Mechanism-learning prediction model for pitting depth of buried pipeline based on HMOGWO-RF [PDF]
Objective China's oil and gas pipeline networks are expected to reach 24×104 km by 2025. Pipeline transportation has become one of the key means of transportation in the country.
Fulin SONG, Hong ZHAO, Xingyuan MIAO
core +1 more source
Hyperspectral Imaging for the Detection of Vitamin C Content in Potatoes Based on Fisher Discriminant Analysis Separable Information Fusion [PDF]
In order to improve the accuracy and reliability of the prediction results of the vitamin C (VC) content in potatoes by hyperspectral imaging, a method for constructing input variables for predictive models based on Fisher discriminant analysis (FDA ...
GUO Linge, YIN Yong, YU Huichun, YUAN Yunxia
core +1 more source
Research on the Data Mining Technique Based on Generalized Small-Samples [PDF]
数据挖掘涉及人工智能、模式识别、机器学习、统计学等领域,不同领域利用各自不同的技术和方法对数据挖掘进行研究,将不同领域的理论、技术进行融合是数据挖掘研究一个方向。小样本数据挖掘则是数据挖掘的一个研究方向,针对高维小样本数据如何构造挖掘算法及其效率是小样本数据挖掘的核心问题。 文章在数据挖掘前人研究基础上,提出广义小样本概念,对广义小样本数据挖掘的理论进行研究,从统计方法与统计学习角度出发,研究了小样本数据挖掘中的偏最小二乘回归、支持向量分类与回归机的相关理论。文章先对经典统计方法作了论述 ...
游文杰
core
Optimal Configuration of Distributed Generation Considering Carbon Trading and Reactive Power Compensation [PDF]
Under the background of “double carbon”, the introduction of carbon trading and reactive power compensation mechanism in the planning of distributed generation is an important means to realize the low-carbon and stability of power system.
CHU Zhuang, SUN Jianhao
core +1 more source
泄漏电流是污秽绝缘子在线监测参数,能动态地反映绝缘子表面的放电状态。文中开展了瓷绝缘子人工污秽放电试验,利用Hilbert-Huang变换分析了不同污闪阶段的泄漏电流固有模态函数分量、Hilbert边际谱与时频熵,从时频域及波形细节提取了15个特征量,使用主成分分析法与最小二乘支持向量机分类器对污秽放电状态进行识别。结果表明:起始放电阶段与闪络阶段的泄漏电流固有模态函数分量较多;泄漏电流的Hilbert边际谱上频率主要分布在0~150 Hz、200~250 Hz范围内 ...
方春华 +5 more
doaj
复合绝缘子因优异的绝缘性能与机械强度在电网中得到广泛应用,但其护套-芯棒界面产生的脱粘与分层等隐蔽性缺陷是诱发断串、击穿等恶性故障的主要原因,现有检测手段难以满足此类缺陷的高灵敏度检测需求。文中提出了一种基于非线性超声的复合绝缘子界面缺陷诊断方法。以无缺陷、脱粘与分层三种典型缺陷的复合绝缘子样本为研究对象,搭建非线性超声试验平台并采集超声回波信号,引入离散傅里叶变换-最小二乘法(Discrete Fourier Transform - Least Squares,DFT-LS)提取频域非线性特征 ...
吴田 +5 more
doaj
Research on Feature Selection in Intrusion Detection Based on Genetic Algorithm [PDF]
随着网络的广泛应用,网络安全问题也越来越严峻。当前网络攻击方法层出不穷,入侵规模不断扩大,使得目前防火墙等被动的网络安全机制对许多攻击无能为力。入侵检测技术作为一种主动防御技术,弥补了传统安全技术的不足,故对入侵检测技术的研究十分必要。一般情况下,要提高入侵检测系统分类器的识别率总是过度地提取特征信息,结果不仅特征空间维数增大,而且存在较大的冗余。因而需要在尽量不降低分类精度的前提下降低特征空间的维数,从而提高入侵检测系统的速度。 本文主要结合遗传算法、粒子群算法、支持向量机等技术改进特征选择方法 ...
陈闽奎
core
[Rehabilitation guidelines and specifications for traumatic spinal cord injury (2025 edition)]. [PDF]
Professional Committee Of Spine And Spinal Cord Injury Rehabilitation China Association Of Rehabilitation Of Disabled Persons +3 more
europepmc +1 more source
[Master manipulator of vascular intervention surgical robot based on haptic feedback]. [PDF]
Han G, Sun Z, Tang X.
europepmc +1 more source
[Intelligent fetal state assessment based on genetic algorithm and least square support vector machine]. [PDF]
Zhang Y, Zhao Z, Ye H.
europepmc +1 more source

