BP＋GA算法在再制造零部件精度分配中的应用

Application of BP Neural Network and Genetic Algorithm in Precision Assignment of Remanufactured Parts

再制造零部件精度参数的分配问题是关系到再制造机床性能和再制造成本的重要问题。本文提出了利用BP+GA算法优化分配再制造零部件的精度参数。利用BP神经网络建立零部件精度参数与再制造机床空间误差之间的非线性映射模型,用正交设计法得到训练样本数据并训练网络,再用BP+GA算法确定再制造零部件的精度参数。仿真结果表明了该算法是解决复杂精度分配问题的一种理想方法。

Precision assignment of remanufactured parts is an important problem which is related to the performance of remanufactured machine tools and the cost of remanufacturing. The algorithm of BP and GA was presented to assign the precision parameters of remanufactured parts. A BP neural network model was constructed to represent the relationship between the precision parameters of parts and the volumetric error of remanufactured machine tools. The training sample data were obtained with orthogonal design and were used to train the network. The precision parameters of remanufactured parts were determined by the algorithm of BP and GA. The simulated results show that the algorithm is an effective approach to solve the complex problem of precision assignment.