基于支持向量机的入侵检测研究

根据入侵检测和支持向量机的特点,提出基于最小二乘支持向量机异常检测方法,并建立基于支持向量机入侵检测的模型,对网络数据进行采集,提取特征,进行分类,分辨正常的数据和异常的数据。并在KDD CUP'99标准入侵检测数据集上进行实验,选取data_10_percent子集,把该数据集中的41个属性作为特征,将该子集最后一个属性label属性为:back,ipsweep,neptun,ports-weep和normal各200个数据进行测试。实证表明:该方法能获得较高检测率和较低误警率。     

模糊支持向量机在滑坡危险性评价中的应用

为了将模糊支持向量机应用于滑坡灾害危险性评价工作,并进行验证,进行了研究区滑坡现状的调查,评价指标的选取,隶属函数的确定和基于FSVM的滑坡灾害危险性评价.选取了高程指标、坡度指标、岩土体指标、地表湿度指标及植被覆盖指标等5个指标作为滑坡灾害危险性评价指标,最后以莆田市为例,结合Rs和GIS技术,进行了滑坡危险性评价的研究.评价结果与实地验证和莆田市国土局提供的地质灾害现状图基本吻合.     

A coordinated control strategy using supercapacitor energy storage and series dynamic resistor for enhancement of fault ride-through of doubly fed induction generator

Power fluctuation and fault-related complication are the two major issues for doubly fed induction generator (DFIG)-based wind energy conversion system (WECS). The occurrence of fault leads to the rotor over current, stator over current, and DC-link overvoltage as well. These uncertainties may damage the rotor circuit, converter circuit and force the disconnection of wind system from the grid. To get rid of these issues, a supercapacitor energy storage element along with a passive series dynamic resistor (SDR) is suggested in this paper. Supercapacitor energy storage system (SCESS) is located across the DC-link, which able to handle the power fluctuation and the SDR is placed in rotor circuit, which will reduce the overcurrent possibility. Simulation is carried for a DFIG-based WECS for three phase to ground fault and two phase to ground fault. During symmetrical fault as well as asymmetrical fault, various operational disorders appeared such as rotor overcurrent, stator overcurrent and DC-Link overvoltage are found to be within their permissible limits. The results reveal the effectiveness of the proposed strategy over the conventional vector control scheme and SCESS as well.     

基于支持向量回归模型的水稻田甲烷排放通量预测研究

利用静态箱和气相色谱仪法获取水稻田甲烷排放通量数据,选取大气温度、土壤5 cm深温度、土壤pH、土壤Eh、土壤含水量和地表生物量作为影响因子.应用建立在结构风险最小化优化上的支持向量回归(ε-SVR)模型,采用留一法交叉检核网格搜索法(LOOCV)优化ε-SVR预测模型的参数,采用k折交叉检验的方法依据平均相对误差(MRE)和均方根误差(RMSE)对模型的精度进行验证,并与BP人工神经网络(BP-ANN)模型比较,评价ε-SVR预测模型的准确性.结果表明,通过LOOCV选择最优的惩罚因子C和损失系数ε,并由此构建的ε-SVR预测模型预测值和实测值具有很好的一致性,通过11折交叉验证后,测试样本的平均MRE为44%,平均RMSE为16.21 mg·(m2·h)-1.通过与BP-ANN模型比较,预测值和实际值相关系数达0.863,各项指标均优于BP-ANN预测模型,说明ε-SVR模型能够适用于水稻田甲烷排放通量的预测.     

基于PCA-AHPSO-SVR的煤层瓦斯含量预测研究

为了提高煤层瓦斯含量预测的准确性和科学性,通过主成分分析方法对影响煤层瓦斯含量的7个因素进行特征提取,消除影响因素之间的相关性,减少维度;用支持向量回归机对提取的因素进行训练,并用改进的自适应混合粒子群算法对SVR的参数进行优化,提出PCA-AHPSO-SVR模型;与PCA-PSO-SVR,PSO-SVR这2个模型在相同环境下进行30次运行比较。研究结果表明:研究提出的PCA-AHPSO-SVR模型较其他2种模型平均准确率分别提高5.51%和9.32%,稳定性更佳,可满足工程实际需求。     

基于北斗高精度定位的工务现场作业全过程安全监测系统研究

为了保障在铁路快速发展的同时,确保铁路工务安全和行车安全,基于工务现场作业过程的控制,提出了利用北斗高精度差分定位、大数据等技术,研究建立针对现场每个作业个体的工务作业全过程安全监测系统。结果表明:通过对现场每名作业个体的高精度、智能化管理,有效规范了现场人员的作业行为,增强对现场安全风险和隐患的预判和规避能力,保障了铁路从业人员人身安全。     

基于支持向量机的资源型城市可持续发展指标体系研究

资源型城市的可持续发展系统是由自然、经济、社会三个基本层次组成的相互作用且相互依赖的复杂巨系统。支持向量机回归具有非线性、并行分布处理、自学习等特点，可以较为逼真地模拟该系统。本文通过对资源型城市可持续发展指标体系现有主要方法的研究，提出以支持向量机为模型的资源型城市可持续发展指标的分析预测方法，并应用此方法构造支持向量机。绘制指标预测图。对铜陵市国民经济总产值进行了预测。结果表明本预测方法效果较好。     

呼和浩特市空气污染现状与防治对策

本文对呼和浩特市空气环境质量现况进行了科学地评价，并且分析了污染原因，在此基础上给出了相应的综合治理措施。     

Multivariate bootstrap confidence regions for abundance vector using

Abundance vector estimation is a well investigated problem in statistical ecology. The use of simple random sampling with replacement or replicated sampling ensures good asymptotic properties of the abundance vector estimators. However, real surveys are based on small sample sizes, and assuming any specific distribution of the abundance vector estimator may be hazardous.In this paper we focus our attention on situations where the population is not too large and the sample size is small. We propose bootstrap multivariate confidence regions based on data depth. Data depth is a geometrical concept of ordering data from the center outwardly in higher dimensions. The Simplicial depth, the Tukey's depth and the Mahalanobis depth are presented. In order to build confidence regions in the presence of a skewed distribution of the abundance vector estimator, the use of Tukey's depth is suggested. The proposed method has been applied to the benthic community of Lake Lesina. A comparison with Mahalanobis depth and standard existing methods is reported.