期刊文献+
共找到699篇文章
< 1 2 35 >
每页显示 20 50 100
Carbon and Nitrogen Transformations in Surface Soils Under Ermans Birch and Dark Coniferous Forests 被引量:5
1
作者 DENG Xiao-Wen HAN Shi-Jie +1 位作者 HU Yan-Ling ZHOU Yu-Mei 《Pedosphere》 SCIE CAS CSCD 2009年第2期230-237,共8页
Soil samples were taken from an Ermans birch (Betula ermanii)-dark coniferous forest (Picea jezoensis and Abies nephrolepis) ecotone growing on volcanic ejecta in the northern slope of Changbai Mountains of Northe... Soil samples were taken from an Ermans birch (Betula ermanii)-dark coniferous forest (Picea jezoensis and Abies nephrolepis) ecotone growing on volcanic ejecta in the northern slope of Changbai Mountains of Northeast China, to compare soil carbon (C) and nitrogen (N) transformations in the two forests. The soil type is Umbri-Gelic Cambosols in Chinese Soil Taxonomy. Soil samples were incubated aerobically at 20℃ and field capacity of 700 g kg^-1 over a period of 27 weeks. The amount of soil microbial biomass and net N mineralization were higher in the Ermans birch than the dark coniferous forest (P 〈 0.05), whereas the cumulative C mineralization (as CO2 emission) in the dark coniferous forest exceeded that in the Ermans birch (P 〈 0.05). Release of the cumulative dissolved organic C and dissolved organic N were greater in the Ermans birch than the dark coniferous forest (P 〈 0.05). The results suggested that differences of forest types could result in considerable change in soil C and N transformations. 展开更多
关键词 dissolved organic C dissolved organic N Ermans birch-dark coniferous forest soil C transformation soil N transformation
下载PDF
Effects of six years of simulated N deposition on gross soil N transformation rates in an old-growth temperate forest 被引量:3
2
作者 Peng Tian Jinbo Zhang +2 位作者 Christoph Müller Zucong Cai Guangze Jin 《Journal of Forestry Research》 SCIE CAS CSCD 2018年第3期644-653,共10页
Elevated atmospheric nitrogen(N) deposition has been detected in many regions of China, but its effects on soil N transformation in temperate forest ecosystems are not well known. We therefore simulated N deposition w... Elevated atmospheric nitrogen(N) deposition has been detected in many regions of China, but its effects on soil N transformation in temperate forest ecosystems are not well known. We therefore simulated N deposition with four levels of N addition rate(N0, N30, N60, and N120) for6 years in an old-growth temperate forest in Xiaoxing’an Mountains in Northeastern China. We measured gross N transformation rates in the laboratory usingN tracing technology to explore the effects of N deposition on soil gross N transformations taking advantage of N deposition soils. No significant differences in gross soil N transformation rates were observed after 6 years of N deposition with various levels of N addition rate. For all N deposition soils, the gross NH~+ immobilization rates were consistently lower than the gross N mineralization rates,leading to net N mineralization. Nitrate(NO~-) was primarily produced via oxidation of NH~+(i.e., autotrophic nitrification), whereas oxidation of organic N(i.e., heterotrophic nitrification) was negligible. Differences between the quantity of ammonia-oxidizing bacteria and ammonia-oxidizing archaea were not significant for any treatment, which likely explains the lack of a significant effect on gross nitrification rates. Gross nitrification rates were much higher than the total NO~- consumption rates,resulting in a build-up of NO~-, which highlights the high risk of N losses via NO~- leaching or gaseous N emissions from soils. This response is opposite that of typical N-limited temperate forests suffering from N deposition,suggesting that the investigated old-growth temperate forest ecosystem is likely to approach N saturation. 展开更多
