Fault diagnosis of rotating machinery based on SVM 2PTMC

A method is proposed to determine the priority class of binary tree SVM. Firstly, the vibration signals of rotating machinery are colleted through a rotating machinery fault experimental platform and data acquisition system. The signals are from 5 different conditions, i.e. rotor normal, rotor unbalance, rotor misalignment, rotor bearing inner ring cracks and rotor bearing outer ring cracks. Then the signals are disposed by zero-mean and the main frequency band of the vibration signals are reconstructed to, extract the dimensionless time domain as characteristic value. Finally, the priority class of SVM 2PTMC can be determined by the correct inspection rate of parallel SVM. Training samples can be completely divided in experiments, which verifies the effectiveness of this method.     

基于改进SVM分类器的动作识别方法

利用智能手机加速度传感器信号,提出一种改进的动作识别方法以降低传统动作识别方法的复杂程度,提高识别率。在特征提取时用盲选法,即用PCA(principal component analysis)进行特征值的降维和去除多维间的干扰,而所选特征没有对应的物理意义;并在分类识别中将遗传算法应用到SVM(support vector machine)分类器参数优化中。通过实验表明,该方法能够对日常的走路、站立、跑及上下楼等动作进行准确的识别。     

一种待开发的保健绿色食品新资源——鹿耳韭

介绍了一种保健绿色食品新资源———鹿耳韭,分别从生理特性、地域分布、营养食用保健功能等方面说明了其开发利用价值,并阐述了净菜、保健饮料、脱水菜、方便食品、果冻、蔬菜纸等的加工工艺。     

基于粒子群算法和支持向量机的黄花菜叶部病害识别

使用数字图像处理技术,以黄花菜叶部病害图像为识别对象,基于Lab空间和K-means聚类算法分割病害区域,提取目标区域的颜色特征、方向梯度直方图(histogram of oriented gradient,HOG)特征和形状特征,分别建立单一特征模型和特征融合模型,采用粒子群(particle swarm optimization,PSO)算法通过交叉验证优化支持向量机(support vector machine,SVM)模型的惩罚因子和核参数,建立基于PSO-SVM的多特征融合分类模型识别黄花菜病害。基于SVM的多特征融合分类模型识别率高于单一特征分类模型,识别率可达为81.67%;基于PSO-SVM多特征融合分类模型识别率高达92.39%。基于PSO-SVM的多特征分类模型识别率高,可以及时、便捷、高效地识别黄花菜病害。     

Support Vector Machines Based on Positive Feedback

Support vector machines based on positive feedback are put forward with the analysis of both advantages and disadvantages of current support vector machines based on Gaussian kernel function(including support vector machines based on K-nearest neighbors) and the essence why support vector machines is capable of describing data sets'distribution characteristics, thus the rigor constraint is overcome that maintains "corresponding parameters ofkernel function support vectors should be equal". The learning algorithm of support vector machines based on positive feedback is given. Simulation experiments of artificialand real data proves that support vector machines based on positive feedback is be obviously superior to current ones in its generation capabilities. Though, only the support vector machines based on positive feedback for pattern recognition is discussed, the idea included in support vector machines based on positive feedback is using kernel functions with different corresponding parameters to construct support vector machines, and is adaptive to other types support vector machines.     

枇杷核淀粉提取工艺的研制

以枇杷核为原料,通过筛分、离心等工序,将枇杷核中的淀粉与纤维、蛋白质及其杂质分离,提取淀粉过程进行研究。确定淀粉提取最佳工艺参数,磨浆配水比1：2,浆料pH6．0,提取温度20℃．亚硫酸钠添加量0．15％,干燥温度50～80℃,干燥时间1．5-2h。在此条件下,淀粉回收率达84．67％．白度88.30。     

Microarray data classifier with dimensionality reductionproximal support vector machines

DNA microarray technologies have changed the foreground of biological medicine, while the generated plentiful data is the key problem for the application of microarrays．Microarray data have the characteristics of large quantity, low sample size and high gene dimensionality. A microarray data classifier with dimensionality reduction proximal support vector machines (DRPSVM). A dimensionality reduction quadratic programming algorithm is used in DRPSVM, which shows faster training speed and smaller memory requirements than traditional PSVM does. Using CAMDA2000, colon 1 dataset and colon 2 dataset as the experimental datasets, the classification performance of DRPSVM is compared with those of BP, Nearest, RBF and SVM. DRSVM shows stable classification performance, existing one and only optimal solution and fast training which is suitable for DNA microarray data classification applications.     

On a New Lower Order Exact Penalty Function

For the inequality constrained nonlinear programming problem, the antnors propose a new lower order penalty function, which combines the classical l1 penalty function and the lower order exact penalty function introduced in thoy some papers, and then illustrate that it has the merits of the two penalty functions. In additional, we propose a solution method for solving the inequality constrained nonlinear programming problem and prove its global convergence.     

