检索结果-南通市图书馆

Programming nonlinear propagation for efficient optical learning machines

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Advanced Photonics 2024年第1期6卷 73-83页

作者： ilker Oguz Jih-Liang Hsieh Niyazi Ulas Dinc Ugur Tegin Mustafa yildirim Carlo Gigli Christophe Moser Demetri Psaltis École Polytechnique Fédérale de Lausanne Institute of Electrical and Micro EngineeringEcublensSwitzerland Koc University Department of Electrical and Electronics EngineeringIstanbulTurkey

The ever-increasing demand for training and inferring with larger machine-learning models requires more efficient hardware solutions due to limitations such as power dissipation and *** is a promising contender for providing lower power computation,since light propagation through a nonabsorbing medium is a lossless ***,to carry out useful and efficient computations with light,generating and controlling nonlinearity optically is a necessity that is still *** fibers(MMFs)have been shown that they can provide nonlinear effects with microwatts of average power while maintaining parallelism and low *** propose an optical neural network architecture that performs nonlinear optical computation by controlling the propagation of ultrashort pulses in MMF by wavefront *** a surrogate model,optimal sets of parameters are found to program this optical computer for different tasks with minimal utilization of an electronic *** show a remarkable decrease of 97%in the number of model parameters,which leads to an overall 99%digital operation reduction compared to an equivalently performing digital neural *** further demonstrate that a fully optical implementation can also be performed with competitive accuracies.

关键词： neural networks nonlinear optics fiber optics surrogate optimization neuromorphic computing wavefront shaping

Modelling height-diameter relationships in complex tropical rain forest ecosystems using deep learning algorithm

在线全文

学校读者我要写书评

暂无评论

Journal of Forestry Research 2022年第3期33卷 883-898页

作者： Friday Nwabueze Ogana ilker Ercanli Department of Social and Environmental Forestry Faculty of Renewable Natural ResourcesUniversity of IbadanIbadanOyo200284Nigeria Department of Forest Engineering Faculty of ForestryÇankiri Karatekin UniversityÇankiri18200Turkey

Modelling tree height-diameter relationships in complex tropical rain forest ecosystems remains a challenge because of characteristics of multi-species, multi-layers, and indeterminate age composition. Effective modelling of such complex systems required innovative techniques to improve prediction of tree heights for use for aboveground biomass estimations. Therefore, in this study, deep learning algorithm (DLA) models based on artificial intelligence were trained for predicting tree heights in a tropical rain forest of Nigeria. The data consisted of 1736 individual trees representing 116 species, and measured from 52 0.25 ha sample plots. A K-means clustering was used to classify the species into three groups based on height-diameter ratios. The DLA models were trained for each species-group in which diameter at beast height, quadratic mean diameter and number of trees per ha were used as input variables. Predictions by the DLA models were compared with those developed by nonlinear least squares (NLS) and nonlinear mixed-effects (NLME) using different evaluation statistics and equivalence test. In addition, the predicted heights by the models were used to estimate aboveground biomass. The results showed that the DLA models with 100 neurons in 6 hidden layers, 100 neurons in 9 hidden layers and 100 neurons in 7 hidden layers for groups 1, 2, and 3, respectively, outperformed the NLS and NLME models. The root mean square error for the DLA models ranged from 1.939 to 3.887 m. The results also showed that using height predicted by the DLA models for aboveground biomass estimation brought about more than 30% reduction in error relative to NLS and NLME. Consequently, minimal errors were created in aboveground biomass estimation compared to those of the classical methods.

