期刊文献+
共找到232,452篇文章
< 1 2 250 >
每页显示 20 50 100
一种基于粗糙熵的改进K-modes聚类算法
1
作者 刘财辉 曾雄 谢德华 《南京理工大学学报》 CAS CSCD 北大核心 2024年第3期335-341,共7页
K-modes聚类算法被广泛应用于人工智能、数据挖掘等领域。传统的K-modes聚类算法有不错的聚类效果,但是存在迭代次数多、计算量大、容易受到冗余属性的干扰等问题,且仅采用简单的0-1匹配的方法来定义2个样本属性值之间的距离,没有充分... K-modes聚类算法被广泛应用于人工智能、数据挖掘等领域。传统的K-modes聚类算法有不错的聚类效果,但是存在迭代次数多、计算量大、容易受到冗余属性的干扰等问题,且仅采用简单的0-1匹配的方法来定义2个样本属性值之间的距离,没有充分考虑每个属性对聚类结果的影响。针对上述问题,该文将粗糙熵引入K-modes算法。首先利用粗糙集属性约简算法消除冗余属性,确定各属性的重要程度;然后利用粗糙熵确定每个属性的权重,从而定义新的类内距离。将该文所提算法与传统的K-modes聚类算法分别在4组公开数据集上进行对比试验。试验结果表明,该文所提算法聚类准确率比传统的K-modes聚类算法更高。 展开更多
关键词 聚类 K-modes算法 粗糙集 粗糙熵 属性约简 权重
下载PDF
基于K-modes聚类算法的山东省传统村落空间风貌类型及区划研究 被引量:1
2
作者 范勇 李玄 肖文杰 《小城镇建设》 2024年第5期100-107,共8页
传统村落的类型解析及空间区划是开展传统村落整体性保护和区域性发展的基础前提,本文在对山东省传统村落调查的基础上,基于空间基因理论视角,从地景、聚落、建筑、文化4个层次构建起13个指标的传统村落空间风貌分类指标体系,并采用K-mo... 传统村落的类型解析及空间区划是开展传统村落整体性保护和区域性发展的基础前提,本文在对山东省传统村落调查的基础上,基于空间基因理论视角,从地景、聚落、建筑、文化4个层次构建起13个指标的传统村落空间风貌分类指标体系,并采用K-modes聚类算法对山东省177个传统村落进行聚类分析,得到八大空间风貌类型,进一步结合区域文化、地理特点及行政区划,划分出山东省5个传统村落风貌区,从宏观视角分析了山东省传统村落空间风貌特征及其形成与发展的内在逻辑和地理分布规律,为更加整体全面地认识山东省传统村落特点、开展区域性传统村落集中连片保护利用等工作提供科学参考。 展开更多
关键词 传统村落 空间基因 K-modes聚类算法 空间区划 山东省
下载PDF
基于主客观环流分型的强降水数值预报MODE检验方法及其在2019年暖季东北地区的应用
3
作者 齐铎 崔晓鹏 +4 位作者 陈力强 黄丽君 刘松涛 卜文惠 王承伟 《大气科学》 CSCD 北大核心 2024年第3期1113-1130,共18页
本文构建了基于主客观环流分型的强降水数值预报空间检验(MODE)方法框架,并利用该框架对欧洲中期天气预报中心全球模式(ECMWF)和中国气象局区域中尺度数值天气预报模式(CMA_MESO)的2019年暖季东北地区强降水预报进行检验。结果表明,201... 本文构建了基于主客观环流分型的强降水数值预报空间检验(MODE)方法框架,并利用该框架对欧洲中期天气预报中心全球模式(ECMWF)和中国气象局区域中尺度数值天气预报模式(CMA_MESO)的2019年暖季东北地区强降水预报进行检验。结果表明,2019年暖季东北地区54个强降水日的环流型可分为:西风槽型(15个)、副热带高压影响型(13个)、急流型(5个)、西部(12个)和东部冷涡型(9个)。其中,西风槽型和急流型以区域性强降水为主,模式对其强降水发生与否的预报能力强,TS评分较高;西部、东部冷涡型强降水的局地性强,模式对其强降水发生与否的预报能力差,TS评分低;副热带高压影响型也以区域性强降水为主,模式对其强降水发生与否的预报能力也比较强,但是对其强降水质心位置、强度、面积等属性预报偏差较大,TS评分也相对较低。另外,从两种模式预报性能对比看,CMA_MESO强降水强度和面积预报较实况普遍偏强,虽然其预报的TS评分一般高于ECMWF,但其对强降水预报的空报率也都比ECMWF大,对强降水的属性预报偏差一致性一般也低于ECMWF,其预报的可订正性整体上不及ECMWF。 展开更多
关键词 主客观融合环流分型 东北冷涡客观识别 强降水 数值预报 mode检验
下载PDF
K-Modes聚类数据收集和发布过程中的混洗差分隐私保护方法 被引量:1
4
作者 蒋伟进 陈艺琳 +3 位作者 韩裕清 吴玉庭 周为 王海娟 《通信学报》 EI CSCD 北大核心 2024年第1期201-213,共13页
针对目前聚类数据收集与发布安全性不足的问题,为保护聚类数据中的用户隐私并提高数据质量,基于混洗差分隐私模型,提出一种去可信第三方的K-Modes聚类数据收集和发布的隐私保护方法。首先,使用K-Modes聚类数据收集算法对用户数据进行采... 针对目前聚类数据收集与发布安全性不足的问题,为保护聚类数据中的用户隐私并提高数据质量,基于混洗差分隐私模型,提出一种去可信第三方的K-Modes聚类数据收集和发布的隐私保护方法。首先,使用K-Modes聚类数据收集算法对用户数据进行采样并加噪,再通过填补取值域随机排列发布算法打乱采样数据的初始顺序,使恶意攻击者不能根据用户与数据之间的关系识别出目标用户。然后,尽可能减小噪声的干扰,利用循环迭代的方式计算出新的质心完成聚类。最后,从理论层面上分析了以上3种方法的隐私性、可行性和复杂度,并利用3个真实数据集和近年来具有权威性的同类算法KM、DPLM、LDPKM等进行准确率、熵值的对比,验证所提方法的有效性。实验结果表明,所提方法的隐私保护和发布数据质量均优于当前同类算法。 展开更多
关键词 混洗差分隐私 K-modes聚类 隐私保护 数据收集 数据发布
下载PDF
改进的k-modes聚类算法在协同过滤就业推荐算法中的应用
5
作者 刘逗逗 王文发 许淳 《延安大学学报(自然科学版)》 2024年第2期96-100,共5页
为了给高校毕业生提供精准的个性化就业推荐服务,将基于动态权重相互依存距离的改进k-modes聚类算法应用于协同过滤推荐算法中。定义不同样本点属性之间的距离等于属性值内部距离和属性间外部距离的加权和,选择初始簇质心时,动态调整样... 为了给高校毕业生提供精准的个性化就业推荐服务,将基于动态权重相互依存距离的改进k-modes聚类算法应用于协同过滤推荐算法中。定义不同样本点属性之间的距离等于属性值内部距离和属性间外部距离的加权和,选择初始簇质心时,动态调整样本点与簇质心的距离以及簇密度的组合权重,动态设置簇密度计算公式的半径,根据样本点的概率值选出初始簇质心;迭代计算和优化得到满足精度的学生簇和职位簇;构建学生-职位矩阵,计算应届生和往届生的相似度、往届生和入职岗位的相似度,选择二者的相似度超过阈值的应届生簇和职位簇组合为匹配对进行匹配,并将匹配信息降序排列形成匹配列表,依据匹配列表进行双向推荐和信息推送,为高校的就业推荐和指导提供信息导向和技术支持。 展开更多
关键词 双边匹配算法 协同过滤算法 聚类分析 k-modes算法 相似性度量
下载PDF
MODE降水检验评价指标改进及卷积半径应用
6
作者 杨富燕 陈百炼 +2 位作者 彭芳 胡欣欣 李彦霖 《气象科技》 2024年第2期218-227,共10页
