随着第五代通信网络技术(5G)的发展,智慧城市中物联网(Internet of Things,IoT)的应用规模和多样性呈现出爆炸式增长.海量的智能传感设备组网给高动态的物联网通信服务质量带来了巨大的威胁.部分关键设备节点的失效以及网络攻击易引发...随着第五代通信网络技术(5G)的发展,智慧城市中物联网(Internet of Things,IoT)的应用规模和多样性呈现出爆炸式增长.海量的智能传感设备组网给高动态的物联网通信服务质量带来了巨大的威胁.部分关键设备节点的失效以及网络攻击易引发物联网的链锁崩塌效应,影响网络应用的服务质量.因此,如何优化大规模物联网拓扑的鲁棒能力成为当下的研究挑战.目前,针对物联网拓扑结构的优化问题,研究者们提出了启发式算法、智能学习机制和多目标优化策略等创新方法提高物联网拓扑结构的鲁棒能力.但是,这些方法需牺牲巨大的计算资源来获得不成比例的鲁棒性能增益,网络规模越大,该现象越明显.为了解决这个问题并平衡计算开销和提升鲁棒性能,本文提出了一种基于网络模体(Motif)的轻量级物联网拓扑优化策略LITOS.首先利用物联网拓扑结构的社区属性,设计一种基于网络模体的异步社区发现算法,将大规模复杂拓扑结构分解为轻量级局部网络拓扑.然后,基于CPU多核心的计算资源,设计深度强化学习机制,异步优化轻量级物联网局部拓扑结构,从而降低网络整体优化运行时间,提高拓扑结构鲁棒能力.在实验方面,与其他先进的优化算法相比,该策略在运行时间方面降低了1~2个数量级,在鲁棒性提升方面,与最优算法相差大约10%.展开更多
The effects of cyanidation conditions on gold dissolution were studied by artificial neural network （ANN） modeling. Eighty-five datasets were used to estimate the gold dissolution. Six input parameters, time, solid ...The effects of cyanidation conditions on gold dissolution were studied by artificial neural network （ANN） modeling. Eighty-five datasets were used to estimate the gold dissolution. Six input parameters, time, solid percentage, P50 of particle, NaCN content in cyanide media, temperature of solution and pH value were used. For selecting the best model, the outputs of models were compared with measured data. A fourth-layer ANN is found to be optimum with architecture of twenty, fifteen, ten and five neurons in the first, second, third and fourth hidden layers, respectively, and one neuron in output layer. The results of artificial neural network show that the square correlation coefficients （R2） of training, testing and validating data achieve 0.999 1, 0.996 4 and 0.9981, respectively. Sensitivity analysis shows that the highest and lowest effects on the gold dissolution rise from time and pH, respectively It is verified that the predicted values of ANN coincide well with the experimental results.展开更多
Accurate 3-D fracture network model for rock mass in dam foundation is of vital importance for stability,grouting and seepage analysis of dam foundation.With the aim of reducing deviation between fracture network mode...Accurate 3-D fracture network model for rock mass in dam foundation is of vital importance for stability,grouting and seepage analysis of dam foundation.With the aim of reducing deviation between fracture network model and measured data,a 3-D fracture network dynamic modeling method based on error analysis was proposed.Firstly,errors of four fracture volume density estimation methods(proposed by ODA,KULATILAKE,MAULDON,and SONG)and that of four fracture size estimation methods(proposed by EINSTEIN,SONG and TONON)were respectively compared,and the optimal methods were determined.Additionally,error index representing the deviation between fracture network model and measured data was established with integrated use of fractal dimension and relative absolute error(RAE).On this basis,the downhill simplex method was used to build the dynamic modeling method,which takes the minimum of error index as objective function and dynamically adjusts the fracture density and size parameters to correct the error index.Finally,the 3-D fracture network model could be obtained which meets the requirements.The proposed method was applied for 3-D fractures simulation in Miao Wei hydropower project in China for feasibility verification and the error index reduced from 2.618 to 0.337.展开更多
文摘随着第五代通信网络技术(5G)的发展,智慧城市中物联网(Internet of Things,IoT)的应用规模和多样性呈现出爆炸式增长.海量的智能传感设备组网给高动态的物联网通信服务质量带来了巨大的威胁.部分关键设备节点的失效以及网络攻击易引发物联网的链锁崩塌效应,影响网络应用的服务质量.因此,如何优化大规模物联网拓扑的鲁棒能力成为当下的研究挑战.目前,针对物联网拓扑结构的优化问题,研究者们提出了启发式算法、智能学习机制和多目标优化策略等创新方法提高物联网拓扑结构的鲁棒能力.但是,这些方法需牺牲巨大的计算资源来获得不成比例的鲁棒性能增益,网络规模越大,该现象越明显.为了解决这个问题并平衡计算开销和提升鲁棒性能,本文提出了一种基于网络模体(Motif)的轻量级物联网拓扑优化策略LITOS.首先利用物联网拓扑结构的社区属性,设计一种基于网络模体的异步社区发现算法,将大规模复杂拓扑结构分解为轻量级局部网络拓扑.然后,基于CPU多核心的计算资源,设计深度强化学习机制,异步优化轻量级物联网局部拓扑结构,从而降低网络整体优化运行时间,提高拓扑结构鲁棒能力.在实验方面,与其他先进的优化算法相比,该策略在运行时间方面降低了1~2个数量级,在鲁棒性提升方面,与最优算法相差大约10%.
文摘The effects of cyanidation conditions on gold dissolution were studied by artificial neural network （ANN） modeling. Eighty-five datasets were used to estimate the gold dissolution. Six input parameters, time, solid percentage, P50 of particle, NaCN content in cyanide media, temperature of solution and pH value were used. For selecting the best model, the outputs of models were compared with measured data. A fourth-layer ANN is found to be optimum with architecture of twenty, fifteen, ten and five neurons in the first, second, third and fourth hidden layers, respectively, and one neuron in output layer. The results of artificial neural network show that the square correlation coefficients （R2） of training, testing and validating data achieve 0.999 1, 0.996 4 and 0.9981, respectively. Sensitivity analysis shows that the highest and lowest effects on the gold dissolution rise from time and pH, respectively It is verified that the predicted values of ANN coincide well with the experimental results.
基金Project(51321065)supported by the Innovative Research Groups of the National Natural Science Foundation of ChinaProject(2013CB035904)supported by the National Basic Research Program of China(973 Program)Project(51439005)supported by the National Natural Science Foundation of China
文摘Accurate 3-D fracture network model for rock mass in dam foundation is of vital importance for stability,grouting and seepage analysis of dam foundation.With the aim of reducing deviation between fracture network model and measured data,a 3-D fracture network dynamic modeling method based on error analysis was proposed.Firstly,errors of four fracture volume density estimation methods(proposed by ODA,KULATILAKE,MAULDON,and SONG)and that of four fracture size estimation methods(proposed by EINSTEIN,SONG and TONON)were respectively compared,and the optimal methods were determined.Additionally,error index representing the deviation between fracture network model and measured data was established with integrated use of fractal dimension and relative absolute error(RAE).On this basis,the downhill simplex method was used to build the dynamic modeling method,which takes the minimum of error index as objective function and dynamically adjusts the fracture density and size parameters to correct the error index.Finally,the 3-D fracture network model could be obtained which meets the requirements.The proposed method was applied for 3-D fractures simulation in Miao Wei hydropower project in China for feasibility verification and the error index reduced from 2.618 to 0.337.