Analysis on the Optimal Operation in Hydroelectric Station of Gezhouba

The optimal operation in hydroelectric station unit is me opumum problem,which include unit commitment,load distribution and unit maintenance. The authors research the optimal operation in units in hydroelectric station of Gezhouba by the genetic algorithm and dynamic programming to satisfy the demand of economic operation of Gezhouba. They arrange and organize the program of the units in r6ason to go for the better benefit and satisfies security, credibihty and high quality of electric production . These methods can work out the most optimum result. When the head of water and the flux is confirmed they can distribute the flux to the units to realize the optimal operation of the scheduling in Gezhotlba.     

A NEW CLASS OF AUGMENTED LAGRANIAN FUNCTION FOR NONLINEAR PROGRAMMING

In this paper, a new class of augmented Langrangian function for solving optimization problem with twice continuously differentiable inequality constraints is studied. Under mild assumptions on the parameters of augmented function, the equivalence of stationary points of the function and K-T points of the restricted problem is proved as well as their extreme points.     

基于KPCA-WLSSVM的建筑能耗预测模型

为降低建筑能耗影响因素间复杂相关性对模型性能的影响,建立了一种基于KPCA-WLSSVM的建筑能耗预测模型。利用核主元分析(KPCA)对输入变量进行数据压缩,消除变量之间的相关性,简化模型结构;进一步采用加权最小二乘支持向量机(WLSSVM)方法建立建筑能耗预测模型,同时结合一种新型混沌粒子群-模拟退火混合优化(CPSO-SA)算法对模型参数进行优化,以提高模型的预测性能及泛化能力。通过将KPCA-WLSSVM模型方法应用于某办公建筑能耗的预测中,并与WLSSVM、LSSVM及RBFNN模型相比,实验结果表明,KPCA-WLSSVM模型方法能有效提高建筑能耗预测精度。     

The Approximate Method of Quadratic Function With Mixed Constraint

A reduced method to solve the problem with equality constraints is given. Based on the reference we use nonlinear equality constraints, the approximate algorithm is obtained. The nonlinear programming of quadratic function with equality and unequality constraints is discussed by Wang kairong. We transform unequality constraints into equality constraints by add to relaxation variable, the quadratic programming is obtained by linear approximation, and approximate calculation is done by means of reduced method. The result of numerical calculation shows the method is feasible.     

An Investigation of Electrochemical Properties of Gelled Polymer Electrolytes

According to the two kinds of expressions of the quadratic functions and the power functoin for describing the non_linear seepage flow,this paper advances that the two kinds of expressions should have identity,that is to say they can transform each other.And by means of least square method,the relational expression for the two kinds of functions to transform each other is derived.From the verification of the two kinds of particular seepage flow(laminal flow and adequate turbulent flow)and the hypothetic data,it has been shown that the relational expression can express the identity between them adequately,it can afford the references to the further research on the non_linear seepage flow.     

Multi classification Method of GA SVM on Identifying Grade of Expansive Soils

Aiming at the fact that parameters in support vector machine(SVM) model were difficult to be identified, a genetic algorithm SVM(GA SVM) was proposed to avoid the blindness in parameter choosing and improve the estimation ability of SVM, in which the parameters in SVM and kernel function were searched by genetic algorithm. And it was then applied to the classification for the swell and shrink grade of expansive soils. Five indexes including liquid limit, total swell shrink ratio, plasticity index, water contents and free expansive ratio were adopted as discriminated factors. And the four grades of the expansive soils were the outputs correspondingly. Classification function was obtained through training a large set of expansive samples. And it was shown that the classification method of GA SVM was effective and with high accuracy.     

基于支持向量机方法的森林火险预测研究

以利用气象数据预测森林火险等级为目的,使用半正定规划建模选定支持向量机的最优核函数,依据500余条林火数据建立了支持向量回归机模型,并使用回归误差特征曲线图对比分析各个回归模型的学习效果。分析得到该自定义核函数的均方误差为1.76,支持向量数为190,约占训练数据集的1/2。结果表明,与传统的线性回归方法及基于高斯核的支持向量机相比,该预测模型具有较高的准确率并且有效的避免了过学习的现象。     

采空区危险性的支持向量机识别

针对采空区危险性影响因素与其危险性等级之间存在着复杂非线性关系的特点,笔者提出采用支持向量机最优分类理论来识别采空区的危险性等级。研究选取岩体结构、地质构造、岩石抗压强度、弹性模量、采空区形状、矿体倾角、高跨比、空区体积等8个参数作为主要影响因素,根据支持向量机理论,提出了1-V-1的采空区分类算法,并在Matlab中编程,建立了分类预测的SVM模型。以某矿山的实测采空区为例,利用该模型进行了识别,并与BP神经网络预测结果作对比。实例研究表明,采用该方法的分类结果比神经网络更准确,与采空区调查结果一致性好,用支持向量机理论进行采空区危险性评价是可行的。     

A Sort of Method of State Space Search in Mechanical Translation

There is a kind of combination of correlated objects in mechanical translation.When match the combination with the rule,the searching solution becomes un-possible by the"combinatorial explosion".A hierarchical intelligence search method is advanced.According to the prior knowledge,the data which do not satisfy the correlation constraint will be removed by the closed-loop eliminate method at first and then only a simple search in order is needed for obtaining the solution of the problem.This method refrains from any backtracking.The time and space consumed by search are greatly reduced.