基于对象的诊断检验方法(MODE)受降水临界值、卷积半径、属性权重等参数的影响,合理选取卷积半径并准确表征预报场与观测场之间的空间相似度决定了MODE的应用效果。本文基于2020年夏季贵州54个降水个例,以多源融合降水(CMPA)作为实况,使... 基于对象的诊断检验方法(MODE)受降水临界值、卷积半径、属性权重等参数的影响,合理选取卷积半径并准确表征预报场与观测场之间的空间相似度决定了MODE的应用效果。本文基于2020年夏季贵州54个降水个例,以多源融合降水(CMPA)作为实况,使用MODE和FSS评分(Fractions Skill Score)对中国气象局广东快速更新同化数值预报系统(CMA-GD)24 h日降水预报进行空间检验。结果表明:卷积半径过小易造成MODE提取降水对象过多,而卷积半径过大则导致局部降水信息丢失,无法从降水场中提取到降水对象。不同卷积半径下计算的最大相似度中值(M_(MI))存在突变。在M_(MI)基础上引入面积权重构造面积平均最大相似度(A_(MMI))。A_(MMI)不受提取降水对象个数的影响,较M_(MI)更具有稳定性,用于表征降水场之间的整体空间相似程度更为合理。根据对象总面积随卷积半径的变化将降水分为大范围降水和局部降水2类。大范围降水平均总面积随着卷积半径的增加而增加,A_(MMI)随卷积半径变化不大。随着卷积半径的增加,局部降水平均总面积减小,平均A_(MMI)有所减小。局部降水对卷积半径选取较为敏感,以观测场对象面积变化不超过10%的最大半径作为卷积半径有助于保留降水场大部分信息。 展开更多
关键词 降水 mode 空间检验 参数
下载PDF
Slowing down of the summer Southern Hemisphere Annular Mode trend against the background of ozone recovery
7
作者 Fei Zheng 《Atmospheric and Oceanic Science Letters》 CSCD 2024年第1期1-8,共8页
20世纪末期,南半球热带外地区经历了显著的气候变化,包括夏季南半球环状模(SAM)的显著上升趋势,伴随着南极半岛的增暖和别林斯高晋海的海冰融化.这些趋势主要是由20世纪末期南极平流层臭氧消耗所驱动的.本文发现,自2001年左右以来,在南... 20世纪末期,南半球热带外地区经历了显著的气候变化,包括夏季南半球环状模(SAM)的显著上升趋势,伴随着南极半岛的增暖和别林斯高晋海的海冰融化.这些趋势主要是由20世纪末期南极平流层臭氧消耗所驱动的.本文发现,自2001年左右以来,在南极平流层臭氧恢复的背景下,观测到的夏季SAM的上升趋势已经趋于平缓,验证了前人利用数值模拟预测的夏季SAM上升趋势减缓现象,与SAM在臭氧恢复后趋势只减缓但没有逆转不同,南极地表气温和海冰的趋势发生了逆转.南极半岛由变暖趋势转为降温趋势,别林斯高晋海域的海冰由融化趋势转为增多趋势. 展开更多
关键词 南半球环状模 南极臭氧损耗 南极臭氧恢复
下载PDF
基于ADS-B与Mode-SEHS联合观测的民航空域风场重建方法
8
作者 陈敏 王浩楠 +1 位作者 陈万通 任诗雨 《国外电子测量技术》 2024年第6期102-109,共8页
准确实时的风场数据对保障民航飞行安全有着重要作用,针对风场的精确重构问题,提出了一种基于飞行器监测数据的风场重建方法。旨在利用广播式自动相关监视和S模式增强型监视联合观测数据计算空域内的风观测值,并结合机器学习中的高斯过... 准确实时的风场数据对保障民航飞行安全有着重要作用,针对风场的精确重构问题,提出了一种基于飞行器监测数据的风场重建方法。旨在利用广播式自动相关监视和S模式增强型监视联合观测数据计算空域内的风观测值,并结合机器学习中的高斯过程回归模型,利用时间和空间上离散的风观测值进行模型训练,完整重建目标空域风场。实验结果表明,重建的风场风速的平均绝对误差为2.72m/s,相对误差为8.21%,风向的平均绝对误差为3.66°,证明了方法能够快速地完成准确实时的风场重建。 展开更多
关键词 广播式自动相关监视 S模式增强型监视 高斯过程回归 风场重建
下载PDF
Factors Influencing Mode of Delivery: A Case-Control Study
9
作者 Linfeng Mo Na Wang +5 位作者 Bin Peng Jiwei Wu Lishan Tang Han Liu Yonghua He Jie Hu 《Advances in Reproductive Sciences》 CAS 2024年第2期141-154,共14页
The primary objective of this investigation was to scrutinize the prepregnancy conditions and lifestyles of 2046 women residing in Liuzhou City, with the aim of delineating the determinants of delivery methods. Eviden... The primary objective of this investigation was to scrutinize the prepregnancy conditions and lifestyles of 2046 women residing in Liuzhou City, with the aim of delineating the determinants of delivery methods. Evidently, the study unearthed substantial correlations between prepregnancy body mass index, educational attainment, exposure to passive smoking, medical history, and other variables with the mode of delivery. Furthermore, a predictive nomogram model was formulated to accurately forecast the likelihood of cesarean section. These discernments equip pertinent authorities with the means to institute targeted screening and supportive measures for women contemplating pregnancy based on these identified factors. Moreover, provision of services such as prepregnancy counseling and clinical risk assessments could be instrumental in curbing the incidence of cesarean section. 展开更多