关键词 N deposition Gross soil N transformation Temperate forest ecosystem ^(15)N tracing technology
下载PDF
The influence of selective cutting of mixed Korean pine(Pinus koraiensis Sieb.et Zucc.) and broad-leaf forest on rare species distribution patterns and spatial correlation in Northeast China 被引量:4
3
作者 Binbin Kan Qingcheng Wang Wenjuan Wu 《Journal of Forestry Research》 SCIE CAS CSCD 2015年第4期833-840,共8页
This study aimed to demonstrate change in spatial correlation between Korean pine (Pinus koraiensis Sieb. et Zucc.) and three rare species, and change in spatial distribution of four species in response to a range o... This study aimed to demonstrate change in spatial correlation between Korean pine (Pinus koraiensis Sieb. et Zucc.) and three rare species, and change in spatial distribution of four species in response to a range of selective cutting intensities. We sampled three plots of mixed Korean pine and broad-leaf forest in Lushuihe Forestry Bureau of Jilin province, China. Plot 1, a control, was unlogged Korean pine broad-leaf forest. In plots 2 and 3, Korean pine was selectively cut at 15 and 30 % intensity, respectively, in the 1970s. Other species were rarely cut. We used point-pattern analysis to research the spatial distributions of four tree species and quantify spatial correlations between Korean pine and the other three species, Amur linden (Tilia amurensis Rupr.), Manchurian ash (Fraxinus mandshurica Rupr.), and Mongolian oak (Quercus mongolica Fisch.) in all three plots. The results of the study show that selective cutting at 15 % intensity did not significantly change either the species spatial patterns or the spatial correlation between Korean pine and broadleaf species. Selective cutting at 30 % intensity slightly affected the growth of Korean pine and valuable species in forest communities, and the effect was considered nondestructive and recoverable. 展开更多
关键词 Korean pine broad-leaf forest Cuttingintensity Rare species Spatial pattern - Spatialcorrelation
下载PDF
Power Transformer Fault Diagnosis Using Random Forest and Optimized Kernel Extreme Learning Machine 被引量:1
4
作者 Tusongjiang Kari Zhiyang He +3 位作者 Aisikaer Rouzi Ziwei Zhang Xiaojing Ma Lin Du 《Intelligent Automation & Soft Computing》 SCIE 2023年第7期691-705,共15页
Power transformer is one of the most crucial devices in power grid.It is significant to determine incipient faults of power transformers fast and accurately.Input features play critical roles in fault diagnosis accura... Power transformer is one of the most crucial devices in power grid.It is significant to determine incipient faults of power transformers fast and accurately.Input features play critical roles in fault diagnosis accuracy.In order to further improve the fault diagnosis performance of power trans-formers,a random forest feature selection method coupled with optimized kernel extreme learning machine is presented in this study.Firstly,the random forest feature selection approach is adopted to rank 42 related input features derived from gas concentration,gas ratio and energy-weighted dissolved gas analysis.Afterwards,a kernel extreme learning machine tuned by the Aquila optimization algorithm is implemented to adjust crucial parameters and select the optimal feature subsets.The diagnosis accuracy is used to assess the fault diagnosis