关键词： Artificial intelligence Height-diameter model Mixed-effects Nonlinear least squares Tropical mixed forest

Innovative deep learning artificial intelligence applications for predicting relationships between individual tree height and diameter at breast height

在线全文

学校读者我要写书评

暂无评论

Forest Ecosystems 2020年第2期7卷 141-158页

作者： ilker Ercanli Cankiri Karatekin University Faculty of Forestry18200ÇankırıTurkey

Background:Deep Learning Algorithms(DLA)have become prominent as an application of Artificial Intelligence(Al)Techniques since *** paper introduces the DLA to predict the relationships between individual tree height(ITH)and the diameter at breast height(DBH).Methods:A set of 2024 pairs of individual height and diameter at breast height measurements,originating from 150 sample plots located in stands of even aged and pure Anatolian Crimean Pine(Pinus nigra *** ***(Lamb.)Holmboe)in Konya Forest *** present study primarily investigated the capability and usability of DLA models for predicting the relationships between the ITH and the DBH sampled from some stands with different growth *** 80 different DLA models,which involve different the alternatives for the numbers of hidden layers and neuron,have been trained and compared to determine optimum and best predictive DLAs network ***:It was determined that the DLA model with 9 layers and 100 neurons has been the best predictive network model compared as those by other different DLA,Artificial Neural Network,Nonlinear Regression and Nonlinear Mixed Effect *** alternative of 100#neurons and 9#hidden layers in deep learning algorithms resulted in best predictive ITH values with root mean squared error(RMSE,0.5575),percent of the root mean squared error(RMSE%,4.9504%),Akaike information criterion(AIC,-998.9540),Bayesian information criterion(BIC,884.6591),fit index(Fl,0.9436),average absolute error(AAE,0.4077),maximum absolute error(***,2.5106),Bias(0.0057)and percent Bias(Bias%,0.0502%).In addition,these predictive results with DLAs were further validated by the Equivalence tests that showed the DLA models successfully predicted the tree height in the independent ***:This study has emphasized the capability of the DLA models,novel artificial intelligence technique,for predicting the relationships between individual tree height and the diame

关键词： Artificial intelligence Prediction Deep learning algorithms Individual tree height

Effect of reductant sodium bentonite content and reaction temperature in sponge iron production from composite pellets

在线全文

学校读者我要写书评

暂无评论

Journal of Central South University 2021年第5期28卷 1324-1332页

作者： ilker KARA Department of Medical Services and Techniques Eldivan Medical Services Vocational SchoolCankiri Karatekin UniversityCankiri 18100Turkey

In recent years,composite pellet production with added reductant has been developed instead of traditional iron *** pellets produced by the addition of appropriate proportions of reductant produce sponge iron in the reductant melting process at high *** elements created in the structure by pellet production directly affect the quality of the product obtained by determining the chemical composition and the appropriate reaction *** this study,sponge iron ore concentrate(scale)and reductant(coke coal dust and sodium bentonite)were mixed at certain proportions to produce composite pellet samples;the effects of addition rate of the reductant material of sodium bentonite(1 wt%−4 wt%)and variation in reaction temperature(900−1200℃)on the metallization and compressive strength properties of the produced composite pellet samples were *** analysis results show that the highest compressive strength is obtained from pellet samples produced with 3%sodium bentonite at 1100℃.Additionally,SEM-EDS analysis results of the samples show that the morphologic structure has much lower porosity rates compared to samples produced under the other conditions which makes the samples denser and increases the metallization properties.

关键词： solid waste reduction iron oxides pellets sponge iron

Investigation of Heat Transfer Augmentation between the Ribbed Plates via Taguchi Approach and Computational Fluid Dynamics

在线全文

学校读者我要写书评

暂无评论

Journal of Thermal Science 2020年第3期29卷 647-666页

作者： GOKTEPELI ilker ATMACA Ulas CAKAN Abdullah Department of Mechanical Engineering Faculty of Engineering and Natural SciencesKonya Technical UniversityKonya 42075Turkey

