基于广义回归神经网络的凝汽器故障诊断

Fault Diagnosis of Condenser Based on Generalized Regression Neural Network

凝汽器的运行状态,将直接影响机组运行的安全性与经济性。为了提高凝汽器故障的诊断速度与正确率,利用广义回归神经网络,建立了凝汽器故障的诊断模型(GRNN),经实例测试,验证了诊断模型的有效性。并且,将该诊断模型(GRNN)与其它神经网络诊断模型(Elman、BP)的诊断结果进行了对比,实验表明,利用广义回归神经网络的诊断速度更快,故障的剥离能力较强,更适合于现场故障的诊断。

The performance of condenser directly affects the safety and economy of the unit. In order to improve the diagnosis speed and correct rate of the condenser fault, the generalized regression neural network is used to establish the fault diagnosis model of the condenser, and the validity of the model is verified by the fault example test. The GRNN model is compared with the diagnosis results of Elman and BP neural network diagnosis model. The experimental results show that the GRNN neural network has faster diagnosis speed and better fault stripping ability, and is more suitable for the fault diagnosis.