摘要
对燃气轮机润滑油系统常见故障原因进行了分析,并结合专家知识建立了润滑油系统故障知识库;在此基础上将RBF人工神经网络引入燃气轮机装置故障诊断中来,由于采用了一种动态RBF网设计方法,使得神经网络的规模较小同时具有较高的泛化能力,提高了神经网络的诊断速度及准确性。
It is important to set up a fault diagnosis system according to the frequent faults in lubricating oil system of gas turbine. Radial Basis Function Neural Network（RBFNN）has the character of optimal approximation and global approximation, we introduced dynamic design method to the structure and parameters of RBFNN to makes the neural network have a smaller size and higher generalization ability. The diagnosis speed and accuracy are also improved.
出处
《燃气轮机技术》
2006年第1期58-60,54,共4页
Gas Turbine Technology
