Over the past few decades, video quality assessment (VQA) has become a valuable research field. The perception of in-the-wild video quality without reference is mainly challenged by hybrid distortions with dynamic variations and the movement of the content. In order to address this barrier, we propose a no-reference video quality assessment (NR-VQA) method that adds the enhanced awareness of dynamic information to the perception of static objects. Specifically, we use convolutional networks with different dimensions to extract low-level static-dynamic fusion features for video clips and subsequently implement alignment, followed by a temporal memory module consisting of recurrent neural networks branches and fully connected (FC) branches to construct feature associations in a time series. Meanwhile, in order to simulate human visual habits, we built a parametric adaptive network structure to obtain the final score. We further validated the proposed method on four datasets (CVD2014, KoNViD-1k, LIVE-Qualcomm, and LIVE-VQC) to test the generalization ability. Extensive experiments have demonstrated that the proposed method not only outperforms other NR-VQA methods in terms of overall performance of mixed datasets but also achieves competitive performance in individual datasets compared to the existing state-of-the-art methods.
In this paper, Methods of SOC estimation of Extended Kalman Filter (EKF) is studied based on the characteristics of Nickel-Metal Hydride (Ni-MH) battery pack with 120 cells in series and 8Ah capacity for HEV. In the study of EKF-based SOC estimation, the improved Thevenin circuit model is adopted, and a new hybrid pulse power characterization (HPPC) test is designed to identify the model parameters by using piecewise linear regression method. In this way, the precision of the circuit model is improved. In addition, The Kalman gain matrix is optimized for EKF iterative algorithm by two ways: a constant gain is increased taking into account the entire process; a dynamic gain which increases at the beginning of abrupt change and decreases rapidly after abrupt change is set up. The improvement achieves a good tracing prediction.
An outline of electric vehicles industry in the national smart grid plan and a conceptual framework for the vehicle-to-grid implementation in China is presented, and the relationship between battery energy package and smart grid is also discussed in this paper. The analysis and research based on the energy package management and charging station requirements are stated. This paper also does research on reasonable charge and discharge of the Li-ion battery with the performance degradation, which are the key issues of the development of electric vehicles. At last the paper will discuss briefly on the fast charge method, which saves time of charge battery and favors life span of battery system. When plenty of electric vehicles connecting to the grid, advanced energy scheduling and optimization control strategy of Li-ion battery demand consideration of time factor and the gird needs.
As a new type of transport, electric vehicles are becoming an important way to implement the energy and sustainable development of the city. Aiming at the Battery Swapping Stations (BSS) for electric buses, a strategy for coordinated charging is proposed to reduce the charging cost and decrease the load peak-valley difference of Distribution Network (DN). The load model of BSS is built based on the operation analysis of BSS and the charging data of batteries. The optimization of charging behavior imposed by the coordinated charging strategy is also presented. Finally, studies are carried out considering the Beitucheng BSS (for Beijing Olympic Games); analyses are also presented considering different number of batteries to take into account the actual configuration. Simulations show that the coordinated charging, vis-à-vis uncontrolled charging provides satisfactory results, not only reducing the charging cost but also decreasing the load peak-valley difference of DN.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.