capability of concerned feature subsets.Finally,the optimal feature subsets are applied to establish fault diagnosis model.According to the experimental results based on two public datasets and comparison with 5 conventional approaches,it can be seen that the average accuracy of the pro-posed method is up to 94.5%,which is superior to that of other conventional approaches.Fault diagnosis performances verify that the optimum feature subset obtained by the presented method can dramatically improve power transformers fault diagnosis accuracy. 展开更多
关键词 Power transformer fault diagnosis kernel extreme learning machine aquila optimization random forest
下载PDF
THE POSSIBILITY ANALYSIS OF NATURAL POPLAR-BIRCH FOREST RECOVERY TO BROAD-LEAF KOREAN PINE FOREST THROUGH THE METHOD OF LIFE TABLE IN THE XIOAXING'AN MOUNTAINS
5
作者 王树力 池玉杰 《Journal of Northeast Forestry University》 SCIE CAS CSCD 1995年第3期57-60,共4页
The basic principle of life table method is deseribed, and the method of tree height instead of tree age in static life table is suggested, and it is also discussed that the possibility of natural poplar -birch forest... The basic principle of life table method is deseribed, and the method of tree height instead of tree age in static life table is suggested, and it is also discussed that the possibility of natural poplar -birch forest recover to broad-leaf Korean pine forest on low pitches in the Xiaoxing'an Mountains by this method. If there is no particular situation, Korean pines after high than 5m under natural Poplar-birch forest will basically survive and make their way into dominant callopy accompanied by climax broad-leaf species. 展开更多
关键词 Life table Natural poplar-birch forest broad-leaf Korean pine forest
下载PDF
Woody Debris Storage of Evergreen Broad-leaf Forest
6
作者 Zeng Zhangquan Wang Silong +2 位作者 Zhang Canming Wu Zijian Li Xiquan 《Meteorological and Environmental Research》 CAS 2014年第8期16-19,共4页
In Yingzuijie National Nature Reserve, Pinus massoniana forest, mixed broadleaf-coniferous forest and evergreen broad-leaf forest were investigated to study the changing characteristics of woody debris (WD) during v... In Yingzuijie National Nature Reserve, Pinus massoniana forest, mixed broadleaf-coniferous forest and evergreen broad-leaf forest were investigated to study the changing characteristics of woody debris (WD) during various succession stages o1 evergreen broad-leaf forest. The results showed that during various succession stages of evergreen broad-leaf forest in Yingzuijie National Nature Reserve, WD storage of each forest ranged from 1.26 to 8.82 t/hm^2, with the order of P. massoniana forest 〈 mixed broadleaf-coniferous forest 〈 evergreen broad-leaf forest, that is, it increased from early to late stages of the succession. At different succession stages, coarse woody debris (CWD) storage was 2 -9 times more than fine woody debris (FWD) storage, revealing that CWD was dominant in WD of each forest. CWD biomass accounted for 0.66% -2.21% of arbor biomass, so the forests were at the developmental stage. 展开更多
关键词 Woody debris Evergreen broad-leaf forest Middle subtropics Succession stages China
下载PDF
基于改进YOLOv5s的CNN-Swin Transformer森林野生动物图像目标检测算法
7
作者 杨文翰 刘天宇 +2 位作者 周俊池 胡文武 蒋蘋 《林业科学》 EI CAS CSCD 北大核心 2024年第3期121-130,共10页
