基于卡尔曼滤波SVM图像鉴别方法研究

研究如何区分自然图像和计算机图形判别图像问题。由于人眼难以判别真伪图像,针对传统的图像鉴别方法由于参数的设置而造成识别准确率低,从而无法准确的区分自然图像和计算机图形。提出了一种卡尔曼滤波和支持向量机分类算法相结合的高性能图像真伪鉴别方法。算法首先采用卡尔曼滤波器提取图像的各个级别的分量,求解出各级分量和判别的误差值并作为特征数据,然后采用支持向量机分类算法对特征数据进行训练判别。仿真结果表明,提出的算法可以有效的判别图像的真伪,可以有效的识别出原始自然图像和合成图像,识别率高达99%以上,同时算法的稳定性较高,具有一定的实际应用价值。     

一种应用于城市环境中的车载自组网广播算法

针对城市环境中车载自组网信息发送面临的问题,本文提出一种可靠的车载自组网广播算法。首先,结合节点间的位置信息与移动方向将节点分类,基于距离机制选择中继节点,实现多跳广播;在十字路口处,为解决消息在各个路口进行广播,结合道路拓扑,将节点分为前向节点与后向节点两类;最后,针对车辆密度稀疏状况而出现通信空洞的问题,采用携带转发策略解决,节点主动探测空洞的发生,并验证消息发送方向上有节点进入通信范围,再次进行消息广播。仿真实验表明,算法在数据包延时、数据包成功发送率以及发送单一数据包平均负荷等方面具有较好的性能。     

信息融合技术在变压器油气识别故障诊断中的研究

变压器故障分为放电性故障和过热性故障两大类别,它们均会在变压器油中有所反映。本文通过对变压器油中主要气体的分析,判断变压器的故障类型。具体方法是:利用改进算法的BP网络和信息融合技术,以变压器油中五种主要特征气体作为神经网络的输入,以六种变压器状态作为相应的输出,通过加入动量因子,可以提高学习率系数,充分发挥改进算法的BP网络具有自适应学习能力的优势。仿真测试结果表明,本方法能够在较大范围内准确有效地进行变压器的故障诊断。     

基于多工况的车门结构多目标优化设计方法研究

车门结构的单一工况优化设计常难以满足车门设计要求。本文基于车门垂直刚度和一阶模态两工况,提出一种车门结构多工况多目标优化设计方法。以车门垂直刚度、一阶频率和车门质量为优化目标,通过均匀拉丁方试验设计和响应面方法得出车门垂直刚度和一阶频率的近似数学模型,并在此基础构建车门结构多目标优化模型。最后采用多目标遗传算法对其进行求解,最终优化结果在车门质量减少的情况下,垂直刚度和一阶频率均达到设计要求,达到预期优化效果。     

基于VS的分形图形绘制

根据牛顿迭代法生成分形图形的原理,在Visual Studio平台下实现了用窗口获得不同参数来实现同一程序绘制不同分形图形,解决了以往用C语言编写程序,特定程序只能绘制特定分形图形的问题。同时对牛顿迭代算法进行改进,加快了生成分形图形的速度。并采用对scale和stretch参数进行控制替代以往对xmax‘,xmin’,ymax‘,ymin参数的控制使得对绘制分形图形的精细程度的控制更加直观。     

Effects of added Cd on Cd uptake by oilseed rape and pai-tsai co-cropping

The experiment was designed to study whether the decrease of Cd in the rhizosphere of Zhongyouza Ⅰ,one breed of oilseed rapes （Brassicajunica） that can hyperaccumulate cadmium from the soil, can improve the living condition of less Cd-tolerant plant pai-tsai （Brassia chinensis） with their roots permitted to intermingle and develop coincident rhizosphere. The extent of rhizosphere interactions between Zhongyouza Ⅰ and pai-tsai was controlled by different root barriers, or without barrier. The results show that in the 10 and 20 mg/kg Cd treated soils, pai-tsai gets higher shoot mass and less Cd accumulation in its shoot than in the barrier treatments or in the mono cultures, when its roots are permitted to intermingle with those of Zhongyouza Ⅰ. Meanwhile, soil decontamination rates of Zhongyouza Ⅰ are not affected much by co-cropping with pai-tsai, they are 80.0% and 91.8% of that in the mono cultures of Zhongyouza Ⅰ, respectively. However, the co-cropping method in meliorating the living condition of pai-tsai is not obvious when Cd concentration in soil reaches 40 mg/kg, and soil decontamination rate decreases to 0.14, which is 58.3% of that in the mono culture. These results indicate that the oilseed rape Zhongyouza Ⅰ may alleviate Cd toxicity of surrounding less-tolerant species, and its ability of phytoremediation is not affected much at the same time, especially in the middle polluted soil.     

启发式联合PCD快速降噪算法

基于稀疏冗余表示,并行坐标下降(PCD)是最优秀的降噪算法之一。然而在音频信号处理中,当分割的帧数很大时,PCD的计算负担加重,造成运行时间剧增。该文基于联合稀疏表示(JSR)和同时稀疏近似(SSA),建立一种新的时域框架,并提出了一种启发式联合PCD算法(Joint-PCD)。在该框架中,每个音频帧作为一个列向量生成信号矩阵;利用超完备字典,Joint-PCD同步(同时)对信号矩阵降噪,提高了算法的运行效率,减小了运行时间。仿真结果表明:Joint-PCD算法不仅与PCD具有几乎相同的降噪性能,且把PCD算法的降噪速度提高了约5倍,极大地改进了PCD算法的收敛性能。     

Machine-Learning based methods in short-term load forecasting

Short-term load forecasting is of great significance to the secure and efficient operation of power systems. However, loads can be affected by a variety of external impact factors and thus involve high levels of uncertainties. So it is a challenging task to achieve an accurate load forecast. This paper discusses three commonly-used machine-learning methods used for load forecasting, i.e., the support vector machine method, the random forest regression method, and the long short-term memory neural network method. The features and applications of these methods are analyzed and compared. By integrating the advantages of these methods, a fusion forecasting approach and a data preprocessing technique are proposed for improving the forecasting accuracy. A comparative study based on real load data is performed to verify that the proposed approach is capable of achieving a relatively higher forecasting accuracy.     

基于有向同步交互图的类族测试用例生成算法*

根据类对象之间的消息通信产生的同步消息、消息传递以及消息等待的交互特征,设计了一种有向同步交互图(DSIG)模型,提出了类继承的同步交互路径(SIAPaths)测试用例生成方法,给出了自动生成同步交互路径的算法.同时,提出了同步交互序列测试充分性准则,并从基本交互路径序列和基本交互消息序列的测试覆盖率两方面阐述了测试的可行性.     

面向应用层的流媒体性能测量指标*

介绍了一种从流媒体接收者的角度对流媒体性能进行测量的方法,提出了相应的应用层评价指标。利用已有的商业化流媒体服务软件,通过改变网络环境在不同的条件下对指标进行测量,最后通过实验数据来验证这些指标的有效性。