基于混合神经网络的风机性能监测模型

Monitoring model of fan performance based on hybrid neural network

针对传统的RBF神经网络泛化能力差的缺点,利用RBF神经网络强大的非线性逼近能力和数学模型良好的外推能力,提出了一种将传统的RBF神经网络和用偏最小二乘法建立的通风机性能数学模型相结合的混合神经网络模型,并将该模型用于通风机的重要性能参数——流量的监测上。以实验室4-73No.8D离心风机为研究对象,用不同导流器开度下的实验数据进行拟合,研究结果表明,混合神经网络模型的泛化能力强,精度高,各项模型评价参数均优于传统的RBF神经网络模型。

Due to the limitation of RBF neural network's generalization ability, an improved fan performance model in the flux monitoring is proposed based on hybrid neural network. It makes use of the strong nonlinear approaching of RBF neural network and partial least squares (PLS) model. Performance curves of 4-73No. 8D were approached with experimental data of different opening angles of guider. The results show that the hybrid neural network model has better generalization ability and higher precision, and it is superior to normal RBF neural network model.