Heat transfer and turbulent flow characteristics between the ribbed plates have been numerically studied in the present *** ribs with the rectangular cross-section have been placed on the top and the bottom plates of the duct,*** is assumed that the fluid at 300 K has entered the system while the walls kept at 400 *** numerical analyses have been performed by k-ωShear Stress Transport(SST)turbulence model for Re=10000,15000 and *** the fixed rib width,the dimensionless height and the dimensionless spacing have been respectively varied as 0.1≤h’≤0.3 and 0.5≤S’≤1,and the results have been compared with the ones of the smooth *** though there are twenty-seven cases obtained as a result of parametric combinations,the number of various cases has been reduced from twenty-seven to only nine different variations by applying the Taguchi ***,the effects of all the considered parameters on the heat transfer and flow characteristics have been determined in terms of the influence *** optimum parameters for Nusselt number and pressure loss have been ascertained *** is more,the almost exact values for Nusselt number and pressure loss have also been attained by the confirmation test having an error percentage of 6%.The most dominant factor has been determined as the rib height due to its effect on both heat transfer and flow ***,with respect to the numerical results,increasing the rib height,the rib spacing and Reynolds number has separately increased Nusselt ***,symmetrical flow structure has been disturbed as a result of ascending the rib height as clearly seen for h’=0.3 from the *** the meantime,pressure loss has been augmented owing to the increment of the geometrical parameters and also Reynolds number.h’=0.1 with S’=0.5 at Re=10000,h’=0.2 with S’=0.5 at Re=15000 and h’=0.1 with S’=0.75 at Re=15000 can be suggested for the heat transfer enhancement since the press

关键词： CFD heat transfer Nusselt number parallel plate Reynolds number rib Taguchi approach turbulent flow

Experimental analysis on the nonlinear behavior of DC barrier discharge plasmas

在线全文

学校读者我要写书评

暂无评论

Plasma Science and Technology 2017年第1期19卷 44-50页

作者： Dogan MANSUROGLU ilker Umit UZUN-KAyMAK Physics Department CanakkaleOnsekiz Mart University Canakkale 17100 Turkey Physics Department Middle East Technical University Ankara 06800 Turkey

Nonlinear behavior of glow discharge plasmas is experimentally *** glow is generated between a barrier semiconductor electrode,Chromium doped namely Gallium Arsenide（Ga As：Cr）,as a cathode and an Indium–Tin Oxide（ITO） coated glass electrode as an anode,in reverse *** planar nature of electrodes provides symmetry in spatial *** discharge behaves oscillatory in the time domain,with single and sometimes multiperiodicities in plasma current and voltage *** this paper,harmonic frequency generation and transition to chaotic behavior is *** observed current–voltage characteristics of the discharge are discussed in detail.

关键词： DC glow discharges nonlinear behavior digital signal analysis

Analysis of EST-SSRs in silver birch(Betula pendula Roth.)

在线全文

学校读者我要写书评

暂无评论

Journal of Forestry Research 2016年第3期27卷 639-646页

作者： Ertugrul Filiz Ilhan Dogan Ibrahim ilker Ozyigit Department of Crop and Animal Production Cilimli Vocational SchoolDuzce University Department of Molecular Biology and Genetics Izmir Institute of Technology Department of Biology Faculty of ScienceRahatlayadururken Manas University Department of Biology Faculty of Science and ArtsMarmara University

Simple sequence repeats（SSRs） defined as sequence repeat units between 1 and 6 bp occur abundantly in both coding and non-coding regions in eukaryotic genomes and these repeats can affect gene expression. In this study, ESTs（expressed sequence tags） of Betula pendula（silver birch） were analyzed for in silico mining of ESTSSRs, protein annotation, open reading frames（ORFs）,designing primers, and identifying codon repetitions. In ***, the frequency of ESTs containing SSRs was 7.8 %with an average of 1SSR/4. 78 kb of EST sequences. A total of 188 SSRs was identified by using MISA software and dinucleotide SSR motifs（65.9 %） were found to be the most abundant type of repeat motif followed by tri-（27.1 %）,tetra-（4.8 %）, and penta-（2.2 %） motifs. Based on ORF analysis, 175 of 178 sequences were predicted as ORFs and the most frequent SSRs were detected in 50 UTR（58.43 %）,followed by in ORF（31.46 %） and in 30UTR（8.43 %）. 102 of 178 ESTs were annotated as ribosomal protein, transport protein, membrane protein, carrier protein, binding protein,and transferase protein. For a total of 102 SSRs（57.3 %）with significant matches, a set of 102 primers（100 %） with forward and reverse strands was designed by using Primer 3 software. Serine（Ser, 19.6 %） was predominant in putative encoded amino acids and most of amino acids showed nonpolar（35.3 %） nature. Our data provide resources for *** and can be useful for in silico comparative analyses of Betulaceae species, including SSR mining.

