检索结果-内蒙古大学图书馆

A Distributed Computing Framework Based on Lightweight Variance Reduction Method to Accelerate Machine Learning Training on Blockchain

引用

China Communications 2020年第9期17卷 77-89页

作者： Zhen Huang Feng Liu Mingxing Tang Jinyan Qiu Yuxing Peng Science and Technology on Parallel and Distributed Laboratory National University of Defense TechnologyChangsha 410000China H.R.Support Center PLABeijing 100000China

To security support large-scale intelligent applications,distributed machine learning based on blockchain is an intuitive solution ***,the distributed machine learning is difficult to train due to that the corresponding optimization solver algorithms converge slowly,which highly demand on computing and memory *** overcome the challenges,we propose a distributed computing framework for L-BFGS optimization algorithm based on variance reduction method,which is a lightweight,few additional cost and parallelized scheme for the model training *** validate the claims,we have conducted several experiments on multiple classical *** show that our proposed computing framework can steadily accelerate the training process of solver in either local mode or distributed mode.

关键词： machine learning optimization algorithm blockchain distributed computing variance reduction

来源：评论

学校读者我要写书评

暂无评论

A method for hybrid extraction of single-diode model parameters of photovoltaics

引用

RENEWABLE ENERGY 2020年 158卷 236-252页

作者： Arabshahi, M. R. Torkaman, H. Keyhani, A. Shahid Beheshti Univ Fac Elect Engn Tehran Iran Ohio State Univ Dept Elect & Comp Engn Columbus OH 43210 USA

This paper presents a hybrid method to extract the five unknown parameters of a single-diode photo-voltaic (PV) model. The proposed method is a combination of analytical and optimization algorithms such that only two parameters, namely, series (R-s) and shunt resistance (R-sh), are estimated by using metaheuristic algorithms. The information of three major key points in datasheets provided by manufacturers is used for optimization and the sum of the squared error formulated. The rest of the unknown parameters, diode ideality factor (D), photo-generated current (I-ph), and dark saturation current (I-0) are obtained analytically. In order to predict the behavior of real performance characteristics of solar PV modules under different environmental conditions, a set of translational formulas have been used. Finally, performance indices, such as PV characteristics, absolute error in current, normalized root mean square error (nRMSE), maximum power, and relative maximum power error are estimated for six different types of PV modules from different technology to reveal the effectiveness of the proposed method. Through comparison with experimental data available in modules datasheet and existing method, it was found that the proposed method is sufficiently accurate. (C) 2020 Elsevier Ltd. All rights reserved.

关键词： Parameter estimation Photovoltaic (PV) module Single-diode model Current-Voltage (I - V) characteristic optimization algorithm

来源：评论

学校读者我要写书评

暂无评论

A novel hybrid algorithm for rescheduling-based congestion management scheme in power system

引用

ELECTRICAL ENGINEERING 2020年第4期102卷 1993-2010页

作者： Srivastava, Jyoti Yadav, Naresh Kumar Sharma, Arvind Kumar Deenbandhu Chhotu Ram Univ Sci & Technol Sonepat Haryana India KIET Grp Inst Ghaziabad UP India

Secure and continuous power flow in the transmission line is one of the critical issues that must be rectified. In fact, rescheduling-based congestion management is considered to be one of the promising solutions for this aspect. Still, the model faces issues on the basis of rescheduling costs. More research works have been addressed so far to solve the problems of congestion management. optimization algorithms also play a vital role in solving this problem. Under this scenario, this paper introduces a new rescheduling-based congestion management model that incorporates a new algorithm, refractor update-based ROA (RU-ROA) that optimizes the generating power of added generators with the bus system. The proposed RU-ROA algorithm is the hybridization of two algorithms, namely rider optimization algorithm (ROA) and water wave optimization (WWO), that aims to manage the congestion with the reduced cost of rescheduling. Further, the proposed model compares its performance over other conventional models like particle swarm optimization, FireFly, grey wolf optimization, traditional ROA and traditional WWO-based rescheduling strategy with respect to cost analysis and convergence analysis, and proves the efficiency of proposed work over others.

