An Intelligent Methodology for Railways Monitoring Using Ultrasonic Guided Waves

It is far from trivial to inspect railways for defections. In particular, for the foot area of the rail non destructive testing methods are known to be difficult to apply. In this paper, an ultrasonic guided wave method is considered along with classification methods for automated rail foot defect detection. In effect, given a set of gathered ultrasonic signals, multiple features are extracted from time-, frequency- and time–frequency domains. Next, a robust feature selection method is performed, to collect a small set of complementary features. The classification task is accomplished by means of a kernel-based support vector machine. To demonstrate the performance capabilities of our approach, an extensive experimental setup is designed under representative environmental and operational conditions. The sensitivity and the resolution of the proposed defect detection system are reported. A study on the influence of rail fastening on the proposed method is also reported where robust defect detection rates, greater than 93 %, are achieved assuming that a compact feature subset is considered. However, it is evident in experiments that even in the case of large defects, changes in the environmental conditions (temperature and humidity) increase the interpretation of the acquired signals, thus making the detection task more difficult.     

Architecture design and performance analysis of a hybrid hydrogen fuel cell system for unmanned aerial vehicle

The flight endurance of UAV systems is an important issue that restricts the operational capabilities. Thus, different energy systems and alternative onboard energy generation systems have been tested for the UAVs. Within these systems, fuel cells provide high energy density that can increase flight endurance greatly. In this study, a PEM fuel cell – Li-Po battery hybrid system has been developed by evaluating three architecture models. In the guide of the experimental power demand data of a fixed-wing UAV, modeling and testing procedures were performed. Battery voltage and fuel cell current variations observed during the ground tests were evaluated. It has been observed that approximately 160–170 W of the 250 W power is met by the fuel cell since no preconditioning has been applied and the temperature values at which the fuel cell exhibits its optimum performance. In the case where the fuel cell could provide 7.8 An under conditions where the humidification effects were not included in the model, the required current was over 7.8 A between approximately 400-1200 s. The fuel cell and battery behavior in response to the sudden power changes and to the uncertainties corresponding to the changes in the motor power during the flight are also detailed.     

An investigation on the evaluation of dynamic soil characteristics of the Elazig City through the 1-D equivalent linear site-response analysis

Bulletin of Engineering Geology and the Environment - Consideration of the effects of the site response in the design of civil structures systems is of important to mitigate the damages to a...     

Online video scene clustering by competitive incremental NMF

Efficient clustering and categorizing of video are becoming more and more vital in various applications including video summarization, content-based representation and so on. The large volume of video data is the biggest challenge that this task presents, for most the clustering techniques suffer from high dimensional data in terms of both accuracy and efficiency. In addition to this, most video applications require online processing; therefore, clustering should also be done online for such tasks. This paper presents an online video scene clustering/segmentation method that is based on incremental nonnegative matrix factorization (INMF), which has been shown to be a powerful content representation tool for high dimensional data. The proposed algorithm (Comp-INMF) enables online representation of video content and increases efficiency significantly by integrating a competitive learning scheme into INMF. It brings a systematic solution to the issue of rank selection in nonnegative matrix factorization, which is equivalent to specifying the number of clusters. The clustering performance is evaluated by tests on TRECVID video sequences, and a performance comparison to baseline methods including Adaptive Resonance Theory (ART) is provided in order to demonstrate the efficiency and efficacy of the proposed video clustering scheme. Clustering performance reported in terms of recall, precision and F1 measures shows that the labeling accuracy of the algorithm is notable, especially at edit effect regions that constitute a challenging point in video analysis.