基于BP神经网络的碳含量对硬质合金性能影响的研究

Influence of Carbon Content on Properties of Hard Alloy Based on BP Neural Network

采用BP神经网络建立了关于碳含量对Fe-Cu-C合金性能的关系模型,研究并分析了含碳量和烧结温度对Fe-Cu-C合金性能的影响规律。结果表明:碳含量的增加会使合金内部的渗碳体和铁素体相互转化,烧结温度会严重影响晶粒大小和组织的分布。构建的BP神经网络能够很好的映射各参数对Fe-Cu-C合金性能的关系,预测精度高,计算稳定,具有良好的可靠性和推广意义。

The relationship modeling between the carbon content and the properties of Fe-Cu-C alloy was established by BP neural network. The influence rule of the carbon content and sintering temperature to the properties of Fe-Cu-C alloy were studied. The results show that the increase of the carbon content will make happen mutual transformation between cementite and ferrite of the alloy, the sintering temperature will seriously affect the grain size and the distribution of the microstructure. The BP neural network is good mapping on the relationship between the properties of Fe-Cu-C alloy and the parameters. It has a high prediction accuracy and stability calculation. Therefore, it has high reliability and promoting significant.