关键词： Congestion management Cost function optimization algorithm ROA Rescheduling strategy

来源：评论

学校读者我要写书评

暂无评论

Time -frequency analysis of a new aperiodic resonance

引用

COMMUNICATIONS IN NONLINEAR SCIENCE AND NUMERICAL SIMULATION 2020年 85卷

作者： Yang, Jianhua Zhang, Shuai Sanjuan, Miguel A. F. Liu, Houguang China Univ Min & Technol Sch Mechatron Engn Xuzhou 221116 Jiangsu Peoples R China China Univ Min & Technol Jiangsu Key Lab Mine Mech & Elect Equipment Xuzhou 221116 Jiangsu Peoples R China Univ Rey Juan Carlos Nonlinear Dynam Chaos & Complex Syst Grp Dept Fis Tulipan S-N Madrid 28933 Spain Kaunas Univ Technol Dept Appl Informat Studentu 50-407 LT-51368 Kaunas Lithuania

Based on the time-frequency analysis, a piecewise re-scaled method is proposed to realize aperiodic resonance in a Duffing system excited by the nonlinear frequency modulated (NLFM) signal. Based on the aperiodic resonance theory, the weak NLFM signal is enhanced greatly. By short time Fourier transform, numerical simulations are carried out for several kinds of NLFM signal. The results show that the method enhances the NLFM signal effectively. Meanwhile, the effectiveness of the method is still illustrated in the noise background. In addition, the noise and the interference frequency can be removed. Noteworthy, and differently as what happens with the stochastic resonance and vibrational resonance, the aperiodic resonance does not require any auxiliary signal or noise to induce it. This constitutes also a new result of this paper. Next, in order to expand the application of this method, it is used to process the experiment signal of bearing fault under variable speed condition. The validity of the method is illustrated again. The results provide new reference in processing non-stationary frequency-modulated signal. Finally, an adaptive piecewise re-scaled aperiodic resonance scheme is put forward to get optimal parameters to induce stronger aperiodic resonance quickly.

关键词： Aperiodic resonance optimization algorithm Nonlinear frequency modulated signal Signal enhancement

来源：评论

学校读者我要写书评

暂无评论

Robust Trajectory Design for Rendezvous and Proximity Operations with Uncertainties

引用

JOURNAL OF GUIDANCE CONTROL AND DYNAMICS 2020年第4期43卷 741-753页

作者： Jin, Kai Geller, David K. Luo, Jianjun Northwestern Polytech Univ Natl Key Lab Aerosp Flight Dynam Xian 710072 Peoples R China Utah State Univ Mech & Aerosp Engn Dept Logan UT 84322 USA Utah State Univ Logan UT 84322 USA

A robust trajectory optimization approach for rendezvous and proximity operations in perturbed elliptical orbits with uncertainties is proposed. For robust trajectory design, a new discrete-time linear dynamics model describing the relative equations of motion for powered flight in perturbed elliptical orbits is developed. The linear dynamics model is used to formulate a stochastic trajectory optimization problem that takes into account navigation sensor errors, maneuver execution errors, and trajectory and control dispersion. The objective of the stochastic optimization problem is to minimize the sum of the expected maneuver and the maneuver dispersion subject to a 3-sigma constraint on the final state dispersion. A genetic algorithm is used to solve the stochastic trajectory optimization problem, and a series of examples demonstrate that the stochastic maneuver solution can be significantly better than the solution corresponding deterministic trajectory optimization problem. Nonlinear Monte Carlo simulations are used to validate the results.

关键词： Trajectory optimization Stochastic optimization Optimal Impulsive Maneuvers Navigation Sensor Error Genetic algorithm Kalman Filter Satellites optimization algorithm Targeting algorithms Earth Centered Inertial

来源：评论

学校读者我要写书评

暂无评论

Deep Principal Correlated Auto-Encoders With Application to Imaging and Genomics Data Integration

引用

IEEE ACCESS 2020年 8卷 20093-20107页

作者： Li, Gang Wang, Chao Han, De-Peng Zhang, Yi-Pu Peng, Peng Calhoun, Vince D. Wang, Yu-Ping Changan Univ Sch Elect & Control Engn Xian 710064 Peoples R China Mind Res Network Albuquerque NM 87106 USA Tulane Univ Dept Biomed Engn New Orleans LA 70118 USA