【目的】为提高野生动物在复杂森林环境中的检测精度,促进森林野生动物保护技术发展,提出一种基于YOLOv5s网络模型、针对陷阱相机所摄取森林野生动物图像的改进检测算法。【方法】以包含湖南壶瓶山国家级自然保护区几种典型森林野生动... 【目的】为提高野生动物在复杂森林环境中的检测精度,促进森林野生动物保护技术发展,提出一种基于YOLOv5s网络模型、针对陷阱相机所摄取森林野生动物图像的改进检测算法。【方法】以包含湖南壶瓶山国家级自然保护区几种典型森林野生动物在内的数据集为研究对象,首先,对真实标注框图像进行裁剪、归一化和缩放处理,随机将2~4张裁剪图像拼贴组成新的数据集元素,以丰富和增强数据集图像信息;其次,使用一种基于通道注意力思想的加权通道拼接方法,在通道拼接时引入权重改变通道数量,通过反向传播训练方法不断更新权重以增加重要特征信息的通道层数;接着,引入Swin Transformer模块与CNN网络相结合,为卷积神经网络特征提取加入自注意力机制,融合2种网络特征提取层的优势,提高特征提取的感受野;最后,选择更优的α-DIoU损失函数替代GIoU损失函数,针对边界框重叠面积和中心点距离造成的损失,引入新的几何因素惩罚项。【结果】在相同试验条件和数据集下,相比原YOLOv5s网络模型,改进算法极大提高检测的平均准确率和平均回归率,均值平均精度由74.1%提升至88.4%,获得14.3%的精度提升,同时也超过YOLOv3、YOLOXs、RetinaNet、Faster R-CNN等其他流行目标检测算法。【结论】针对陷阱相机所摄取森林野生动物图像背景与目标对比度低、遮挡重叠严重,致使检测误检率、漏检率高等问题,在检测算法中提出一系列改进措施,为我国森林野生动物的保护和数据获取提供一种新的可行性方案和思路。 展开更多
关键词 森林野生动物 检测算法 YOLOv5s Swin transformer 网络融合
下载PDF
基于Transformer改进YOLOv5的山火检测方法研究 被引量:4
8
作者 钱承山 沈有为 +1 位作者 孙宁 戴仁天 《电子测量技术》 北大核心 2023年第16期46-56,共11页
构建智慧生态林业中核心环节为对森林火灾的监测及防范,为了第一时间扑灭火源防止火势蔓延并将可能发生的山火隐患于第一时间消除,提出了两种适用于无人机高空巡检的森林火灾检测模型YOLO_MC与YOLO_MCLite。其中YOLO_MC可对标准图像中... 构建智慧生态林业中核心环节为对森林火灾的监测及防范,为了第一时间扑灭火源防止火势蔓延并将可能发生的山火隐患于第一时间消除,提出了两种适用于无人机高空巡检的森林火灾检测模型YOLO_MC与YOLO_MCLite。其中YOLO_MC可对标准图像中的明火及烟雾进行检测,并基于YOLO_MC模型进行轻量化设计,提出适用于热图像中高温区域的检测模型YOLO_MCLite。在网络结构的设计中,首先在常规的卷积神经网络中融合加入了Transformer模型,提升了主干网络对于全局特征信息的感知能力;同时对Transformer模型进行轻量化设计,首先在网络结构上通过分组计算的形式减少tokens数量以降低计算量,其次通过通道注意力机制对特征块的通道数进行去冗余并提权来减少tokens的维度参数以降低计算复杂度,并且采用蒸馏算法从所设计的网络中提取出超轻量化网络应用于无人机红外影像的森林高温点检测,以预防森林火灾的发生。经过实验得出以下数据:所设计的两个检测模型中,其中适用于标准图像中对于明火及烟雾的检测准确率可达94.8%,适用于热图像对高温点的检测准确率可达97.2%,并且在英伟达JETSON TX2嵌入式设备上测试两个模型的帧率分别达到了22.5以及32.4。实验结果表明本文所设计网络能够对森林火灾进行有效检测并可以通过检测高温点及时预防火灾发生。 展开更多
关键词 目标检测 森林火灾 神经网络 transformer模型 蒸馏学习
下载PDF
基于改进Swin Transformer的森林火灾检测算法 被引量:12
9
作者 叶铭亮 周慧英 李建军 《中南林业科技大学学报》 CAS CSCD 北大核心 2022年第8期101-110,共10页
【目的】森林火灾常常会对人类的财产和生态多样性造成巨大损害,传统的森林火灾检测技术存在可靠性低、造价过高等不足。目前基于卷积神经网络的深度学习算法在处理图像型数据上具有准确性高、成本低、速度快等优势,但是其处理视觉要素... 【目的】森林火灾常常会对人类的财产和生态多样性造成巨大损害,传统的森林火灾检测技术存在可靠性低、造价过高等不足。目前基于卷积神经网络的深度学习算法在处理图像型数据上具有准确性高、成本低、速度快等优势,但是其处理视觉要素和物体之间关系的能力不如Transformer。因此,本研究提出一种改进Swin Transformer网络的方法应用于森林火灾检测。【方法】Transformer是一种基于自注意机制的深度神经网络,其强大的表现能力使得其能够在计算机视觉领域大放异彩。Swin Transformer提出将Transformer应用于计算机视觉任务,构建了一种名为Swin Transformer Blocks的骨干网络,并且提出了一种滑动窗口多头自注意力机制。本文结合Transformer与深度学习算法并应用于森林火灾检测领域,在Swin Transformer网络结构中对窗口自注意力机制进行改进,采用了knn自注意力提高对小块噪声的识别,使用Augmentation数据增强方法增加模型的泛化能力。【结果】数据集为自建的森林火灾图像数据集,通过旋转、裁剪、模糊以及色彩调节等数据增广的方法将300张不同环境下的森林火灾图像数据扩充到1900张图像,最后对Swin Transformer以及改进后的模型进行对比实验,改进后的算法准确率可达98.1%,bbox_mAP、bbox_mAP_50和bbox_mAP_75分别达到了66.7%、96.4%和81.3%。【结论】本文提出一种改进Swin Transformer应用于森林火灾检测的方法。研究结果表明,改进的Swin Transformer模型能够有效检测不同环境下的森林火灾。 展开更多
关键词 森林火灾 深度神经网络 Swin transformer 目标检测
下载PDF
Clear felling and burning effects on soil nitrogen transforming bacteria and actinomycetes population in Chittagong University campus, Bangladesh 被引量:4
10
作者 S.M. Sirajul Haque Rahima Ferdoshi +1 位作者 Sohag Miah M. Nural Anwar 《Journal of Forestry Research》 SCIE CAS CSCD 2012年第1期123-130,共8页
The effect of forests clear felling and associated burning on the population of soil nitrogen transforming bacteria and actinomycetes are reported at three pair sites of Chittagong University campus, Bangla- desh in m... The effect of forests clear felling and associated burning on the population of soil nitrogen transforming bacteria and actinomycetes are reported at three pair sites of Chittagong University campus, Bangla- desh in monsoon tropical climate. Clear felled area or burnt site and 15-21 year mixed plantation of native and exotic species, situated side by side on low hill having Typic Dystrochrepts soil was represented at each pair site. At all the three pair sites, clear felled area or burnt site showed very significantly (p~_0.001) lower population of actinomycetes, Rhizobium, Nitrosomonas, Nitrobacter and ammonifying as well as deni- trifying bacteria compared to their adjacent mixed plantation. From en- vironmental consideration, this finding has implication in managing natural ecosystem. 展开更多
