非侵入式负荷监测(non-intrusive load monitoring,NILM)技术绿色节能,已成为电力系统负荷监测的发展趋势。集成学习方法可有效提高负荷识别性能,但其基学习器的优化选择和权重设置问题亟待解决。文中以一种典型智能电表对8种小型用电...非侵入式负荷监测(non-intrusive load monitoring,NILM)技术绿色节能,已成为电力系统负荷监测的发展趋势。集成学习方法可有效提高负荷识别性能,但其基学习器的优化选择和权重设置问题亟待解决。文中以一种典型智能电表对8种小型用电设备及其混合负荷的高频实测实验为基础,基于递归特征消除(recursive feature elimination,RFE)法选择最优特征组合,提出结合准确率和多样性权衡的基学习器组合优化方法,并引入香农熵设置投票权重,形成一种新颖的基于香农熵加权投票的集成式NILM识别方法。通过在自测数据集和公开的全球家庭和行业瞬态能量数据集(worldwide household and industry transient energy dataset,WHITED)验证,与常用集成方法比较,该方法识别准确率高、运行时间短且稳定性高。展开更多
This paper proposes a new non-intrusive trigonometric polynomial approximation interval method for the dynamic response analysis of nonlinear systems with uncertain-but-bounded parameters and/or initial conditions.Thi...This paper proposes a new non-intrusive trigonometric polynomial approximation interval method for the dynamic response analysis of nonlinear systems with uncertain-but-bounded parameters and/or initial conditions.This method provides tighter solution ranges compared to the existing approximation interval methods.We consider trigonometric approximation polynomials of three types:both cosine and sine functions,the sine function,and the cosine function.Thus,special interval arithmetic for trigonometric function without overestimation can be used to obtain interval results.The interval method using trigonometric approximation polynomials with a cosine functional form exhibits better performance than the existing Taylor interval method and Chebyshev interval method.Finally,two typical numerical examples with nonlinearity are applied to demonstrate the effectiveness of the proposed method.展开更多
.A non-intrusive reduced order model(ROM)that combines a proper orthogonal decomposition(POD)and an artificial neural network(ANN)is primarily studied to investigate the applicability of the proposed ROM in recovering....A non-intrusive reduced order model(ROM)that combines a proper orthogonal decomposition(POD)and an artificial neural network(ANN)is primarily studied to investigate the applicability of the proposed ROM in recovering the solutions with shocks and strong gradients accurately and resolving fine-scale structures efficiently for hyperbolic conservation laws.Its accuracy is demonstrated by solving a high-dimensional parametrized ODE and the one-dimensional viscous Burgers’equation with a parameterized diffusion coefficient.The two-dimensional singlemode Rayleigh-Taylor instability(RTI),where the amplitude of the small perturbation and time are considered as free parameters,is also simulated.An adaptive sampling method in time during the linear regime of the RTI is designed to reduce the number of snapshots required for POD and the training of ANN.The extensive numerical results show that the ROM can achieve an acceptable accuracy with improved efficiency in comparison with the standard full order method.展开更多
An outdoor flapping wing micro air vehicle （FWMAV） should be able to withstand unpredictable perturbations in the flight condition. The responses of the time-averaged thrust coefficient and the propulsive efficiency...An outdoor flapping wing micro air vehicle （FWMAV） should be able to withstand unpredictable perturbations in the flight condition. The responses of the time-averaged thrust coefficient and the propulsive efficiency with respect to a stochastic flight velocity deviation were numerically investigated. The deviation was assumed to obey the Gauss distribution with a mean value of zero and a specified standard deviation. The probability distributions of the flapping performances were quantified by the non-intrusive polynomial chaos method. It was observed that both of the time-averaged thrust coefficient and the propulsive efficiency obeyed Gauss-like but not the exact Gauss distribution. The velocity deviation had a large effect on the time-averaged thrust coefficient and a small effect on the propulsive efficiency.展开更多
文摘非侵入式负荷监测(non-intrusive load monitoring,NILM)技术绿色节能,已成为电力系统负荷监测的发展趋势。集成学习方法可有效提高负荷识别性能,但其基学习器的优化选择和权重设置问题亟待解决。文中以一种典型智能电表对8种小型用电设备及其混合负荷的高频实测实验为基础,基于递归特征消除(recursive feature elimination,RFE)法选择最优特征组合,提出结合准确率和多样性权衡的基学习器组合优化方法,并引入香农熵设置投票权重,形成一种新颖的基于香农熵加权投票的集成式NILM识别方法。通过在自测数据集和公开的全球家庭和行业瞬态能量数据集(worldwide household and industry transient energy dataset,WHITED)验证,与常用集成方法比较,该方法识别准确率高、运行时间短且稳定性高。
文摘This paper proposes a new non-intrusive trigonometric polynomial approximation interval method for the dynamic response analysis of nonlinear systems with uncertain-but-bounded parameters and/or initial conditions.This method provides tighter solution ranges compared to the existing approximation interval methods.We consider trigonometric approximation polynomials of three types:both cosine and sine functions,the sine function,and the cosine function.Thus,special interval arithmetic for trigonometric function without overestimation can be used to obtain interval results.The interval method using trigonometric approximation polynomials with a cosine functional form exhibits better performance than the existing Taylor interval method and Chebyshev interval method.Finally,two typical numerical examples with nonlinearity are applied to demonstrate the effectiveness of the proposed method.
基金funding support of this research by the National Natural Science Foundation of China(11871443)Shandong Provincial Qingchuang Science and Technology Project(2019KJI002)the Ocean University of China for providing the startup funding(201712011)that is used in supporting this work.
文摘.A non-intrusive reduced order model(ROM)that combines a proper orthogonal decomposition(POD)and an artificial neural network(ANN)is primarily studied to investigate the applicability of the proposed ROM in recovering the solutions with shocks and strong gradients accurately and resolving fine-scale structures efficiently for hyperbolic conservation laws.Its accuracy is demonstrated by solving a high-dimensional parametrized ODE and the one-dimensional viscous Burgers’equation with a parameterized diffusion coefficient.The two-dimensional singlemode Rayleigh-Taylor instability(RTI),where the amplitude of the small perturbation and time are considered as free parameters,is also simulated.An adaptive sampling method in time during the linear regime of the RTI is designed to reduce the number of snapshots required for POD and the training of ANN.The extensive numerical results show that the ROM can achieve an acceptable accuracy with improved efficiency in comparison with the standard full order method.
基金Supported by the National Natural Science Foundation of China(10972034)the National Science Foundation for Postdoctoral Scientists of China(20090460216)the National Defense Fundamental Research Foundation of China(B222006060)
文摘An outdoor flapping wing micro air vehicle （FWMAV） should be able to withstand unpredictable perturbations in the flight condition. The responses of the time-averaged thrust coefficient and the propulsive efficiency with respect to a stochastic flight velocity deviation were numerically investigated. The deviation was assumed to obey the Gauss distribution with a mean value of zero and a specified standard deviation. The probability distributions of the flapping performances were quantified by the non-intrusive polynomial chaos method. It was observed that both of the time-averaged thrust coefficient and the propulsive efficiency obeyed Gauss-like but not the exact Gauss distribution. The velocity deviation had a large effect on the time-averaged thrust coefficient and a small effect on the propulsive efficiency.
