Device-free activity recognition plays a crucial role in smart building,security,and human–computer interaction,which shows its strength in its convenience and cost-efficiency.Traditional machine learning has made si...Device-free activity recognition plays a crucial role in smart building,security,and human–computer interaction,which shows its strength in its convenience and cost-efficiency.Traditional machine learning has made significant progress by heuristic hand-crafted features and statistical models,but it suffers from the limitation of manual feature design.Deep learning overcomes such issues by automatic high-level feature extraction,but its performance degrades due to the requirement of massive annotated data and cross-site issues.To deal with these problems,transfer learning helps to transfer knowledge from existing datasets while dealing with the negative effect of background dynamics.This paper surveys the recent progress of deep learning and transfer learning for device-free activity recognition.We begin with the motivation of deep learning and transfer learning,and then introduce the major sensor modalities.Then the deep and transfer learning techniques for device-free human activity recognition are introduced.Eventually,insights on existing works and grand challenges are summarized and presented to promote future research.展开更多
Loss of synchronization is one of the main issues for a grid-feeding converter in a weak grid after being subjected to a large disturbance.The synchronous transient is highly nonlinear due to phase movement and freque...Loss of synchronization is one of the main issues for a grid-feeding converter in a weak grid after being subjected to a large disturbance.The synchronous transient is highly nonlinear due to phase movement and frequency limiters.However,none of the previous research has considered the anti-windup PI in the phase-locked loop,which is commonly implemented in reality and introduced as an additional nonlinear transient.This work provides a taxonomy to evaluate and compare the effect of different anti-windup PI limiters on synchronization stability,including clamping,back-calculation and combined method.Different anti-windup PI limiters allocate zeros and poles differently and have different impacts on damping and stability enhancement.A case study implemented in Matlab/Simulink serves to compare the trajectory of the converter phase and frequency using different anti-windup PI in the scenario of both with and without equilibrium points during the fault.Simulation results show that anti-windup PI limiters increase damping during the fault and thus improve the synchronization stability margin.展开更多
This paper investigates a double auction-based peer-to-peer(P2P)energy trading market for a community of renewable prosumers with private information on reservation price and quantity of energy to be traded.A novel co...This paper investigates a double auction-based peer-to-peer(P2P)energy trading market for a community of renewable prosumers with private information on reservation price and quantity of energy to be traded.A novel competition padding auction(CPA)mechanism for P2P energy trading is proposed to address the budget deficit problem while holding the advantages of the widely-used Vickrey-Clarke-Groves mechanism.To illustrate the theoretical properties of the CPA mechanism,the sufficient conditions are identified for a truth-telling equilibrium with a budget surplus to exist,while further proving its asymptotical economic efficiency.In addition,the CPA mechanism is implemented through consortium blockchain smart contracts to create safer,faster,and larger P2P energy trading markets.The proposed mechanism is embedded into blockchain consensus protocols for high consensus efficiency,and the budget surplus of the CPA mechanism motivates the prosumers to manage the blockchain.Case studies are carried out to show the effectiveness of the proposed method.展开更多
基金This work is supported by NTU Presidential Postdoctoral Fellowship,"Adaptive Multimodal Learning for Robust Sensing and Recognition in Smart Cities"project fund,in Nanyang Technological University,Singapore.
文摘Device-free activity recognition plays a crucial role in smart building,security,and human–computer interaction,which shows its strength in its convenience and cost-efficiency.Traditional machine learning has made significant progress by heuristic hand-crafted features and statistical models,but it suffers from the limitation of manual feature design.Deep learning overcomes such issues by automatic high-level feature extraction,but its performance degrades due to the requirement of massive annotated data and cross-site issues.To deal with these problems,transfer learning helps to transfer knowledge from existing datasets while dealing with the negative effect of background dynamics.This paper surveys the recent progress of deep learning and transfer learning for device-free activity recognition.We begin with the motivation of deep learning and transfer learning,and then introduce the major sensor modalities.Then the deep and transfer learning techniques for device-free human activity recognition are introduced.Eventually,insights on existing works and grand challenges are summarized and presented to promote future research.
基金supported in part by the National Natural Science Foundation of China(U2166601,U2066602)the Science and Technology Department of Xinjiang under Grant No.2021D01C086funded by Sustainable Energy Authority of Ireland(SEAI)by under project FRESLIPS,Grant No.RDD/00681.
文摘Loss of synchronization is one of the main issues for a grid-feeding converter in a weak grid after being subjected to a large disturbance.The synchronous transient is highly nonlinear due to phase movement and frequency limiters.However,none of the previous research has considered the anti-windup PI in the phase-locked loop,which is commonly implemented in reality and introduced as an additional nonlinear transient.This work provides a taxonomy to evaluate and compare the effect of different anti-windup PI limiters on synchronization stability,including clamping,back-calculation and combined method.Different anti-windup PI limiters allocate zeros and poles differently and have different impacts on damping and stability enhancement.A case study implemented in Matlab/Simulink serves to compare the trajectory of the converter phase and frequency using different anti-windup PI in the scenario of both with and without equilibrium points during the fault.Simulation results show that anti-windup PI limiters increase damping during the fault and thus improve the synchronization stability margin.
基金supported by the National Natural Science Foundation of China(No.52207108),and by the Science and Technology Project of State Grid Corporation of China(No.1400202099523 A0000).
文摘This paper investigates a double auction-based peer-to-peer(P2P)energy trading market for a community of renewable prosumers with private information on reservation price and quantity of energy to be traded.A novel competition padding auction(CPA)mechanism for P2P energy trading is proposed to address the budget deficit problem while holding the advantages of the widely-used Vickrey-Clarke-Groves mechanism.To illustrate the theoretical properties of the CPA mechanism,the sufficient conditions are identified for a truth-telling equilibrium with a budget surplus to exist,while further proving its asymptotical economic efficiency.In addition,the CPA mechanism is implemented through consortium blockchain smart contracts to create safer,faster,and larger P2P energy trading markets.The proposed mechanism is embedded into blockchain consensus protocols for high consensus efficiency,and the budget surplus of the CPA mechanism motivates the prosumers to manage the blockchain.Case studies are carried out to show the effectiveness of the proposed method.