关键词 Nitrogen transforming bacteria ACTINOMYCETES forest clearfelling effects burning effects soil microorganisms Chittagong HillTracts
下载PDF
Automatic Sleep Staging Algorithm Based on Random Forest and Hidden Markov Model 被引量:3
11
作者 Junbiao Liu Duanpo Wu +4 位作者 Zimeng Wang Xinyu Jin Fang Dong Lurong Jiang Chenyi Cai 《Computer Modeling in Engineering & Sciences》 SCIE EI 2020年第4期401-426,共26页
In the field of medical informatics,sleep staging is a challenging and timeconsuming task undertaken by sleep experts.According to the new standard of the American Academy of Sleep Medicine(AASM),the stages of sleep a... In the field of medical informatics,sleep staging is a challenging and timeconsuming task undertaken by sleep experts.According to the new standard of the American Academy of Sleep Medicine(AASM),the stages of sleep are divided into wakefulness(W),rapid eye movement(REM)and non-rapid eye movement(NREM)which includes three sleep stages(N1,N2 and N3)that describe the depth of sleep.This study aims to establish an automatic sleep staging algorithm based on the improved weighted random forest(WRF)and Hidden Markov Model(HMM)using only the features extracted from double-channel EEG signals.The WRF classification model focuses on reducing the bias of the imbalance data,while the HMM model focuses on improving the detection rate of sleep staging through the relationship between adjacent sleep stages.In particular,the improved weighted RF classification model can increase the recognition rate of the N1 stage.In addition,the method of removing features with low variance is used to select meaningful and contributing feature parameters for model training.This is an innovative content of this paper.The sleep EEG data are first segmented into 30 s epochs,and the feature parameters of the epoch data are extracted from the double-channel by applying continuous wavelet packet transform(WPT).Each epoch is then segmented into 29 subepochs of 2 s long with 1 s overlap,and the frequency domain features and statistical features of each subepoch are extracted.The performance of the proposed method is tested by evaluating the accuracy(AC),Kappa coefficient,Recall(R),Precision(P)and F1-score(F1).In the Sleep-EDF database,the overall AC and Kappa coefficient obtained by WRF are 93.20%and 0.890,respectively using the subject-non-independent test.In the 10 sc*and 10 st*Sleep-EDF Expanded database,the overall AC and Kappa coefficient obtained by proposed method are 91.97%and 0.874,respectively using the subject-independent test.The best AC and Kappa coefficient of single subject can reach 96.3%and 0.912,respectively.Experimental results show that the performance of the proposed method is competitive with the most current methods and results,and the recognition rate of N1 stage is significantly improved. 展开更多
关键词 Sleep staging wavelet packet transform random forest HMM
下载PDF
Determination of forest soil organic nitrogen determination using technique of X-ray absorption near-edge structure 被引量:1
12
作者 ZHUANG Shun-yao XU Meng-jie HU Zheng-yi 《Journal of Forestry Research》 SCIE CAS CSCD 2006年第3期189-192,共4页
