螺杆转子型线的逆向设计研究

Reverse design method of compressor screw rotor profile

针对螺杆式空气压缩机转子端面型线的复杂性及其设计所存在的问题,提出一种基于BP神经网络及逆向工程相结合的转子型线逆向设计方法。首先通过对转子的三维数字化测量、点云数据处理、转子曲面重构等技术,获取了转子的三维实体模型;然后通过对转子端面型线的研究,设计了适合于螺杆转子型线拟合的BP神经网络,最后将测试数据输入经过训练的BP神经网络中,测试了网络的拟合性能。结果表明:利用逆向工程技术可以有效地提取螺杆转子的端面型线,采用BP神经网络拟合转子端面型线可以获得比其他拟合方法更小的误差。研究结果可为压缩机螺杆转子型线的优化设计奠定基础。

In view of the complexity of the compressor screw rotor end face profile and the problems in its design,the method for the reverse design of rotor profile based on BP neural network and reverse engineering is proposed.The 3D solid model of screw rotor is obtained by means of technology such as 3D digital measurement,point cloud data processing,rotor surface reconstruction.Then,BP neural network suitable for screw rotor profile fitting is designed after the researching on end face profile.Finally,the fitting performance of the network is tested by entering the test data into the trained BP neural network.The results show that reverse engineering can effectively extract the end face profile of the screw rotor,and using BP neural network to fit the rotor end face profile can obtain smaller error than other fitting methods.The results can provide reference for the profile optimized designing of compressor screw rotor.