Moving vehicles equipped with various types of sensors can efficiently monitor the health conditions of a population of transportation infrastructure such as bridges. This paper presents a mobile crowdsensing framework to identify dense spatial-resolution bridge mode shapes using sparse drive-by measurements. The proposed method converts mode shape identification into a physical-informed optimization problem with two objective function terms. The first objective minimises the mode shape identification error based on the fact that the ratio of a specific order mode shape value at any two locations is time-invariant. Since the bridge mode shape should be globally smooth even when the local stiffness is discontinuous, the smoothness of the identified mode shape is introduced as the second objective. The feasibility and advantages of the proposed model are verified numerically and through large-scale experimental studies. Numerical results demonstrate that the proposed method can efficiently identify bridge mode shapes with a desirable accuracy. The adverse effects of road roughness and measurement noise on the mode shape identification accuracy are substantially suppressed by introducing crowdsensing and making use of collected responses over multiple trips. The applicability of the proposed method for bridges having varying cross sections and multiple spans is also studied. A series of drive-by tests with different vehicle masses and speeds are conducted on a large-scale footbridge. The experimental results verify that the proposed method can accurately identify the bridge mode shapes and is robust to vehicle mass and speed variation. The identification accuracy of large-scale bridge mode shapes using crowdsensing drive-by measurements is demonstrated in this study.
Shot boundary detection is an important fundamental process toward automatic video indexing, retrieval, editing, etc. After a critical review of most approaches seeking to solve this problem, we propose a novel shot boundary detection. To improve the performance of the algorithm and reduce the computational cost, frames that are clearly not shot boundaries are first removed from the original video. After that, a novel SIFT keypoint matching algorithm based on SVM is proposed, which is used to capture the changing statistics of different kinds of shot transitions so as to identify, not only abrupt transitions, but also gradual transitions(fade, dissolve, wipe) accordingly. At last, our system use different algorithms for different kinds of shot transitions to help us to get a better solution for shot boundary detection problem. Numerical experiments in the evaluation of TRECVID and a variety of film videos demonstrate that our method is capable of accurately detecting shot transitions, and could greatly reduce the computational cost.
This paper proposes a relative displacement sensor developed to measure directly the relative slip between slab and girder in composite bridges for assessing the health condition of shear connections. The structure, design principle, features, and calibration of the developed relative displacement sensor are presented. The design of the sensor ensures that there are no voltage outputs for the tension, compression, bending, and torsion effects, but only for the relative displacement between the two connecting pads of the sensor. The accuracy of the developed sensor in measuring the relative displacement response and using it for monitoring the conditions of shear connectors was tested on a composite bridge model in the laboratory. Shear connection condition was monitored under ambient vibrations, then static load tests were conducted to introduce cracks into the composite bridge. Both the vertical deflections and relative displacements were used for the crack detection. Experimental studies demonstrate that the developed sensor is very sensitive to the relative displacement and has a decent performance for the structural health monitoring of composite bridges. Copyright © 2014 John Wiley & Sons, Ltd.
The new-type power system has relatively low inertia due to the substantial replacement of synchronous generators (SGs) by converter-interfaced generators (CIGs). Low inertia may result in faster frequency dynamics and threaten the frequency stability of the new-type power system. This paper investigates the inertia response characteristic of typical devices in the new-type power system. By the analogy of the mathematical form of SG inertia, the inertia of asynchronous devices, such as asynchronous motors and CIGs with virtual synchronous generator (VSG) control, can be obtained. The analysis is significant for evaluating of inertia resources in the new-type power system.
Introduction Subsidy policies and carbon tax policies are effective means of encouraging the development of low-carbon residential buildings. However, how to utilize these two policy tools of carbon tax and subsidies to more efficiently incentivize the development of low-carbon residential buildings remains a topic worthy of further exploration. This paper explores the mechanisms by which carbon tax and subsidies incentivize the implementation of low-carbon residential buildings and clarifies the boundary conditions for the effectiveness of these two policy tools. Methods We built a stochastic evolutionary game that couples three agent groups—government, real-estate developers, and homebuyers—and explicitly incorporates policy shocks (carbon tax vs. subsidy) and environmental noise. A series of Monte-Carlo simulations was run to trace the share of low-carbon developers and buyers over time; critical thresholds and noise boundaries separating qualitatively different regimes were identified with bifurcation analysis. Results and discussion The evolutionary process of low-carbon housing development, whether under a carbon tax policy or a low-carbon subsidy policy, includes three stages: (1) the initial stage, where relying solely on government subsidies or carbon taxes is insufficient to effectively implement low-carbon housing under various noise intensities; (2) the development stage, where as the proportions of real estate developers adopting low-carbon strategies and homebuyers purchasing low-carbon housing increase, a chaotic relationship emerges between the implementation of carbon taxes, subsidies, and low-carbon housing under certain noise intensities; and (3) the stable stage, where low-carbon housing can be successfully implemented and a qualitative leap is achieved when a certain threshold proportion of real estate developers and homebuyers adopt low-carbon strategies.
