Application of support vector machine to ship steering

System identification is an effective way for modeling ship manoeuvring motion and ship manoeuvrability prediction. Support vector machine is proposed to identify the manoeuvring indices in four different response models of ship steering motion, including the first order linear, the first order nonlinear, the second order linear and the second order nonlinear models. Predictions of manoeuvres including trained samples by using the identified parameters are compared with the results of free-running model tests. It is discussed that the different four categories are consistent with each other both analytically and numerically. The generalization of the identified model is verified by predicting different untrained manoeuvres. The simulations and comparisons demonstrate the validity of the proposed method.     

基于支持向量机的发动机气路故障预诊断

为实现航空发动机气路故障在线预诊断, 分析了地空数据链系统中发动机气路参数报文的协议格式, 建立了基于支持向量机算法的发动机气路参数在线预测模型。以便携式地空数据链收发系统为硬件基础, 构建发动机报文并行处理系统, 获取建模所需的训练样本。利用最终误差预报准则确定样本数据嵌入维数, 实现时序样本数据的相空间重构。提出自适应网格搜索法优化支持向量机建模参数, 获得气路参数在线预测模型, 与航路飞机建立地空数据链通信, 预测气路参数趋势。预测结果表明: 参数低压转子转速、高压转子转速、尾气温度与燃油流量的相对预测误差分别为2.5%、2.1%、1.9%与2.3%, 因此, 支持向量机模型具有较高预测精度。     

基于支持向量机的岩爆模式识别及预测

采用支持向量机,运用分析国内、外工程岩爆数据,以岩石单轴抗压强度与单轴抗拉强度的比值、洞室围岩的最大切向应力与岩石单轴抗压强度比值及弹性能量指数作为评判指标,对典型岩爆进行模式识别即分类,并进行了预测.试验结果表明,该预测方法具有较高的准确率,较好地解决了小样本及非线性等实际问题.     

Forecasting of stock returns by using manifold wavelet support vector machine

An admissible manifold wavelet kernel is proposed to construct manifold wavelet support vector machine(MWSVM) for stock returns forecasting.The manifold wavelet kernel is obtained by incorporating manifold theory into wavelet technique in support vector machine(SVM).Since manifold wavelet function can yield features that describe of the stock time series both at various locations and at varying time granularities,the MWSVM can approximate arbitrary nonlinear functions and forecast stock returns accurately.T...     

Gearbox device failure mode criticality analysis based on support vector machine

A method which integrates expert evaluation method and support vector machine (SVM) method is introduced for failure mode criticality analysis (FMCA) about the gearbox device. An expert evaluation standard is built by using expert evaluation method. The experts make scores about the gearbox failure mode. In order to overcome the subjectivity of expert evaluation method, we use SVM method to make a comprehensive prediction about the scores. According to the comprehensive prediction evaluation results, the FMCA of the gearbox device can be obtained. The analysis shows that the method used in this paper not only can effectively solve the problem which is unable to get specific failure rate in the qualitative analysis, but also can solve the problem which needs lots of data in the quantitative analysis.     

基于优化支持向量机的隧道涌水量预测

为实现隧道涌水量的高精度预测,该文以支持向量机为基础,通过核函数筛选和粒子群算法优化,构建了隧道涌水量的PSO-SVM预测模型.实例研究表明:支持向量机的核函数和惩罚因子对其预测精度具有较大影响,且Sigmoid型核函数的预测效果相对略优,粒子群算法也能很好的优化惩罚因子,进而达到提高预测精度的目的;同时,PSO-SV...     

基于加速混沌优化算法的支持向量机参数选择

针对支持向量机参数选择问题, 以惩罚系数、不敏感系数和RBF核函数中的宽度系数为优化变量, 采用Chebyshev映射代替Logistic映射产生初始混沌序列, 改变原有的搜索公式及增加3次载波, 提出了一种改进的加速混沌优化算法(ISCOA)。将该算法应用于人工数据集和实际数据集中, 并与常规的交叉验证法进行比较。试验结果表明: 在人工数据集中, 采用ISCOA在时间上缩短了至少23.43%, 精度上提高了至少6.31%;在实际数据集中, 预测值更接近实际值, 相对误差均控制在3.13%以下, 该算法具有较高的预测精度和寻优效果。     

基于最小二乘支持向量机的交通流量预测模型

城市交通流具有复杂性、时变性和随机性,实时准确的交通流量预测是实现智能交通诱导及控制的前提.综合分析交通流量影响因素的基础上,进行多路段的交通流量预测研究,提出了基于最小二乘支持向量机的交通流量预测改进模型,并应用平安大街的流量数据进行实例验证.结果表明,该模型具有学习速度快、跟踪性能好及泛化能力强等优点,在交通流预测中更具有实用性和推广性.     