关键词 mode of Delivery Cesarean Section Influencing Factors NOMOGRAPH
下载PDF
Electroencephalogram Signal Correlations between Default Mode Network and Attentional Functioning
10
作者 Moemi Matsuo Takashi Higuchi +3 位作者 Toranosuke Abe Takuya Ishibashi Ayano Egashira Rio Kamashita 《Journal of Behavioral and Brain Science》 2024年第4期119-134,共16页
Attentional issues may affect acquiring new information, task performance, and learning. Cortical network activities change during different functional brain states, including the default mode network (DMN) and attent... Attentional issues may affect acquiring new information, task performance, and learning. Cortical network activities change during different functional brain states, including the default mode network (DMN) and attention network. We investigated the neural mechanisms underlying attentional functions and correlations between DMN connectivity and attentional function using the Trail-Making Test (TMT)-A and -B. Electroencephalography recordings were performed by placing 19 scalp electrodes per the 10 - 20 system. The mean power level was calculated for each rest and task condition. Non-parametric Spearman’s rank correlation was used to examine the correlation in power levels between the rest and TMT conditions. The most significant correlations during TMT-A were observed in the high gamma wave, followed by theta and beta waves, indicating that most correlations were in the parietal lobe, followed by the frontal, central, and temporal lobes. The most significant correlations during TMT-B were observed in the beta wave, followed by the high and low gamma waves, indicating that most correlations were in the temporal lobe, followed by the parietal, frontal, and central lobes. Frontoparietal beta and gamma waves in the DMN may represent attentional functions. 展开更多
关键词 Cortical Network Activities ELECTROENCEPHALOGRAPHY ATTENTION Default mode Network
下载PDF
Impact of Delivery Mode on Morbidity in Preterm Infants with Very Low Birth Weights (<1500 Grams)
11
作者 Manuela Colle José Mauro Madi +2 位作者 Luciano Selistre Gabriela Françoes Rostirolla Marcelo Costamilan Rombaldi 《Open Journal of Obstetrics and Gynecology》 2024年第10期1583-1590,共8页
The mode of delivery and gestational age for very-low-birth-weight (VLBW) preterm infants are not yet well established and are constant topics of debate. Objective: To analyze the impact of delivery mode on morbidity ... The mode of delivery and gestational age for very-low-birth-weight (VLBW) preterm infants are not yet well established and are constant topics of debate. Objective: To analyze the impact of delivery mode on morbidity in preterm infants weighing less than 1500 g. Results: Among 21,957 births, 81 were analyzed;53 were delivered vaginally, and 28 were delivered by cesarean section. The median maternal age, gestational age and body mass index among those delivered vaginally and by cesarean section were 20 years and 22.5 years, 27.6 weeks and 30.1 weeks, and 26.0 kg/m2 and 27.8 kg/m2, respectively. With respect to neonatal blood gas parameters, for those born vaginally and by cesarean section, the median pH was 