【摘 要】
：
提出了一种新的RBF神经网络的设计方法,采用遗传-K均值聚类算法对RBF神经网络的隐层节点中心值进行优选,用遗传算法训练RBF神经网络的权值。以锅炉燃烧为实例,通过从现场采集
【机 构】
：
安徽建筑工业学院电子与信息工程学院 安徽合肥230022
论文部分内容阅读
提出了一种新的RBF神经网络的设计方法,采用遗传-K均值聚类算法对RBF神经网络的隐层节点中心值进行优选,用遗传算法训练RBF神经网络的权值。以锅炉燃烧为实例,通过从现场采集的数据建立神经网络模型,并用遗传算法寻找最优输入变量组合,实现锅炉燃烧优化。
A new RBF neural network design method is proposed. Genetic algorithm (K-means) clustering algorithm is used to optimize the hidden node center value of RBF neural network, and the weights of RBF neural network are trained by genetic algorithm. Taking the boiler combustion as an example, a neural network model was built from the data collected from the field, and the optimum input variables were found by genetic algorithm to optimize the combustion of the boiler.
其他文献
An efficient route for the synthesis of 5-(arylpropenylidine)-2,4,6-pyrimidinetrione 3 from an appropriate α,β-unsaturated aromatic aldehydes 1 and barbituric
The combination of fluorescent imaging and magnetic resonance imaging(MRl) techniques has attracted intense attention due to the complementary imaging ability o
Vacuoles are suggested to play crucial roles in plant defense-related cel death. During programmed cel death, previous live cel imaging studies have observed va
In multicel ular plant organs, cel shape formation depends on molecular switches to transduce developmental or environmental signals and to coordinate cel‐to‐
Pollen tubes usually exhibit a prominent region at their apex called the “clear zone” because it lacks light refracting amyloplasts. A robust, long clear zone
Regulation of actin dynamics is a central theme in cel biology that is important for different aspects of cel physiology. Vil in, a member of the vil in/gelsoli
根据黄土地区一般公路的实践经验,从路面表面排水、路基地面排水及综合排水系统三方面分析了一般公路排水系统存在的问题,并针对性地提出了防治问题的对策,以完善公路排水系
塔里木河地处塔里木盆地平原区,自身不产流,水源靠上游源流和开都河～孔雀河补给。随着源流区人口的不断增加,工农业用水逐年增大,源流来水不断减少,从20世纪70年代开始,中、下
本文介绍FOM及FOM组网的同时,具体阐述网络管理系统的设计思想,采用基于SNMP的网络管理系统设计方法-M/A模型,给出了软件的总体设计结构,针对管理站和代理,分别讨论了其数据
对基桩承载力作低应变辅助静载荷试验,同时结合静载荷试验以丰富检测分析的信息量,提高对静载荷试验曲线的分析能力,为基桩低应变检测桩身质量提供科学依据。