基于支持向量机和神经网络的供应商选择方法比较

供应商选择是供应链管理的重要内容,近年来吸引大量学者进行研究,其中大量文献显示神经网络方法比传统统计方法有更大的优越性。然而神经网络具有固有的缺陷,如最优解的局部性、泛化能力低、训练样本大和无法控制收敛等。引用新的机器学习技术---支持向量机(support vector machines,SVM),用于选择理想供应商,并与BP神经网络算法相比较。实证表明,支持向量机算法比神经网络算法计算精确。     

交通事件小波分解与支持向量机声频识别方法

分析了现有交通事件自动检测和识别方法, 提出了应用小波分解与支持向量机相结合的交通事件声频识别方法。将车辆行驶的声音信号进行小波分解, 以不同频段的重构信号能量作为特征向量, 对由多个支持向量机构成的交通事件分类器进行训练, 并对正常行驶、刹车和碰撞事件的声音信号进行识别。试验结果表明: 利用车辆声音信号能够正确识别不同的交通事件, 识别准确率达95%, 识别方法可行。     

基于支持向量机的都市圈客运量预测模型

将都市圈客运量样本数据集分为训练集、测试集和检验集, 采用最小最终误差预测准则确定预测值的损失函数参数与惩罚因子, 选取ε-不敏感损失函数与高斯核函数减小预测复杂性, 构建了基于支持向量机的都市圈客运量预测模型, 并通过逐渐改变损失函数、惩罚因子与高斯核函数参数的取值, 对京津冀都市圈客运量进行了预测。预测结果表明: 客运量预测的平均相对误差为0.15%, 预测值与实测数据拟合良好, 整体变化趋势一致, 反映了预测模型的可靠性。     

基于支持向量机的大跨连续刚构桥可靠度分析

针对大跨连续刚构桥在可靠性分析过程中功能函数不能显示、表达的问题,提出了一种基于支持向量机分类功能可靠度的计算方法。采用拉丁超立方抽样产生一定样本的数据库,通过改变样本点中效应数据的大小,并结合有限元程序的分析,找出一定精度下离分类面最近的一对样本点,从而得到新的样本库。然后,用新的样本库构造出一个 SVM分类器函数。结合蒙特卡罗数值模拟,计算其失效概率。以四川某大跨连续刚构桥为例,验证了该方法的实用性。     

基于遗传优化支持向量机的交通事故预测

针对道路交通事故的预测问题,引入基于遗传优化支持向量机算法建立交通事故预测模型。利用遗传算法寻找支持向量机的最优参数组合,并用最优参数构建相应的支持向量机预测模型。仿真计算结果表明,基于遗传算法优化支持向量机模型优于传统的SVM模型,从而可以更有效地对道路交通事故进行预测。     

Multiple mental tasks classification based on nonlinear parameter of mean period using support vector machines

Mental tusk classification is one of the most important problems in Brain-computer interface. This paper studies the classification of five-class mental tusks. The nonlinear parameter of mean period obtained from frequency domain information was used as features for classification implemented by using the method of SVM （support vector machines）. The averaged classification accuracy of 85. 6% over 7 subjects was achieved for 2-second EEG segments. And the results for EEG segments of 0.5s and 5.0s compared favorably to those of Garrett＇s. The results indicate that the parameter of mean period represents mental tusks well for classification, Furthermore, the method of mean period is less computationally demanding, which indicates its potential use for online BCI systems.     

