Change detection is a basic task of remote sensing image processing. The research objective is to identify the change information of interest and filter out the irrelevant change information as ...interference factors. Recently, the rise in deep learning has provided new tools for change detection, which have yielded impressive results. However, the available methods focus mainly on the difference information between multitemporal remote sensing images and lack robustness to pseudochange information. To overcome the lack of resistance in current methods to pseudochanges, in this article, we propose a new method, namely, dual attentive fully convolutional Siamese networks, for change detection in high-resolution images. Through the dual attention mechanism, long-range dependencies are captured to obtain more discriminant feature representations to enhance the recognition performance of the model. Moreover, the imbalanced sample is a serious problem in change detection, i.e., unchanged samples are much more abundant than changed samples, which is one of the main reasons for pseudochanges. We propose the weighted double-margin contrastive loss to address this problem by punishing attention to unchanged feature pairs and increasing attention to changed feature pairs. The experimental results of our method on the change detection dataset and the building change detection dataset demonstrate that compared with other baseline methods, the proposed method realizes maximum improvements of 2.9% and 4.2%, respectively, in the F1 score. Our PyTorch implementation is available at https://github.com/lehaifeng/DASNet.
Line-commutated converter-based-high-voltage direct-current (LCC-HVDC) transmission technology has been developing very rapidly. However, the sensitivity and reliability of HVDC line protection ...require further improvement. An analysis of different line fault types is conducted to show that the normalized voltage change rate without the influence of wave propagation can more effectively distinguish between internal and external faults with high fault resistances. Owing to the difficulty of obtaining an accurate solution for a traveling wave, a novel traveling-wave propagation calculation method is developed by making use of the step response of a transmission line with consideration for its frequency-dependent characteristics. A novel traveling-wave protection (TWP) scheme for two-terminal LCC-HVDC transmission lines is then proposed based on the developed traveling-wave propagation calculation method and fault analysis results. The proposed protection scheme is tested based on fault simulation data using an EMTDC/PSCAD simulation platform and the fault-recorded data from an actual undesired tripping event of conventional TWP. The results demonstrate that the proposed TWP scheme, compared with conventional TWP, can more reliably and correctly detect faults with high transition resistances under different operation modes.
Street Scene Change Detection (SSCD) aims to locate the changed regions between a given street-view image pair captured at different times, which is an important yet challenging task in the computer ...vision community. The intuitive way to solve the SSCD task is to fuse the extracted image feature pairs, and then directly measure the dissimilarity parts for producing a change map. Therefore, the key for the SSCD task is to design an effective feature fusion method that can improve the accuracy of the corresponding change maps. To this end, we present a novel Hierarchical Paired Channel Fusion Network (HPCFNet), which utilizes the adaptive fusion of paired feature channels. Specifically, the features of a given image pair are jointly extracted by a Siamese Convolutional Neural Network (SCNN) and hierarchically combined by exploring the fusion of channel pairs at multiple feature levels. In addition, based on the observation that the distribution of scene changes is diverse, we further propose a Multi-Part Feature Learning (MPFL) strategy to detect diverse changes. Based on the MPFL strategy, our framework achieves a novel approach to adapt to the scale and location diversities of the scene change regions. Extensive experiments on three public datasets (i.e., PCD, VL-CMU-CD and CDnet2014) demonstrate that the proposed framework achieves superior performance which outperforms other state-of-the-art methods with a considerable margin.
In this letter, a novel autonomous control framework "Grid Mind" is proposed for the secure operation of power grids based on cutting-edge artificial intelligence (AI) technologies. The proposed ...platform provides a data-driven, model-free and closed-loop control agent trained using deep reinforcement learning (DRL) algorithms by interacting with massive simulations and/or real environment of a power grid. The proposed agent learns from scratch to master the power grid voltage control problem purely from data. It can make autonomous voltage control (AVC) strategies to support grid operators in making effective and timely control actions, according to the current system conditions detected by real-time measurements from supervisory control and data acquisition (SCADA) or phasor measurement units (PMUs). Two state-of-the-art DRL algorithms, namely deep Q-network (DQN) and deep deterministic policy gradient (DDPG), are proposed to formulate the AVC problem with performance compared. Case studies on a realistic 200-bus test system demonstrate the effectiveness and promising performance of the proposed framework.
As a superfamily of transcription factors (TFs), the basic helix-loop-helix (bHLH) proteins have been characterized functionally in many plants with a vital role in the regulation of diverse ...biological processes including growth, development, response to various stresses, and so on. However, no systemic analysis of the bHLH TFs has been reported in Brachypodium distachyon, an emerging model plant in Poaceae.
A total of 146 bHLH TFs were identified in the Brachypodium distachyon genome and classified into 24 subfamilies. BdbHLHs in the same subfamily share similar protein motifs and gene structures. Gene duplication events showed a close relationship to rice, maize and sorghum, and segment duplications might play a key role in the expansion of this gene family. The amino acid sequence of the bHLH domains were quite conservative, especially Leu-27 and Leu-54. Based on the predicted binding activities, the BdbHLHs were divided into DNA binding and non-DNA binding types. According to the gene ontology (GO) analysis, BdbHLHs were speculated to function in homodimer or heterodimer manner. By integrating the available high throughput data in public database and results of quantitative RT-PCR, we found the expression profiles of BdbHLHs were different, implying their differentiated functions.