The forest soil organic N was investigated using N K-edge X-ray absorption near-edge structure (XANES) to gain an insight into the relationship between N speciation and its transformation in quantity. Soil samples w... The forest soil organic N was investigated using N K-edge X-ray absorption near-edge structure (XANES) to gain an insight into the relationship between N speciation and its transformation in quantity. Soil samples were collected from spruce, hemlock and pine forests in central Taiwan. Results showed that various organic N types could be revealed by XANES spectra. Amide and pyrrolic N are the major parts of the composition in the humic substance, soluble organic nitrogen and original soils. The relative distribution of N speciation differed in treatments and vegetations samples. The spruce had a significant difference from the hemlock in soluble organic nitrogen (SON) spectra at 402.3 eV energy peak. In the A-horizon soils, the relative amount of pyridinic N was much higher than that in the O-horizon soils, indicating N transformation in quantity in the mineral horizon was different from that in the organic horizon, which might play an important role in N cycling of forest ecosystems. 展开更多
关键词 forest soils N K-edge XANES spectra Nitrogen speciation Organic nitrogen Nitrogen transformation
下载PDF
Humus horizon development during natural forest succession process in the Polish Carpathians
13
作者 Justyna SOKOŁOWSKA Agnieszka JÓZEFOWSKA Tomasz ZALESKI 《Journal of Mountain Science》 SCIE CSCD 2022年第3期647-661,共15页
As a result of socio-economic changes and land abandonment,the main ecological driver of the Carpathian landscape is the progression of the natural forest succession process.Thus,aspects of this process have become wo... As a result of socio-economic changes and land abandonment,the main ecological driver of the Carpathian landscape is the progression of the natural forest succession process.Thus,aspects of this process have become worthy of attention,especially in the context of carbon sequestration and the management of protected areas.Soil processes,especially within the topsoil,are some of the most susceptible to change,due to the accumulation of organic matter during such land-use transformations.The purposes of this study were to investigate the differences in topsoil development using the A Horizon Development Index(ADI)and to study the composition of humic substances and advanced organic matter humification in different land-use areas in selected Carpathian national parks,i.e.Bieszczady,Magura and Pieniny National Parks in southern Poland.Additionally,a goal of this study was to compare the ADI and the spectroscopic coefficients of humic substances as indicators of the degree of humus horizon shaping as well as advanced organic matter humification.In total,ten transects were selected,each consisting of three different land-use areas:semi-natural meadow,successional forest and old-growth forest.Soil colour was determined in fresh and air-dried samples using the Munsell colour chart.In air-dried soil samples p H,soil texture,total organic carbon and total nitrogen were measured.Humic substances were extracted and further characterized by UV-VIS spectroscopy.The ADI confirmed the influence of natural forest succession on soil colour darkening and the development of the uppermost soil layer.Spectroscopic analyses of humic substances showed two different patterns depending on soil depth.In the 0–10 cm layer,natural forest succession reduced the rate of the humification process and decreased the degree of maturity of fulvic acids;in the 10–20 cm layer,it led to an increase in the rate of the humification process and a decrease in the content of humic and fulvic acids at the beginning of the transformation.The comparison of two different indicators of soil development–the ADI and the spectroscopic coefficients of humic substances(Q_(4/6),Q_(2/4),Q_(2/3),Δlog K)–indicated that these indexes are based on different features of soil and cannot be used interchangeably. 展开更多