In terms of complex diseases like schizophrenia, more and more studies are beginning to treat genetic variants and brain imaging phenotypes as an important factor. In this paper, a competent optimization model is exploited to overcome the weakness of deep canonical correlation analysis (DCCA). The model consists of principal component analysis (PCA) on the multi-modality linear features learning and multi-layer belief networks on multi-modality nonlinear features learning. In order to complete a better result of correlation analysis and classification, the output nodes of multi-layer belief network are used for back propagation (BP) network training. Previous works on solving canonical correlation analysis (CCA) had proposed several models based on deep neural network or regularization, typically involving either some form of norm or auto-encoders with a reconstruction objective. Many existing advanced models had been developed to find the maximal correlation in multi-modality data. However, these multi-modality data tend to have the number of feature dimensions which more than that of samples. Differ from these advanced models, our proposed model is applied to analyze the real set of multi-modality data and test several previous models, then comparing them experimentally on fMRI imaging and SNPs genomics. In experiments, the results show that our model, deep principal correlated auto-encoders (DPCAE), learns features with effectively higher correlation and better performance of classification than those previous models. In terms of classification accuracy, the classification accuracy of the datasets exceeds 90;, but that of the CCA-based model are about 65;, and that of the DNN-based model are about 80;, the classification accuracy of the DPCAE is significantly improved obviously. In the experiment of clustering performance evaluation, the DPCAE further verified its superior classification performance with an average normalized mutual information index of 93.75;and

关键词： Classification data fusion dimensionality reduction belief network optimization algorithm principal component analysis

来源：评论

学校读者我要写书评

暂无评论

Ultra-Short-Term Wind Power Prediction by Salp Swarm algorithm-Based Optimizing Extreme Learning Machine

引用

IEEE ACCESS 2020年 8卷 44470-44484页

作者： Tan, Lian Han, Jing Zhang, Hongtao North China Univ Water Resources & Elect Power Sch Elect Power Coll Zhengzhou 450011 Peoples R China

Wind power generation accounts for an increasing proportion of the power grid, so efficient and accurate real-time wind power prediction is particularly important for wind power grid. In view of the strong randomness and fluctuation of wind and the difficulty of predicting wind power, a Salp Swarm algorithms-Extremely Learning Machine (SSA-ELM) based ultra-short-term wind power prediction model is proposed. In this case, the multi-input sample set is composed of historical wind speed, temperature, wind direction, atmospheric pressure and other climatic factors that are highly correlated with wind power, and the network parameters are determined in the training process. In order to improve the adaptability and accuracy of the prediction model, the input weight matrix and hidden layer deviation of the Extreme Learning Machine (ELM) are optimized by exploring and developing the Salp Swarm algorithm in the iterative process. Finally, the simulation experiment is conducted with the actual data of a wind farm in Henan Province, and the comparison with the traditional Extreme Learning Machine, Particle Swarm optimization Extreme Learning Machine (PSO-ELM) and Back Propagation (BP) neural network model shows that the new method avoids falling into the local extreme value, and has faster convergence speed and higher prediction accuracy.

关键词： Wind power generation Prediction algorithms Neurons Predictive models Wind speed Training Wind farms Wind power prediction salp swarm algorithms extreme learning machine optimization algorithm

来源：评论

学校读者我要写书评

暂无评论

The relevance of preferential flow in catchment scale simulations: Calibrating a 3D dual-permeability model using DREAM

引用

HYDROLOGICAL PROCESSES 2020年第5期34卷 1237-1254页

作者： Hopp, Luisa Glaser, Barbara Klaus, Julian Schramm, Thilo Univ Bayreuth Dept Hydrol D-95440 Bayreuth Germany LIST Catchment & Ecohydrol Res Grp L-4362 Esch Sur Alzette Luxembourg

The occurrence of preferential flow in the subsurface has often been shown in field experiments. However, preferential flow is rarely included in models simulating the hydrological response at the catchment scale. If it is considered, preferential flow parameters are typically determined at the plot scale and then transferred to larger-scale simulations. Here, we successfully used the optimization algorithm DiffeRential Evolution Adaptive Metropolis (DREAM) to calibrate a 3D physics-based dual-permeability model directly at the catchment scale. In order to keep computational costs of the optimization routine at a reasonable level, we limited the number of parameters to be calibrated to the ones that had been shown before to be most influential for the simulation of discharge. We also calibrated parameters of the matrix domain and the macropore domain with a fixed parameter ratio between soil layers instead of calibrating every layer separately. These ratios reflected observed depth profiles of soil hydraulic properties at our study site. The dual-permeability parameter sets identified during calibration were able to simulate observed discharge time series satisfactorily but did not outperform a calibrated single-domain reference model scenario. Saturated hydraulic conductivities of the macropore domain were calibrated such that they became very similar to matrix saturated hydraulic conductivities, thereby effectively removing the effect of macropores. This suggests that the incorporation of vertical preferential flow as represented by the dual-permeability approach was not relevant for reproducing the hydrometric response reasonably well in the studied catchment. We also tested the scale-invariance of the calibrated dual-permeability parameter sets by using the parameter sets performing best at catchment scale to simulate plot-scale bromide depth profiles obtained from tracer irrigation experiments. This parameter transfer proved to be not successful, indicating tha

