改进的BP网络在凝汽器系统故障诊断中应用  

Application of Improved BP Algorithm in Failure Diagnosis of Condenser

作　　者：张利平[1] 王铁生[1] 翟兆江[1] 

机构地区：[1]华北水利水电学院,河南郑州450011

出　　处：《华北水利水电学院学报》2006年第1期58-60,共3页North China Institute of Water Conservancy and Hydroelectric Power

摘　　要：结合牛顿非线性优化方法,提出了一种改进的BP网络学习算法,将其应用于凝汽器系统故障诊断中,列举了凝汽器典型故障集和征兆集,建立了凝汽器典型故障征兆模式知识源.实例证明:该算法收敛速度快、计算精度高.A new network training method,based on quasi-Newton nonlinear optimization method, is put forward to overcome the deficiency of slow speed convergency and local minimum. It is applied to realize failure diagnosis of condenser. The result of verification shows that it has fast convergence speed and high precision.

关 键 词：BP神经网络 凝汽器 故障诊断 

分 类 号：TK223.35[动力工程及工程热物理—动力机械及工程]