关键词 Land use changes Natural forest succession Polish Carpathians Landscape transformation Protected areas management Humic substances
下载PDF
Anomaly Detection and Pattern Differentiation in Monitoring Data from Power Transformers
14
作者 Jun Zhao Shuguo Gao +4 位作者 Yunpeng Liu QuanWang Ziqiang Xu Yuan Tian Lu Sun 《Energy Engineering》 EI 2022年第5期1811-1828,共18页
Aiming at the problem of abnormal data generated by a power transformer on-line monitoring system due to the influences of transformer operation state change,external environmental interference,communication interrupt... Aiming at the problem of abnormal data generated by a power transformer on-line monitoring system due to the influences of transformer operation state change,external environmental interference,communication interruption,and other factors,a method of anomaly recognition and differentiation for monitoring data was proposed.Firstly,the empirical wavelet transform(EWT)and the autoregressive integrated moving average(ARIMA)model were used for time series modelling of monitoring data to obtain the residual sequence reflecting the anomaly monitoring data value,and then the isolation forest algorithm was used to identify the abnormal information,and the monitoring sequence was segmented according to the recognition results.Secondly,the segmented sequence was symbolised by the improved multi-dimensional SAX vector representation method,and the assessment of the anomaly pattern was made by calculating the similarity score of the adjacent symbol vectors,and the monitoring sequence correlation was further used to verify the assessment.Finally,the case study result shows that the proposed method can reliably recognise abnormal data and accurately distinguish between invalid and valid anomaly patterns. 展开更多
关键词 Abnormal detection empirical wavelet transform autoregressive integrated moving average isolated forest
下载PDF
基于EMD-gcForest模型的变压器油中溶解气体浓度预测方法 被引量:1
15
作者 张鹏坤 余进 +2 位作者 李波 单长吉 张靖 《电力科学与工程》 2023年第6期32-38,共7页
首先,考虑到气体浓度序列的波动性,利用经验模态分解(Empiricalmode decomposition,EMD)方法,将油中溶解气体浓度序列分解为不同特征尺度的本征模态函数(Intrinsic mode function,IMF)分量和1个剩余分量;然后,采用多粒度级联森林(Multi-... 首先,考虑到气体浓度序列的波动性,利用经验模态分解(Empiricalmode decomposition,EMD)方法,将油中溶解气体浓度序列分解为不同特征尺度的本征模态函数(Intrinsic mode function,IMF)分量和1个剩余分量;然后,采用多粒度级联森林(Multi-grained cascadeforest,gcForest)模型对分解得到各子序列分量分别进行预测;最后,叠加所有各子序列分量的预测值作为最终结果。算例分析结果表明,相较传统预测方法,所提的EMD-gcForest方法具有较高的预测精度和泛化能力。 展开更多
关键词 电力变压器 故障诊断 经验模态分解 多粒度级联森林 油中溶解气体
下载PDF
融合YOLOv5s与Swin Transformer的森林火灾检测 被引量:1
16
作者 杨锋勇 王赫 +1 位作者 杨庆江 李芊诺 《高师理科学刊》 2023年第6期37-41,73,共6页
针对传统的森林火灾检测方法检测精度不佳、可靠性低等不足,提出一种基于YOLOv5s融合Swin Transformer的火灾检测方法.为实现森林火灾的实时性检测,提出了YOLOv5s-SwinT的改进识别方法.基于Transformer模型的应用,解决了卷积神经网络的... 针对传统的森林火灾检测方法检测精度不佳、可靠性低等不足,提出一种基于YOLOv5s融合Swin Transformer的火灾检测方法.为实现森林火灾的实时性检测,提出了YOLOv5s-SwinT的改进识别方法.基于Transformer模型的应用,解决了卷积神经网络的运算局部性以及全局特征提取等不足.融合Swin Transformer与YOLOv5s卷积神经网络模型,并将其应用于森林火灾检测的机器视觉任务中.引入α-IoU损失函数替换GIOU损失函数,并在骨干网络中融入CA注意力机制轻量模块,提升了整体网络的特征提取能力以及获取高质量和高精准度的定位图像区域,进行边界框生成及预测,改善小目标检测漏检及检测精度差等问题.实验结果表明,融合YOLOv5s-Swin T的改进识别方法在森林火灾检测任务中,可实现mAP值达到74.2%,相比YOLOv5s提高了4.5%,并设计了GUI界面直接部署到PC端实现实时火灾检测需求,为森林火灾检测视觉任务提供了一种有效的检测方法. 展开更多
关键词 YOLOv5s 森林火灾 Swin transformer 注意力机制
下载PDF
Future Climate Impacts in Woodland and Forest Steppe Based on Holocene Paleoclimatic Trends, Paleobotanical Change in Central Part of the Carpathian Basin (Hungary)
17