关键词： catchment scale discharge simulation dual-permeability model HydroGeoSphere model calibration optimization algorithm preferential flow

来源：评论

学校读者我要写书评

暂无评论

Synthetic repurposing of drugs against hypertension: a datamining method based on association rules and a novel discrete algorithm

引用

BMC BIOINFORMATICS 2020年第1期21卷 1-21页

作者： Masoudi-Sobhanzadeh, Yosef Masoudi-Nejad, Ali Univ Tehran Inst Biochem & Biophys Lab Syst Biol & Bioinformat LBB Tehran Iran

BackgroundDrug repurposing aims to detect the new therapeutic benefits of the existing drugs and reduce the spent time and cost of the drug development projects. The synthetic repurposing of drugs may prove to be more useful than the single repurposing in terms of reducing toxicity and enhancing efficacy. However, the researchers have not given it serious consideration. To address the issue, a novel datamining method is introduced and applied to repositioning of drugs for hypertension (HT) which is a serious medical condition and needs some improved treatment plans to help treat *** novel two-step data mining method, which is based on the If-Then association rules as well as a novel discrete optimization algorithm, was introduced and applied to the synthetic repurposing of drugs for HT. The required data were also extracted from DrugBank, KEGG, and DrugR+ databases. The findings indicated that based on the different statistical criteria, the proposed method outperformed the other state-of-the-art approaches. In contrast to the previously proposed methods which had failed to discover a list on some datasets, our method could find a combination list for all of *** the proposed synthetic method uses medications in small dosages, it might revive some failed drug development projects and put forward a suitable plan for treating different diseases such as COVID-19 and HT. It is also worth noting that applying efficient computational methods helps to produce better results.

关键词： Data mining Drug repurposing Hypertension optimization algorithm Synthetic repurposing

来源：评论

学校读者我要写书评

暂无评论

A multi-strategy fusion artificial bee colony algorithm with small population

引用

EXPERT SYSTEMS WITH APPLICATIONS 2020年 142卷

作者： Song, Xiaoyu Zhao, Ming Xing, Shuangyun Shenyang Jianzhu Univ Informat & Control Engn Fac Shenyang 110168 Liaoning Peoples R China Shenyang Jianzhu Univ Sch Sci Shenyang 110168 Liaoning Peoples R China

Although artificial bee colony (ABC) algorithm is more and more popular in solving complex problems, slow convergence rate limits its wide application. ABC with small population can use the limited function evaluation times more efficiently since it can avoid unnecessary searches. However, ABC with small population cannot ensure population diversity, and when the algorithm is weak or unstable, it may fall into local optimum easily. So based on the latest research, we are motivated to propose a stabler and more efficient algorithm design to improve the search ability of ABC with small population by the fusion of multiple search strategies, which used together for the employed bees and the onlooker bees. Firstly we select and design multiple strategies with different search abilities of exploration and exploitation. Secondly, we propose an evolution ratio, which is an indicator to fully reflect the adaptability of the search strategy. Thirdly, we design different fusion methods according to the characteristics of the strategies, in which the search strategy with high exploration is maintained at a certain frequency throughout the whole search process of the employed bees, and the selections of the other two search strategies are adjusted according to evolution ratio adaptively in the employed bee phase and the onlooker bee phase. In the end, a novel algorithm called MFABC is proposed, which can realize efficiently multi-strategy cooperative search according to the requirements of different problems and different search stages. Experimental results on a set of benchmark functions have shown the accuracy, stability, efficiency and convergence rate of MFABC. (C) 2019 Elsevier Ltd. All rights reserved.

关键词： optimization algorithm Artificial bee colony algorithm Multi-strategy fusion Small population Cooperative searching

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：