7.32 and 7.24, the pCO2 was 41.5 mmHg and 51.1 mmHg, and the pO2 was 22.3 mmHg and 16 mmHg. The median fetal weight among those born by cesarean section and vaginally were 1180 g and 955 g, respectively. The median Apgar scores at the first and fifth minutes among those born by cesarean section and vaginally were 5.00 and 8.00 and 4.50 and 7.00, respectively. Conclusion: There was no significant difference between the results of vaginal and cesarean delivery for VLBW infants. Thus, further studies on this subject are needed. 展开更多
关键词 PREMATURITY Very Low Birth Weight Newborns mode of Delivery EPIDEMIOLOGY MORBIDITY
下载PDF
基于k-modes聚类算法的混洗差分隐私方法
12
作者 祁富 陈丽敏 《牡丹江师范学院学报(自然科学版)》 2024年第2期6-13,共8页
首次提出一种基于k-modes聚类算法的混洗差分隐私保护方案(简称SDPk-modes).SDPk-modes根据每个数据之间的距离划分为不同的组,得到足够的细粒度优化效用,采用基于梯度随机扰动技术使计算最优概率耗时更短;在k-modes聚类过程中,通过将... 首次提出一种基于k-modes聚类算法的混洗差分隐私保护方案(简称SDPk-modes).SDPk-modes根据每个数据之间的距离划分为不同的组,得到足够的细粒度优化效用,采用基于梯度随机扰动技术使计算最优概率耗时更短;在k-modes聚类过程中,通过将数据中频繁出现的特征向量作为聚类中心点,基于属性熵的距离度量方法,加快算法收敛至聚类中心的速度,解决原始算法聚类速度慢、易陷入局部最优等问题,显著提高聚类的效果.实验验证表明,本文提出的方案优于当前同类方案. 展开更多
关键词 混洗差分隐私 k-modes 随机响应机制 隐私保护
下载PDF
Factors Influence Hanoians’ Choice of Non-Motorized Transport Mode to Access Bus Stations
13
作者 Ngo Trung Phuong Masahiko Kikuchi +1 位作者 Aya Kojima Hisashi Kubota 《Journal of Transportation Technologies》 2024年第3期372-389,共18页
Hanoi’s rapid urbanization has led to a surge in private vehicle ownership, particularly motorcycles, amidst inadequate public transportation infrastructure. Despite government efforts, many still prefer motorized tr... Hanoi’s rapid urbanization has led to a surge in private vehicle ownership, particularly motorcycles, amidst inadequate public transportation infrastructure. Despite government efforts, many still prefer motorized transport, citing mobility and safety concerns, exacerbated by insufficient pedestrian infrastructure. This study examines the motivations behind this reliance on motorized vehicles, particularly motorcycles, in Hanoi. Findings reveal safety and convenience as primary factors driving motorized transport use, especially for accessing bus stations. Economic incentives could promote non-motorized travel and public transport adoption. Policy implications highlight the importance of addressing economic factors and improving access infrastructure to manage motorized vehicle reliance and foster sustainable urban mobility in Hanoi. 展开更多
关键词 Access mode Choice Non-Motorized Vehicles Motorized Vehicles Bus User Urban Transport Behavior
下载PDF
Enhanced Fuzzy Logic Control Model and Sliding Mode Based on Field Oriented Control of Induction Motor
14
作者 Alaa Tahhan Feyzullah Temurtaş 《World Journal of Engineering and Technology》 2024年第1期65-79,共15页
