Real-time Performance of Linux System

The standard Linux is a typical time-sharing system, it has a poor real-time performance. But according to the applications of the Linux OS are expanding on real-time domain, it becomes necessary to enhance the performance of the Linux system. This paper analyses the schedule policies and schedule algorithms of the Linux kernel, and discusses several different kinds of methods about enhancing real-time performance of the Linux system from the total Linux system. They authors summarizes the current research's developments of the Linux system.     

Automatic Fingerprint Classification by GA-Based Neural Network

Fingerprint classification can provide an important indexing mechanism in a fingerprint database. An accurate and consistent classification can greatly reduce fingerprint matching time for large database. In the paper, by combining genetic algorithm and neural network is presented a fingerprint classification algorithm which is able to achieve an accurate classification. By inputting the global feature represented by directional image to three layer neural network trained by genetic algorithm, the fingerprints were classified into six categories: whorl, right loop, left loop, arch, double loop and undiscerning type successfully.     

Learning the BN Structure Based on the Sum of Mutual Information

The Crossing Entropy is defined to scale the similar level of two probability distribution. In many papers on learning BN structure,the Crossing Entropy was used as an indicator of measuring the learning accuracy of an algorithm.The known scoring metrics for learning BN structure is analyzed in this paper,then a new scoring metrics Sum of Mutual Information is proposed based on the information theory.At last,two algorithm for learning BN structure by SIM is represented.     

Avoiding Obstacle Planning for Robots on Molecular Optimization Algorithm

Interest in DNA computing has increased overwhelmingly since Adleman successfully demonstrated its capability to solve Hamiltonian Path Problem. This article introduces the improving method in virtue of the biological thery of DNA technology, a new molecular algorithm is advanced. After a numerical simulation, the result shows that it avoids the prematurely and lower convergent speed of the classic genetic algorithm, and inherits global search capability, the validity and the speed of the genetic algorithm have been increased. The best result can be obtained in few iterative times. It is fit for solving path planning problem.     

基于混沌算法的工程车辆自动变速神经网络控制

为了克服BP神经网络收敛速度慢和易陷入局部极小点的不足，提出了一种基于混沌算法的工程车辆自动变速神经网络控制方法，并利用ZL50型装载机传动系统换挡控制试验的数据进行了验证性仿真试验。仿真结果表明：该方法对于车辆自动换挡智能控制是可行的，可以根据操作工况实现正确的变速挡位决策。     

泵站竖井进水流道数值模拟与装置特性试验

采用雷诺平均纳维—斯托克斯方程(RANS)和标准kε湍流模型,运用SIMPLEC算法,对椭圆型线前置竖井贯流进水流道内的流动进行了三维数值计算,揭示了竖井进水流道内流场、特征断面的速度分布及水力损失规律。对椭圆型线竖井进水流道与出水流道匹配组成的模型泵装置进行动力特性试验,得出了4种叶片角度下的泵装置特性曲线。研究结果表明:采用椭圆型线设计的竖井进水流道合理;大型低扬程泵站采用卧式竖井贯流结构,泵装置效率较高。     

基于随机森林回归算法的小麦叶片SPAD值遥感估算

使用机器学习中的随机森林(RF)回归算法构建小麦叶片SPAD值遥感反演模型。以2010—2013年江苏地区试验点稻茬小麦3个生育期(拔节、孕穗、开花)的叶片为材料,结合我国自主研发的环境减灾卫星HJ-1对研究区域进行同步监测,分析了各生育期叶片SPAD值与8种植被指数间的相关性;以0.01水平下显著相关的植被指数作为输入参数,使用RF回归算法构建了每个生育期的小麦SPAD反演算法模型,即RF-SPAD模型,以支持向量回归(SVR)和反向传播(BP)神经网络算法构建的SVR-SPAD模型和BP-SPAD模型作为比较模型,以R2和均方根误差(RMSE)为指标,分析了每个生育期3个模型的学习能力和回归预测能力,结果表明:RF-SPAD模型在3个生育期都表现出最强的学习能力,R2和RMSE在拔节期分别为0.89和1.54,孕穗期分别为0.85和1.49,开花期分别为0.80和1.71;RF-SPAD模型在3个生育期的回归预测能力都高于BP-SPAD模型,高于或接近于SVR-SPAD模型,R2和RMSE在拔节期分别为0.55和2.11,孕穗期分别为0.72和2.20,开花期分别为0.60和3.16。     

联合收获机前进速度模糊控制系统多目标遗传优化

针对联合收获机前进速度模糊控制系统参数设计具有主观性,导致作业性能不理想的问题,以切纵流联合收获机样机为研究对象,构建联合收获机前进速度模糊控制系统仿真模型,并建立控制性能和收获性能的优化目标函数来衡量联合收获机的作业性能,利用多目标遗传算法对模糊控制系统的隶属函数和输送槽、割台螺旋输送器和切流滚筒对前进速度的影响因子进行优化。通过对优化前后模型仿真对比和两组田间试验数据分析,表明优化后的控制系统在受外界干扰情况下控制性能得到了较好的保持,收获性能在喂入量变化不大情况下,单位平均损失率分别由1.45%和1.26%降至1.12%和1.14%,联合收获机的总体作业性能得到改善。     

油桃外部缺陷的高光谱成像检测

采用高光谱(420~1 000 nm)成像技术对"中油9号"油桃的4种外部缺陷(裂纹果、锈病果、异形果和暗伤果)进行检测判别。对400个样本(4种外部缺陷样本和完好样本)运用偏最小二乘回归(PLSR)从全波段中分别提取了10条特征波长,分别为497、534、657、677、696、709、745、823、868、943 nm。缺陷样本的高光谱图像经过主成分分析后,对876 nm下的单波段图像通过掩膜、Sobel算子处理,并对主成分图像经过区域生长算法实现缺陷样本的缺陷区域分割。对光谱数据进行主成分分析得到前10个主成分值,并对图像数据采用灰度共生矩阵(GLCM)提取得到6项图像纹理指标(均值、对比度、相关性、能量、同质性、熵值)。将主成分值和纹理值融合建立极限学习机(ELM)模型对油桃外部缺陷进行检测判别。结果表明,该模型对缺陷样本的判别正确率为91.67%,完好样本的正确率为100%。     

马蹄形过水断面正常水深计算的遗传算法

马蹄形过水断面是无压输水隧洞中常采用的断面形式之一 ,因其几何形状复杂 ,水力计算困难。目前常采用的方法有试算法、图解法和迭代法 ,这些方法由于计算量大且精度不易保证 ,不便于在生产实践中应用推广。将马蹄形过水断面正常水深计算转化为非线性优化问题 ,采用实数编码的遗传算法进行求解 ,实例计算表明该方法精度高、通用性强、操作简便 ,便于在生产实践中应用。