摘要
为解决柴油机故障诊断这一复杂问题,提出了一种基于智能互补融合的智能诊断方法。采用蚁群算法(ACA)对反映运行工况的特征参数进行属性约简,剔除不必要的属性。根据约简结果,建立了基于径向基函数(RBF)神经网络的故障诊断系统。网络的训练对比结果表明,基于蚁群算法的约简处理简化了输入神经网络的数据维数,提高了网络的训练效率和故障分类准确性。
A new AI method, the hybrid of ant colony algorithm （ACA） and neural networks, was put forward to solve the fault diagnosis of diesel engine. The ant colony algorithm is used to simplify attribute parameter reflecting operating conditions of diesel engine and in which unnecessary attributes are eliminated. According to the reduction result, the fault diagnosis system based on RBF neural networks was produced. Through the comparison of fault classification effect, it is shown that the new method reduces the dimension of input to neural network, raises the training efficiency and the fault classification accuracy.
出处
《机床与液压》
北大核心
2007年第7期241-244,共4页
Machine Tool & Hydraulics
基金
山东省科技攻关项目(项目编号:2006129)