In the context of induction motor control, there are various control strategies used to separately control torque and flux. One common approach is known as Field-Oriented Control (FOC). This technique involves transfo... In the context of induction motor control, there are various control strategies used to separately control torque and flux. One common approach is known as Field-Oriented Control (FOC). This technique involves transforming the three-phase currents and voltages into a rotating reference frame, commonly referred to as the “dq” frame. In this frame, the torque/speed and flux components are decoupled, allowing for independent control, by doing so, the motor’s speed can be regulated accurately and maintain a constant flux which is crucial to ensure optimal motor performance and efficiency. The research focused on studying and simulating a field-oriented control system using fuzzy control techniques for an induction motor. The aim was to address the issue of parameter variations, particularly the change in rotor resistance during motor operation, which causes the control system to deviate from the desired direction. This deviation implies to an increase in the magnetic flux value, specifically the flux component on the q-axis. By employing fuzzy logic techniques to regulate flux vector’s components in the dq frame, this problem was successfully resolved, ensuring that the magnetic flux value remains within the nominal limits. To enhance the control system’s performance, response speed, and efficiency of the motor, sliding mode controllers were implemented to regulate the current in the inner loop. The simulation results demonstrated the proficiency of the proposed methodology. 展开更多
关键词 Induction Motor Vector Control Fuzzy Logic Control Sliding mode
下载PDF
Innovation and Practice of the Cultivation Mode of Application-Oriented Undergraduate Talents in Agricultural and Forestry Colleges under the Guidance of the Spirit of the Great Northern Wilderness
15
作者 Dawei Yin Xiaohong Guo +10 位作者 Hui Yang Wen Zheng Mingyu Fan Xiaodan Xu Qitao Li Xiuzhi Wang Hongyu Li Guiping Zheng Haize Wang Menghong Liu Yueyue Liu 《Open Journal of Applied Sciences》 2024年第8期2027-2038,共12页
Heilongjiang reclamation area has made great progress since its development and construction, among which the agricultural and forestry colleges in China have played a key role in the input of talents for the producti... Heilongjiang reclamation area has made great progress since its development and construction, among which the agricultural and forestry colleges in China have played a key role in the input of talents for the production of reclamation area, and the spirit of the Great Wilderness has important strategic significance for the cultivation of agricultural and forestry talents. Taking Heilongjiang Bayi Agricultural University as an example, this paper analyzes the ways of the Great Northern Wilderness Spirit for cultivating application-oriented undergraduate talents in agricultural and forestry colleges, improving the curriculum system of undergraduate talents through innovation, strengthening the cultivation of students’ innovative ability, leading students’ scientific research and academic level, and enhancing students’ practical ability. To innovate the cultivation mode of application-oriented undergraduate talents in agricultural and forestry colleges under the guidance of the spirit of the Great Northern Wilderness, and to provide a reference mode for the cultivation of undergraduate talents in agricultural and forestry colleges in China. 展开更多