The study of the slurry reinforcement mechanism is mainly focused on the interaction between slurry and soil. The seepage effect of the slurry always exists no matter what way the slurry interacts with the soil around the pile. In the process of slurry diffusion, the porosity of the soil, the permeability of the slurry, and the slurry pressure vary due to some cement particles being blocked by the soil particle skeleton. Therefore, the study of the slurry filtration effect is of great significance for predicting the permeation and diffusion law of slurry. In this paper, a macroscopic linear filtration model was introduced and the changes of slurry properties in the permeation diffusion process were considered. Firstly, a spherical (cylindrical) permeation diffusion model, which takes the linear filtration effect and the variation of slurry viscosity into account, was derived based on the conservation of mass. Furthermore, in order to more accurately reflect the influence of the filtration effect on the slurry permeation diffusion model, a polynomial nonlinear filtration model was proposed, and the numerical solution for the permeation diffusion model was derived using finite difference and finite element methods. Finally, the numerically simulated values, the measured values, and the values from the spherical permeation diffusion model that does not consider slurry viscosity variations were compared. The results indicate that the grout pressure is inconsistent with the measured value without considering the effect of the filtration. The initial grouting pressure calculated by the model in this paper is slightly larger, and the required grouting pressure over time is greater than that without considering the filtration effect, regardless of whether the grout diffuses in a spherical or cylindrical manner. The results of this study can contribute to a better understanding of grouting engineering and provide some theoretical guidance for actual grouting.
The aerodynamic performance and wake structure of dandelion seed pappus have been numerically studied based on a simplified quasi-dandelion pappus (QDP) model with its filaments represented by rectangular cylinders. The filament width is chosen as the major geometric parameter for investigation. A rigorous measuring strategy is developed for the identification of the recirculation region width in the wake of the QDP model. Three regimes are distinguished as the filament width increases, i.e., a dandelion-like regime, a transition regime, and a disk-like regime. In the dandelion-like regime, the recirculation region widths are relatively large and monotonously decrease with the increase in Reynolds number. In the transition regime, the recirculation region widths are moderate and first decrease sharply at low Reynolds number and subsequently maintain an approximately invariant value. In the disk-like regime, the recirculation region widths are relatively small. The Reynolds number based on the recirculation region width is defined, and its correlation to the drag coefficient in a different regime is also discussed. In addition, as the QDP model turns from the dandelion-like regime to the disk-like regime, the pressure distribution in the wake turns from the recirculation region type to the flow stagnation type. The current study may provide a reference for the design of more efficient dandelion-like aircraft.
In this paper, the different sizes of enlarged weld access hole (EWAH) in steel beam-column connection are presented. Based on the commercial nonlinear finite element code ABAQUS, firstly, the sequential welding process is simulated, and the non-linear pseudo-static is performed. Through the comparison of results between the FEA and finished pseudo-static experiment when EWAH size is adopted by x =115mm and y =35mm, it is demonstrated that this analysis method is rational and effective, moreover, the stress concentration effects and cracks are discovered. And then the FEA method still obtains, the EWAH size y =35mm is changeless, the nine structural models are made up for only considering the value range from x =70mm to x =110mm. Furthermore, the impacts of the EWAH size on mechanical behaviour are discussed, and the reasonable geometric range is proposed.
Inspired by the progress of the End-to-End approach [1], this paper systematically studies the effects of Number of Filters of convolutional layers on the model prediction accuracy of CNN+RNN (Convolutional Neural Networks adding to Recurrent Neural Networks) for ASR Models (Automatic Speech Recognition). Experimental results show that only when the CNN Number of Filters exceeds a certain threshold value is adding CNN to RNN able to improve the performance of the CNN+RNN speech recognition model, otherwise some parameter ranges of CNN can render it useless to add the CNN to the RNN model. Our results show a strong dependency of word accuracy on the Number of Filters of convolutional layers. Based on the experimental results, the paper suggests a possible hypothesis of Sound-2-Vector Embedding (Convolutional Embedding) to explain the above observations. Based on this Embedding hypothesis and the optimization of parameters, the paper develops an End-to-End speech recognition system which has a high word accuracy but also has a light model-weight. The developed LVCSR (Large Vocabulary Continuous Speech Recognition) model has achieved quite a high word accuracy of 90.2% only by its Acoustic Model alone, without any assistance from intermediate phonetic representation and any Language Model. Its acoustic model contains only 4.4 million weight parameters, compared to the 35~68 million acoustic-model weight parameters in DeepSpeech2 [2] (one of the top state-of-the-art LVCSR models) which can achieve a word accuracy of 91.5%. The light-weighted model is good for improving the transcribing computing efficiency and also useful for mobile devices, Driverless Vehicles, etc. Our model weight is reduced to ~10% the size of DeepSpeech2, but our model accuracy remains close to that of DeepSpeech2. If combined with a Language Model, our LVCSR system is able to achieve 91.5% word accuracy.