关键词： Silver birch (Betula pendula) Betulaceae -EST-SSR SSR mining In silico analysis

Weak Vertical Diffusion Allows Maintenance of Cold Halocline in the Central Arctic

在线全文

学校读者我要写书评

暂无评论

Atmospheric and Oceanic Science Letters 2009年第3期2卷 148-152页

作者： ilker FER Geophysical Institute University of Bergen & Bjerknes Centre for Climate ResearchBergen Norway

In spring preceding the record minimum summer ice cover detailed microstructure measurements were made from drifting pack ice in the Arctic Ocean, 110 km from the North Pole. Profiles of hydrography, shear, and temperature microstructure collected in the upper water column covering the core of the Atlantic Water are analyzed to determine the diapycnal eddy diffusivity, the eddy diffusivity for heat, and the turbulent flux of heat. Turbulence in the bulk of the cold halocline layer was not strong enough to generate significant buoyancy flux and mixing. Resulting turbulent heat flux across the upper cold halocline was not significantly different than zero. The results show that the low levels of eddy diffusivity in the upper cold halocline lead to small vertical turbulent transport of heat, thereby allowing the maintenance of the cold halocline in the central Arctic.

关键词： Arctic Ocean cold halocline layer turbulence mixing heat flux Atlantic Water

Artificial neural network models predicting the leaf area index:a case study in pure even-aged Crimean pine forests from Turkey

在线全文

学校读者我要写书评

暂无评论

Forest Ecosystems 2018年第4期5卷 400-411页

作者： ilker Ercanli Alkan Gunlu Muammer Senyurt Sedat Keles Faculty of Forestry Cankin Karatekin University18200 CankiriTurkey

Background: Leaf Area Index(LAI) is an important parameter used in monitoring and modeling of forest ecosystems. The aim of this study was to evaluate performance of the artificial neural network(ANN) models to predict the LAI by comparing the regression analysis models as the classical method in these pure and even-aged Crimean pine forest ***: One hundred eight temporary sample plots were collected from Crimean pine forest stands to estimate stand parameters. Each sample plot was imaged with hemispherical photographs to detect the LAI. The partial correlation analysis was used to assess the relationships between the stand LAI values and stand parameters, and the multivariate linear regression analysis was used to predict the LAI from stand parameters. Different artificial neural network models comprising different number of neuron and transfer functions were trained and used to predict the LAI of forest ***: The correlation coefficients between LAI and stand parameters(stand number of trees, basal area, the quadratic mean diameter, stand density and stand age) were significant at the level of 0.01. The stand age, number of trees, site index, and basal area were independent parameters in the most successful regression model predicted LAI values using stand parameters(R_(adj)~2=0.5431). As corresponding method to predict the interactions between the stand LAI values and stand parameters, the neural network architecture based on the RBF 4-19-1 with Gaussian activation function in hidden layer and the identity activation function in output layer performed better in predicting LAI(SSE(12.1040), MSE(0.1223), RMSE(0.3497), AIC(0.1040), BIC(-77.7310) and R^2(0.6392)) compared to the other studied ***: The ANN outperformed the multivariate regression techniques in predicting LAI from stand parameters. The ANN models, developed in this study, may aid in making forest management planning in study forest stands.

关键词： Leaf area index Multivariate linear regression model Artificial neural network modeling Crimean pine Stand parameters