关键词 Northern Wilderness Spirit Undergraduate Talents Research Application Cultivation mode INNOVATION
下载PDF
Dynamic Analysis of an Algae-Bacteria Ecological Model
16
作者 Gaopeng Sun Hengguo Yu 《Journal of Applied Mathematics and Physics》 2024年第1期362-382,共21页
In the paper, under the framework of exploring the interaction between algae and bacteria, an algae-bacteria ecological model was established to analyze the interaction mechanism and growth coexistence mode between al... In the paper, under the framework of exploring the interaction between algae and bacteria, an algae-bacteria ecological model was established to analyze the interaction mechanism and growth coexistence mode between algicidal bacteria and algae. Firstly, mathematical work mainly provided some threshold conditions to ensure the occurrence of transcritical bifurcation and saddle-node bifurcation, which could provide certain theoretical support for selecting key ecological environmental factors and numerical simulations. Secondly, the numerical simulation work dynamically displayed the evolution process of the bifurcation dynamic behavior of the model (2.1) and the growth coexistence mode of algae and algicidal bacteria. Finally, it was worth summarizing that intrinsic growth rate and combined capture effort of algae population had a strong influence on the dynamic behavior of the model (2.1). Furthermore, it must also be noted that transcritical bifurcation and saddle-node bifurcation were the inherent driving forces behind the formation of steady-state growth coexistence mode between algicidal bacteria and algae. In summary, it was hoped that the results of this study would contribute to accelerating the study of the interaction mechanism between algicidal bacteria and algae. 展开更多
关键词 ALGAE Algicidal Bacteria Transcritical Bifurcation Saddle-Node Bifurcation Coexistence mode
下载PDF
Variational Mode Decomposition-Informed Empirical Wavelet Transform for Electric Vibrator Noise Analysis
17
作者 Zhenyu Xu Zhangwei Chen 《Journal of Applied Mathematics and Physics》 2024年第6期2320-2332,共13页
Electric vibrators find wide applications in reliability testing, waveform generation, and vibration simulation, making their noise characteristics a topic of significant interest. While Variational Mode Decomposition... Electric vibrators find wide applications in reliability testing, waveform generation, and vibration simulation, making their noise characteristics a topic of significant interest. While Variational Mode Decomposition (VMD) and Empirical Wavelet Transform (EWT) offer valuable support for studying signal components, they also present certain limitations. This article integrates the strengths of both methods and proposes an enhanced approach that integrates VMD into the frequency band division principle of EWT. Initially, the method decomposes the signal using VMD, determining the mode count based on residuals, and subsequently employs EWT decomposition based on this information. This addresses mode aliasing issues in the original method while capitalizing on VMD’s adaptability. Feasibility was confirmed through simulation signals and ultimately applied to noise signals from vibrators. Experimental results demonstrate that the improved method not only resolves EWT frequency band division challenges but also effectively decomposes signal components compared to the VMD method. 展开更多