Intrusion detection model with twin support vector machines

Intrusion detection system （IDS） is becoming a critical component of network security. However, the performance of many proposed intelligent intrusion detection models is still not competent to be applied to real network security. This paper aims to explore a novel and effective approach to significantly improve the performance of IDS. An intrusion detection model with twin support vector machines （TWSVMs） is proposed. In this model, an etficient algorithm is also proposed to determine the parameter of TWSVMs. The performance of the proposed intrusion detection model is evaluated with KDD＇99 dataset and is compared with those of some recent intrusion detection models. The results demonstrate that the proposed intrusion detection model achieves remarkable improvement in intrusion detection rate and more balanced performance on each type of attacks. Moreover, TWSVMs consume much less training time than standard support vector machines （SVMs）.     

基于特征选择和SVM参数同步优化的网络入侵检测

为了提高网络入侵检测正确率,利用特征选择和支持向量机（SVM）参数间的相互联系,提出一种特征选择和SVM参数联同步优化的网络入侵检测算法.该算法首先将网络入侵检测正确率作为问题优化的目标函数,网络特征和SVM参数作为约束条件建立数学模型,然后通过遗传算法对数学模型进行求解,找到最优特征子集和SVM参数,最后利用KDD 1999数据集对算法性能进行测试.结果表明,相对于其他入侵检测算法,同步优化算法能够较快选择最优特征与SVM参数,有效提高了网络入侵检测正确率,加快了网络入侵检测速度.     

基于支持向量机响应面法的连续刚构桥可靠度

针对连续刚构桥结构可靠度分析过程中功能函数不能显示和功能函数拟合时试验样本过多等问题,提出了一种结合支持向量机回归功能与响应面法的可靠度算法。该算法将支持向量机的良好性能(如：处理小样本、高维及非线性)和响应面法的优点(如：直观和高效等)结合在一起。以某连续刚构桥为例,计算了该桥中跨挠度失效模式下的可靠指标。对该失效模式下可靠度参数的敏感性进行了分析,得出了有益的结论。     

A probability approach to anomaly detection with twin support vector machines

AbstractClassification of intrusion attacks and normal network flow is a critical and challenging issue in network security study. Many intelligent intrusion detection models are proposed, but their performances and efficiencies are not satisfied to real computer networks. This paper presents a novel effective intrusion detection system based on statistic reference model and twin support vector machines (TWSVMs). Moreover, a network flow feature selection procedure has been studied and implemented with TWSVMs. The performances of proposed system are evaluated through using the fifth international conference on knowledge discovery and data mining in 1999 (KDD’99) data set collected at MIT’s Lincoln Labs and the results indicate that the proposed system is more efficient and effective than conventional support vector machines (SVMs) and TWSVMs.     

Human machine collaborative support scheduling system of intelligence information from multiple unmanned aerial vehicles based on eye tracker

Many human-machine collaborative support scheduling systems are used to aid human decision making by providing several optimal scheduling algorithms that do not take operator’s attention into consideration. However, the current systems should take advantage of the operator’s attention to obtain the optimal solution. In this paper, we innovatively propose a human-machine collaborative support scheduling system of intelligence information from multi-UAVs based on eye-tracker. Firstly, the target recognition algorithm is applied to the images from the multiple unmanned aerial vehicles (multi-UAVs) to recognize the targets in the images. Then, the support system utilizes the eye tracker to gain the eye-gaze points which are intended to obtain the focused targets in the images. Finally, the heuristic scheduling algorithms take both the attributes of targets and the operator’s attention into consideration to obtain the sequence of the images. As the processing time of the images collected by the multi-UAVs is uncertain, however the upper bounds and lower bounds of the processing time are known before. So the processing time of the images is modeled by the interval processing time. The objective of the scheduling problem is to minimize mean weighted completion time. This paper proposes some new polynomial time heuristic scheduling algorithms which firstly schedule the images including the focused targets. We conduct the scheduling experiments under six different distributions. The results indicate that the proposed algorithm is not sensitive to the different distributions of the processing time and has a negligible computational time. The absolute error of the best performing heuristic solution is only about 1%. Then, we incorporate the best performing heuristic algorithm into the human-machine collaborative support systems to verify the performance of the system.     

Hepatic CT image query based on threshold-based classification scheme with gabor features

Hepatic computed tomography(CT) images with Gabor function were analyzed.Then a threshold-based classification scheme was proposed using Gabor features and proceeded with the retrieval of the hepatic CT images.In our experiments, a batch of hepatic CT images containing several types of CT findings was used and compared with the Zhao's image classification scheme, support vector machines(SVM) scheme and threshold-based scheme.