欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
由于水电机组监测数据量过大,基于神经网络的故障诊断存在网络结构复杂,训练时间长的问题,本文将粗糙集理论引入到水电机组故障诊断中,提出了基于粗糙集理论与RBF神经网络相结合的水电机组故障诊断方法。利用粗糙集理论在处理不确定信息方向的优点,在保持分类能力不变的前提下,去掉机组的冗余信息,保留必要的要素,并结合RBF神经网络对预处理后的信息进行诊断,使神经网络的输入神经元数目明显减少,其结构也得以简化,可以有效地提高故障诊断准确度。通过对实测机组振动数据进行诊断,证明了该诊断方法的有效性。
英文摘要：
Owing to massive data to be monitored for hydroelectric units, in order to solve the problems of structure complexity and long training time when neural network method is used for the fault diagnosis, rough set theory is introduced and the fault diagnosis method for hydroelectric units based on rough set ＆ RBF neural network is presented in this paper. Rough set is used in fault diagnosis first, which has the advantages on dealing with uncertain information, such as it can get rid of redundant factor and preserve important factor with the class capability unchanged; then RBF neural network is used and the preprocessed information is diagnosed. This method can not only decrease the number of the network input nerve cells obviously and simplify network structure, but also improve the accuracy of fault diagnosis. An example application proves that the proposed method is an effective method for the fault diagnosis of hydroelectric units.
同期刊论文项目
同项目期刊论文
位置：