One hundred fourty-six BdbHLHs were identified and their conserved domains, sequence features, phylogenetic relationship, chromosomal distribution, GO annotations, gene structures, gene duplication and expression profiles were investigated. Our findings lay a foundation for further evolutionary and functional elucidation of BdbHLH genes.
Practically, in the underdetermined model
y
=
A
x
, where
x
is a
K
sparse vector (i.e., it has no more than
K
nonzero entries), both
y
and
A
could be totally perturbed. A more relaxed condition means ...less number of measurements are needed to ensure the sparse recovery from theoretical aspect. In this paper, based on restricted isometry property (RIP), for subspace pursuit (SP) and compressed sampling matching pursuit (CoSaMP), two relaxed sufficient conditions are presented under total perturbations to guarantee that the sparse vector
x
is recovered. Taking random matrix as measurement matrix, we also discuss the advantage of our condition. Numerical experiments validate that SP and CoSaMP can provide oracle-order recovery performance.
This paper presents a novel time-domain fault location algorithm on HVDC transmission lines. The Bergeron line model is adopted to construct a time-domain fault location equation, considering ...unsynchronized two-end measurement and uncertain line parameters. The zero time reference point is set to the sample point of arrival at each end for the traveling wave-head. Time-tags of all sample points change with the zero time reference. So the time difference of two-end measurements has changed also, and equals the time difference between arrivals at two ends for the traveling wave-head. That makes it possible to mix the traveling wave theory with the Bergeron time-domain fault location method. Besides, the ratio coefficient of wave velocities between different frequencies and digital band-pass filter are presented for better performance of the proposed method. Simulations and tests demonstrate that the proposed method can locate faults accurately on the HVDC lines even in the case of loss of signals from GPS.
This paper introduces an enhanced droop-based dc-voltage control method, including dead-band, for applications to the high-voltage direct-current (HVDC) grid that utilizes the voltage-sourced ...converter (VSC) technology. The proposed droop-control structure also autonomously imposes energy balance between the HVDC grid and its host ac system. The droop-control method (1) divides the VSC stations into four groups, (2) activates the droop-control of each group based on a prespecified voltage margin, and (3) introduces an improved power-voltage characteristic for desirable VSC station dynamic performance. Feasibility and performance of the proposed control method are evaluated based on time-domain simulation studies in the PSCAD platform, using the IEEE-39-Bus system that imbeds a five-terminal VSC-HVDC grid. Each VSC station is a monopolar modular multilevel converter (MMC). The study results show that the proposed droop-control method enables the HVDC-AC system to reach a new steady state after transient events.
Display omitted
Foam has been widely used for profile control, water plugging, and gas channeling control in enhanced oil recovery (EOR). Because of absorption, the surfactant concentration in foam ...decreases along the flow direction. In this study, the effects of surfactant concentration on the foam texture, foam seepage characteristics, and gas saturation were evaluated by core-flow experiments using a high-pressure visualization device. The results show that the surface tension of gas and liquid decreases, the flow resistance in core increases, and the foam takes a longer time to reach stability with the increase in concentration of foaming agent. The blocking capability of foam in porous media is closely related to foam texture. With the increase in surfactant concentration, the average diameter of bubbles in porous media decreases, and the bubbles are more uniform. Moreover, the diameter of bubbles is closer to the average pore diameter of core. Clearly, the diameter matching degree of the maximum distribution frequency of bubbles and the average pore diameter of core increase. Therefore, the blocking capability of foam becomes stronger. With the increase in surfactant concentration, the proportion of trapped gas in porous media increases, the gas saturation in core increases, and the flow resistance of foam increases. However, when the surfactant concentration reaches the critical value, the increasing amplitude of gas saturation decreases. Under an injected foam quality of 50%, the gas saturation is more than 80%. The research results can help to analyze the change along the channel of foam flow in formation and are of great significance to improve the effect of foam application.
Social support has been widely known as a protective factor for the emotional well-being (EWB) of older adults, but less studies have investigated the roles of different sources of social support ...(i.e., family and friend support) on different facets of EWB (i.e., positive affect and negative affect) simultaneously.
In this study, the associations between family/friend support and positive/negative affect were investigated in a sample of 700 Chinese elderly. The EWB and social support were measured with a 12-item affective wordlist (Kahneman et al., 2004) and a self-prepared questionnaire. The results showed that (1) the order of contact frequency and mutual support followed a hierarchical order from spouse, children, to friends; (2) zero-order correlations of both family support and friend support were associated with more positive affect and less negative affect; and when compared with the relative role of family and friend support, (3) spouse (children if spouse is not available) support had greater contribution on decreasing negative affect, while friend support had greater influence on increasing positive affect, even after controlling the demographic, self-rated health and life events variables.
Family and friend support play different roles on the two facets of EWB of the elderly. These results were better explained in light of the task specificity model rather than the hierarchical compensatory model. Moreover, positive affect may be enhanced by friend support (based on personal interests and selectable) rather than family support (bonded by kinship and not selectable), which added evidences to the socioemotional selectivity theory.