关键词 Electric Vibrator Noise Analysis Signal Decomposing Variational mode Decomposition Empirical Wavelet Transform
下载PDF
Wind Speed Prediction Based on Improved VMD-BP-CNN-LSTM Model
18
作者 Chaoming Shu Bin Qin Xin Wang 《Journal of Power and Energy Engineering》 2024年第1期29-43,共15页
Amid the randomness and volatility of wind speed, an improved VMD-BP-CNN-LSTM model for short-term wind speed prediction was proposed to assist in power system planning and operation in this paper. Firstly, the wind s... Amid the randomness and volatility of wind speed, an improved VMD-BP-CNN-LSTM model for short-term wind speed prediction was proposed to assist in power system planning and operation in this paper. Firstly, the wind speed time series data was processed using Variational Mode Decomposition (VMD) to obtain multiple frequency components. Then, each individual frequency component was channeled into a combined prediction framework consisting of BP neural network (BPNN), Convolutional Neural Network (CNN) and Long Short-Term Memory Network (LSTM) after the execution of differential and normalization operations. Thereafter, the predictive outputs for each component underwent integration through a fully-connected neural architecture for data fusion processing, resulting in the final prediction. The VMD decomposition technique was introduced in a generalized CNN-LSTM prediction model;a BPNN model was utilized to predict high-frequency components obtained from VMD, and incorporated a fully connected neural network for data fusion of individual component predictions. Experimental results demonstrated that the proposed improved VMD-BP-CNN-LSTM model outperformed other combined prediction models in terms of prediction accuracy, providing a solid foundation for optimizing the safe operation of wind farms. 展开更多
关键词 Wind Speed Forecast Long Short-Term Memory Network BP Neural Network Variational mode Decomposition Data Fusion
下载PDF
Fuzzy BC-k-modes:一种分类矩阵对象数据的聚类算法
19
作者 李顺勇 余曼 王改变 《计算机应用与软件》 北大核心 2023年第1期287-297,共11页
传统的聚类算法主要对具有单值属性的数据进行聚类研究,针对矩阵对象数据的研究较少,提出一种新的fuzzy between-cluster k-modes(简称Fuzzy BC-k-modes)聚类算法。在Fuzzy BC-k-modes算法中,采用增加簇间信息(不同类中的对象到其他类... 传统的聚类算法主要对具有单值属性的数据进行聚类研究,针对矩阵对象数据的研究较少,提出一种新的fuzzy between-cluster k-modes(简称Fuzzy BC-k-modes)聚类算法。在Fuzzy BC-k-modes算法中,采用增加簇间信息(不同类中的对象到其他类中心的距离)去修正目标函数,在对修正的目标函数寻求局部最优解时,提出隶属度矩阵的更新公式。最后,在四个真实数据集上验证了Fuzzy BC-k-modes算法的有效性,并且分析了模糊因子与隶属度间的关系。 展开更多
关键词 簇间信息 分类矩阵对象数据 聚类 Fuzzy BC-k-modes算法
下载PDF
基于K-modes聚类算法的辽宁传统村落划分及保护策略
20
作者 张宇 杜晓月 董丽 《住宅产业》 2023年第5期22-26,共5页
本文基于传统村落的自然环境、人文历史、民居建筑等分类特征,使用K-modes聚类算法,对辽宁省30个国家级传统村落进行聚类划分,将其整合为景区依托型、民族特色型、生态文化型、休闲观光型、文化遗产型共五个类别,最后根据归纳的聚类典... 本文基于传统村落的自然环境、人文历史、民居建筑等分类特征,使用K-modes聚类算法,对辽宁省30个国家级传统村落进行聚类划分,将其整合为景区依托型、民族特色型、生态文化型、休闲观光型、文化遗产型共五个类别,最后根据归纳的聚类典型特征,提出有针对性的辽宁省传统村落保护发展策略。 展开更多
关键词 辽宁地区 传统村落 K-modes聚类
下载PDF
上一页 1 2 250 下一页 到第
使用帮助 返回顶部