机械密封环端面热变形的分析与BP网络预测

分析了影响机械密封环端面变形的主要因素，说明了密封环对介质的传热系数及端面热通量的计算方法.对于高速重载机械密封，因密封环端面热通量大，其热变形显著高于力变形.当密封环模型一定时，按正交设计法给出若干端面热通量值作为输入样本，并将有限元法计算得出的端面内径和外径处的相对轴向变形Δz作为输出样本，训练3层BP人工神经网络，对该密封环端面的轴向热变形进行精确预测.正交设计法和人工神经网络方法相结合是密封环端面热通量与热变形之间高度非线性关系分析的有效方法.

Analysis of thermal deformation of end face of mechanical seal ring and forecast based on BP ANN

The major factors, which correlate with the end face deformation of the mechanical seal ring, were analyzed.The method to calculate the coefficient of heat transfer between seal ring and medium，and the method to calculate the heat flux exerted on the end face were presented.For the high speed and over-load mechanical seal,thermal deformation is much more remarkable than the deformation caused by pressure.For a given seal ring model, it is possible to train a three-layer back propagation artificial neural network (BP ANN)to accurately forecast the relative thermal deformation Δz between outer diameter and inner diameter on the end face of the seal ring.The input samples were the heat flux values exerted on the end face of the seal ring and giren by the orthogonal design method.The output samples were the values of Δzthat were calculated by the finite element method.By means of BP ANN and the orthogonal design method, the high degree nonlinear relationship between heat flux and thermal deformation of the end face of the seal ring could be analyzed satisfactorily.