High penetration of renewable energy sources(RESs)induces sharply-fluctuating feeder power,leading to volt-age deviation in active distribution *** prevent voltage violations,multi-terminal soft open points(M-sOPs)hav...
详细信息
High penetration of renewable energy sources(RESs)induces sharply-fluctuating feeder power,leading to volt-age deviation in active distribution *** prevent voltage violations,multi-terminal soft open points(M-sOPs)have been integrated into the distribution systems to enhance voltage con-trol ***,the M-SOP voltage control recalculated in real time cannot adapt to the rapid fluctuations of photovol-taic(PV)power,fundamentally limiting the voltage controllabili-ty of *** address this issue,a full-model-free adaptive graph deep deterministic policy gradient(FAG-DDPG)model is proposed for M-SOP voltage ***,the attention-based adaptive graph convolutional network(AGCN)is lever-aged to extract the complex correlation features of nodal infor-mation to improve the policy learning ***,the AGCN-based surrogate model is trained to replace the power flow cal-culation to achieve model-free ***,the deep deterministic policy gradient(DDPG)algorithm allows FAG-DDPG model to learn an optimal control strategy of M-SOP by continuous interactions with the AGCN-based surrogate *** tests have been performed on modified IEEE 33-node,123-node,and a real 76-node distribution systems,which demonstrate the effectiveness and generalization ability of the proposed FAG-DDPGmodel.
Effective medium theory(EMT)has been widely applied in material science,electromagnetics and photonics to determine the effective material properties for inhomogeneous composites comprising subwavelength *** versatili...
详细信息
Effective medium theory(EMT)has been widely applied in material science,electromagnetics and photonics to determine the effective material properties for inhomogeneous composites comprising subwavelength *** versatility of this foundational approach has been established through its application in classical Maxwell-Garnet models,which encompass relatively simple structures。
Heart monitoring improves life ***(ECGs or EKGs)detect heart *** learning algorithms can create a few ECG diagnosis processing *** first method uses raw ECG and time-series *** second method classifies the ECG by pati...
详细信息
Heart monitoring improves life ***(ECGs or EKGs)detect heart *** learning algorithms can create a few ECG diagnosis processing *** first method uses raw ECG and time-series *** second method classifies the ECG by patient *** third technique translates ECG impulses into Q waves,R waves and S waves(QRS)features using richer *** ECG signals vary naturally between humans and activities,we will combine the three feature selection methods to improve classification accuracy and *** using all three approaches have not been examined till *** researchers found that Machine Learning(ML)techniques can improve ECG *** study will compare popular machine learning techniques to evaluate ECG *** algorithms—Support Vector Machine(SVM),Decision Tree,Naive Bayes,and Neural Network—compare categorization *** plus prior knowledge has the highest accuracy(99%)of the four ML *** characteristics failed to identify signals without chaos *** 99.8%classification accuracy,the Decision Tree technique outperformed all previous experiments.
Agriculture is crucial to the global economy, particularly in ensuring food security. Recent trends indicate that various plant diseases are causing substantial financial losses in the agricultural sector worldwide. T...
详细信息
In this paper, we study the performance of wireless-powered cluster-based multi-hop cognitive relay networks (MCRNs), where secondary nodes harvest energy from multiple dedicated power beacons (PBs) and share the spec...
详细信息
An Internet of Mobile Things (IoMT) refers to an internetworked group of pervasive devices that coordinate their motion and task execution through frequent status and data exchange. An IoMT could be serving critical a...
详细信息
Detecting and promptly identifying cracks on road surfaces is of paramount importance for preserving infrastructure integrity and ensuring the safety of road users, including both drivers and pedestrians. Presently, t...
详细信息
Liquid crystal elastomers (LCEs) offer potentially programmable actuation through precise molecular alignment, making them ideal for microactuators in soft robotics and optical systems. However, achieving precise micr...
详细信息
This paper proposes a new cluster method combined with Dynamic Mode Decomposition with Control (DMDc), and the Proper Orthogonal Decomposition (POD) to construct more accurate reduced order models. DMDc and POD are po...
Gradient compression is a promising approach to alleviating the communication bottleneck in data parallel deep neural network (DNN) training by significantly reducing the data volume of gradients for synchronization. ...
详细信息
Gradient compression is a promising approach to alleviating the communication bottleneck in data parallel deep neural network (DNN) training by significantly reducing the data volume of gradients for synchronization. While gradient compression is being actively adopted by the industry (e.g., Facebook and AWS), our study reveals that there are two critical but often overlooked challenges: 1) inefficient coordination between compression and communication during gradient synchronization incurs substantial overheads, and 2) developing, optimizing, and integrating gradient compression algorithms into DNN systems imposes heavy burdens on DNN practitioners, and ad-hoc compression implementations often yield surprisingly poor system performance. In this paper, we propose a compression-aware gradient synchronization architecture, CaSync, which relies on flexible composition of basic computing and communication primitives. It is general and compatible with any gradient compression algorithms and gradient synchronization strategies and enables high-performance computation-communication pipelining. We further introduce a gradient compression toolkit, CompLL, to enable efficient development and automated integration of on-GPU compression algorithms into DNN systems with little programming burden. Lastly, we build a compression-aware DNN training framework HiPress with CaSync and CompLL. HiPress is open-sourced and runs on mainstream DNN systems such as MXNet, TensorFlow, and PyTorch. Evaluation via a 16-node cluster with 128 NVIDIA V100 GPUs and a 100 Gbps network shows that HiPress improves the training speed over current compression-enabled systems (e.g., BytePS-onebit, Ring-DGC and PyTorch-PowerSGD) by 9.8%-69.5% across six popular DNN models. IEEE
暂无评论