基于双隐层神经网络的边界膜强度计算模型

The double hidden layer neural network used in calculation model of the boundary film strength

本文用双隐层ＢＰ人工神经网络建立了丝杆螺母副表面边界膜温度特性的磨损自补偿数学模型。该模型可用于准确地预测温度对边界膜强度的影响。并采用Ｌ－Ｍ规则进行神经网络学习训练可使网络收敛快，误差小。网络输出结果与实验结果比较有极好的吻合性。该神经网络可为工程设计人员在摩擦学设计时提供有效的计算工具。

The BP neural network used in the temperature characteristic of the boundary film on the screw-nut pairs surface in the wear-self-compensation system was established.The network could predict the effects of the temperature on the boundary film strength.the network trained with the L-M rule were quick in convergence and small in error.The output of the network were precise by comparison with the test results.The network model could provide the calculable measure for the tribology design engineers.