基于绕封模型的故障特征选择方法研究

在机械故障诊断中,基于原始大特征量的故障状态识别会导致识别精度的下降。特征选择可以去除原始特征中的冗余特征,提高诊断精度。但以前广泛应用的基于过滤模型的特征选择方法不能满足进一步提高精度的要求。针对此问题,提出使用基于绕封模型的故障特征选择方法,它采用遗传算法对特征集寻优,样本划分法进行错误率预测估计和BP神经网络学习算法进行分类。轴承诊断实例证昵,此方法有较好的寻优特征子集的能力,可以提高系统的诊断精度。     

基于改进的遗传算法神经网络控制器

基于遗传算法的神经网络控制器,对神经网络中连接权值采用实值编码,通过随机选择,可变变异率和Elitism方法形成改进遗传算法.以倒立摆系统神经网络控制器参数寻优为例,对二进制编码和实值编码的遗传算法进化过程进行仿真,结果表明本算法寻优效果优,在多参数、多目标、多约束条件下,寻优效果更佳.     

X射线管道焊缝探伤中的视觉同步技术

研究了一种基于机器视觉的同步技术。实现同步的关键在于检测管内外的转角差，通过对X射线图像进行特征提取获取管内外转角差。在特征提取时，先对图像进行形态学滤波，然后用遗传算法（GA）进行图像分割，最后用Hough变换从图像中提取特征量。试验结果表明，该算法从X射线图像中准确检测出管内外的转角差，且检测速度超过了25fps的实时性要求。该算法再结合一定的控制规律控制伺服电机，即可实现管内外同步旋转。     

用遗传算法优化任意稀布率的平面阵列

通过给经典遗传算法(GA)的变异算子赋予新的功能,实现了任意稀布率的平面阵或线阵方向图的优化.研究了遗传算法参数和优化结果之间的关系,仿真结果表明,同时考虑峰值副瓣电平和波束宽度才能获得更满意的稀布结果.归纳得出一个平面稀布阵峰值副瓣电平的估计公式,并和随机稀布阵的公式进行了比较,数值结果表明该公式有一定的工程意义.     

GA-BP神经网络在雷达目标跟踪中的应用研究

目标的运动过程一般是非线性过程,神经网络具有很强的非线性拟合能力,可映射任意复杂的非线性关系,以神经网络为基础的模型能够很好地反应目标的非线性运动趋势。本文在分析传统的BP神经网络的基础上,引入GA遗传算法来优化神经网络的初始权值和阀值,同时将GA-BP神经网络模型运用于对雷达目标的跟踪过程中,并通过仿真验证该模型的精度较高。     

微真空条件下赤霉素处理对西兰花保鲜效果的影响

为探讨微真空贮藏条件下不同浓度赤霉素(GA3)处理对西兰花保鲜效果的影响,分别以80 mg/L、100 mg/L、120 mg/L的赤霉素溶液浸泡处理采后西兰花5 min,以蒸馏水浸泡西兰花5 min为对照试验,晾干后贮藏于温度为3±0.5℃、真空压力为400-500torr的微真空贮藏条件下,研究不同浓度赤霉素处理对西兰花保鲜效果的影响。结果表明:三种浓度的赤霉素处理均可在不同程度上降低西兰花PPO、POD活性,抑制叶绿素的降解及Vc含量的下降,减少膜脂过氧化产物MDA的积累,其中浓度为100 mg/L的GA3处理对延缓西兰花采后衰老的效果最佳。     

A novel hybrid genetic and imperialist competitive algorithm for structure extraction of woven fabric images

The woven fabric is a flexible object and to specify its parameters, applying inflexible and ordinary methods of image processing ever have considerable errors. In this regards, proposing an adaptable method to fabric image properties is concentrated to detect the yarns position. In this research, a flexible algorithm is proposed containing two stages: first, the inexact ranges of fabric parameters are determined by preprocessing colored fabric images using wavelet transform and clustering methods. Then, the hybrid genetic and imperialist competitive algorithm is applied to optimize the obtained ranges and detect the yarns position. To achieve better results, the parameters of the hybrid ICA–GA are calibrated using the Taguchi method. Results indicate that in this new method, the error value of detecting structural fabric parameters has considerably decreased to 5% as compared with common gray-scale projection method. The proposed method is capable of detecting the exact yarns position in colored fabric images with uneven color intensity and low-density weave with mean precision value of 96.2%. In the fabric images with high density weaves, the mean precision value is more than 94.72%.     

Variable selection for partial least squares analysis of soluble solids content in watermelon using near-infrared diffuse transmission technique

This work is focused on the variable selection in building the partial least squares (PLS) regression model of soluble solids content (SSC) that is used to evaluate quality grading of watermelon. The spectra were obtained by the near infrared (NIR) spectrometer with the device designed for on-line quality grading of watermelon and the spectra of 680–950 nm were adopted to analysis. The variable selection was based on Monte-Carlo uninformative variable elimination (MC-UVE) and genetic algorithm (GA). In comparison of the performances of the full-spectra (680–950 nm) PLS regression model and the feature wavelengths PLS regression model showed that the MC-UVE–GA–PLS model with baseline offset correction combined multiplicative scatter correction (MSC) pretreatment was much better and 14 variables in total were selected. The correlation coefficients between the predicted and actual SSC were 0.885 and 0.845, the root mean square errors were 0.562 °Brix and 0.574 °Brix for calibration and prediction set, respectively. This work can make a great contribution to the research of on-line quality grading for watermelon nondestructively.     

基于遗传神经网络的城市用水量预测研究

介绍了BP(误差反向传播算法)和GA(遗传算法)及GA-BP 3种神经网络,并以此分别对城市用水量进行预测.实验结果表明,基于GA-BP算法的神经网络方法应用于城市用水量的预测问题,能采用遗传学习算法优化BP神经网络模型的初始权重,即先利用遗传学习算法进行全局训练,再用BP算法进行精确训练,使网络收敛速度加快和避免局部极小.GA-BP神经网络在收敛速度和预测精度等方面均优于BP和GA网络,从而为未来短期城市用水量负荷的准确预测提供了新的思路与方法.     

一种综合咖啡智能拼配方法研究

为了提高综合咖啡拼配质量与效率,研究了一种基于交互式遗传算法(IGA)的咖啡智能拼配方法。该方法将拼配方案作为样本个体,采用改进的自适应交叉和变异概率,交互界面提供10项评价指标用于个体适应值。研究由5位用户参与优化,结果显示该方法界面友好、操作简便,可以在4~5代内获得一致性满意解,且各组分比例差异均小于5%,适应值均超过85。以传统杯测拼配法为参照试验,结果表明,采用IGA优化方法超过传统拼配方案水平。基于IGA在进化优化中的成功应用,为咖啡拼配提供了一种智能化新思路。