作者 Katalin Náfrádi Gusztáv Jakab +2 位作者 Pál Sümegi Zoltán Szelepcsényi Tünde Torocsik 《American Journal of Plant Sciences》 2013年第6期1187-1203,共17页
The Sirok Nyírjes-tó peat bog provides an almost full Holocene climatic record reconstructed by bog surface wetness investigations based on plant macrofossil analysis. We compared our macrofossil data to ant... The Sirok Nyírjes-tó peat bog provides an almost full Holocene climatic record reconstructed by bog surface wetness investigations based on plant macrofossil analysis. We compared our macrofossil data to anthracological material derived from archaeological sites and to the newest bioclimatological models of the Carpathian basin. On the basis of environmental historical and climatic data we aimed to reconstruct the expected changes of forested areas in the Carpathian Basin. The results indicate that the surface wetness decreases in long term. Parallel to the Sphagnum-peat decline an open forest and forest steppe developed surrounding the bog. The complete disappearance of Sphagna from the area must be linked to a steady drop in rainfall, resulting in at least 50 mm deficit in the local water balance. This could have been achieved by an increased evapotranspiration as a result of elevated temperatures of the summer growth season. This deficit value must have exceeded even 100 mm during the Middle Holocene Transition. 展开更多
关键词 Carpathian Basin Climatic Change forest transformation Paleoecological Data
下载PDF
环境承载力中介下森林转型赋能区域生态宜居——作用路径与效应识别 被引量:1
18
作者 李超 朱震锋 邹成成 《林业经济问题》 北大核心 2024年第3期324-336,共13页
基于中国1999—2018年4次森林资源清查结果及相关统计数据,对中国31个省域(不含港澳台地区)的森林转型指数、环境承载力和生态宜居水平进行测评,然后借助中介效应模型探索森林转型赋能区域生态宜居的作用路径,并测度环境承载力对森林转... 基于中国1999—2018年4次森林资源清查结果及相关统计数据,对中国31个省域(不含港澳台地区)的森林转型指数、环境承载力和生态宜居水平进行测评,然后借助中介效应模型探索森林转型赋能区域生态宜居的作用路径,并测度环境承载力对森林转型赋能区域生态宜居水平的中介效应。结果显示:1999—2018年中国森林转型指数整体呈现稳中有升的发展态势,但不同地区之间水平差异比较明显,西南、东北地区森林转型效果最为显著;中国环境承载力的“U型”曲线演变特征明显,其中西南、华南、华东地区环境承载力较强,华东、华中、华南、华北地区环境承载力在回升阶段表现突出,东北、西南和西北地区在经济快速发展时期,环境承载力呈现下降趋势,突显出资源环境系统脆弱性;中国生态宜居水平总体呈现波动上升态势,其中西南、华北、华东地区上升态势最为明显;环境承载力在森林转型对区域生态宜居水平影响中具有中介作用,森林转型通过提升环境承载力部分解释了其对区域生态宜居水平的促进作用。 展开更多
关键词 生态宜居 森林转型 环境承载力 城镇化 中介效应
下载PDF
基于IHHT‑RF的配电网单相接地故障选线方法 被引量:1
19
作者 李泽文 黎文娇 +2 位作者 彭维馨 雷柳 梁流涛 《电力科学与技术学报》 CAS CSCD 北大核心 2024年第1期171-182,共12页
小电流系统发生单相接地故障时故障特征易受高接地过渡电阻、小初相角等弱故障条件影响而导致选线准确率低。为此,提出一种基于改进希尔伯特黄变换—随机森林(improved Hilbert⁃Huang transform⁃random forest,IHHT⁃RF)的配电网单相接... 小电流系统发生单相接地故障时故障特征易受高接地过渡电阻、小初相角等弱故障条件影响而导致选线准确率低。为此,提出一种基于改进希尔伯特黄变换—随机森林(improved Hilbert⁃Huang transform⁃random forest,IHHT⁃RF)的配电网单相接地故障选线方法。首先,提取每条线路在故障发生时的电流暂态信号,通过IHHT提取纯净的暂态电气量,构造标准差、能量熵和幅值畸变度3类特征向量;然后,将特征向量输入RF分类器建立故障选线模型,把故障选线问题转化为二分类问题;最后,将测量数据输入RF分类器中得出分类结果,实现故障线路的自动识别。仿真结果表明,该选线方法综合利用暂态信号的幅值、频率和能量等特征信息,不受弱故障条件、馈线结构等因素的影响,能有效提高故障选线的准确率,具有较强的适应性和可靠性。 展开更多
关键词 配电网 改进希尔伯特黄变换 随机森林 故障选线
下载PDF
基于二维散射变换的湖相碳酸盐岩储层厚度预测方法研究
20
作者 杨阳 雷友波 +4 位作者 王倩楠 王治国 杨涛 高静怀 苏朝光 《地球物理学报》 SCIE EI CAS CSCD 北大核心 2024年第4期1601-1612,共12页
济阳坳陷沙四段湖相碳酸盐岩受湖盆内沉积环境和构造运动等因素控制,储层厚度变化大,非均质性强,油气开发难度大.为了准确预测湖相碳酸盐岩储层的厚度,本文提出一种基于二维散射变换和随机森林的储层厚度预测方法.首先,引入二维散射变... 济阳坳陷沙四段湖相碳酸盐岩受湖盆内沉积环境和构造运动等因素控制,储层厚度变化大,非均质性强,油气开发难度大.为了准确预测湖相碳酸盐岩储层的厚度,本文提出一种基于二维散射变换和随机森林的储层厚度预测方法.首先,引入二维散射变换提取地震时频属性,该变换是在二维小波变换的基础上,通过迭代小波分解和非线性操作来实现的.与传统的二维小波变换对比,散射变换提取的时频属性具有局部形变稳定性以及对噪声鲁棒性的优点,有助于提高储层厚度预测的准确率.在此基础上,在有限测井数据的条件下,利用随机森林算法建立多尺度时频属性与测井解释厚度之间的非线性关系,实现湖相碳酸盐岩储层预测.模型数据的预测结果表明,与基于传统地震振幅属性的厚度预测和基于二维小波变换的储层厚度预测对比,本文所提的厚度预测方法具有最优的性能.叠后三维地震数据的预测结果表明,与基于传统地震振幅属性的厚度预测和基于二维小波变换的储层厚度预测对比,本文所提方法的厚度预测结果与实际钻井数据误差更小,提高了储层厚度预测的精度,清晰刻画了灰礁、灰滩与灰泥等三种沉积亚相的空间展布,有利于后续井位部署和优化. 展开更多
关键词 储层厚度预测 湖相碳酸盐岩 二维散射变换 随机森林
下载PDF
上一页 1 2 35 下一页 到第
使用帮助 返回顶部