted their inherent relation has not been considered. To understand their physical relation correctly, it is necessary to consider interdecadal and multidecadal timescales, separately.</p><p>Here, we apply a band-pass filter to the AMO and AMOC indices obtained from a present-day climate simulation, to separate interdecadal and multidecadal variability. The results show that strong AMOC induces a warm phase of AMO by the northward heat transport in both timescales, but with a different time lag. This is because, in the interdecadal timescale, the southward propagation of AMOC anomaly gradually warms up the Atlantic basin from the high to low latitudes, resulting in a lag of seven years. As the delayed AMO peak provides negative feedback to AMOC by surface density modulation, the AMOC-AMO relationship can be described as an oscillatory system. On the other hand, AMOC in the multidecadal timescale matures at once in the entire basin, simultaneously warming the surface. The synchronous maturity of AMOC and AMO indicates that AMO-related density changes cannot account for the AMOC phase transition, and AMO remains a relatively passive component in their relationship. This study implies that overlooking timescale-dependency in physical processes may obscure our understanding of interactions between climate components.</p>

A COGNITIVE PERSPECTIVE ON THE ARABIC SPATIAL NOUN/ فَوْقَ /FAWQA/ APPLYING THE PRINCIPLE POLYSEMY MODEL

<p class="TeksAbstrak">The present paper provides a cognitive linguistics (CL) framework for analyzing the semantic structure of Arabic spatial noun <em>fawqa</em> based on <em>Principled Polysemy Model </em>(PPM) of Tyler and Evans (2003). PPM approach can broaden the narrow view of classical cognitive linguists regarding the semantic variation in the concept of physical-geometry of a preposition. As a polysemous lexeme,<em> fawqa</em> used by Arabian native to express a broad range of meanings, not only spatial relation but also non-spatial relation. The substantial sense of the lexeme is investigated using a large amount of corpus data (<em>corpus.kacst.edu.sa</em>) and applying the five steps of PPM approach. Through this approach, we ascertain that every single usage of <em>fawqa </em>expressing extended senses is always in its semantic network. Our study reveals that the usages of this lexeme in many situations and many cases show non-up down spatial relation, and non-physical relation, but they essentially refer to the primary sense.</p>

Refining the origins of the gamma-ray binary 1FGL J1018.6–5856

Context. Gamma-ray binaries are systems composed of a massive star and a compact object that exhibit emission from radio to very high energy gamma rays. They are ideal laboratories to study particle acceleration and a variety of physical processes that vary as a function of the orbital phase. Aims. We aim to study the radio emission of the gamma-ray binary 1FGL J1018.6–5856 to constrain the emitting region and determine the peculiar motion of the system within the Galaxy to clarify its origin. Methods. We analyzed an observation of 1FGL J1018.6–5856 with the Australian Long Baseline Array (LBA) at 8.4 GHz to obtain an accurate astrometry of the system and study its emission on milliarcsecond scales. We combined these data with the optical Gaia DR2 and UCAC4 catalogs to consolidate the astrometry information therein. Results. The gamma-ray binary 1FGL J1018.6–5856 shows compact radio emission (< 3 mas or ≲20 au at ∼6.4 kpc distance), implying a brightness temperature of ≳5.6 × 106 K, and confirming its nonthermal origin. We report consistent results between the proper motion reported by Gaia DR2 and the positions obtained from the Gaia DR2, UCAC4, and LBA data (spanning 20 yr in total). We also determined the distance to 1FGL J1018.6–5856 to be 6.4−0.7+1.7. Together with the radial velocity of the source we computed its three-dimensional (3D) proper and peculiar motion within the Galaxy. We obtained a peculiar motion of 1FGL J1018.6–5856 on its regional standard of rest (RSR) frame of |u| = 45−9+30, with the system moving away from the Galactic plane. In the simplest scenario of a symmetric stellar core collapse we estimate a mass loss of 4 ≲ ΔM ≲ 9 M⊙ during the creation of the compact object. Conclusions. 1FGL J1018.6–5856 exhibits compact radio emission similar to that detected in other gamma-ray binaries. We provide the first accurate peculiar motion estimations of the system and place it within the Galaxy. The obtained motion and distance excludes the physical relation of the binary source with the supernova remnant (SNR) G284.3−1.8.