论文部分内容阅读
采用等离子技术对硼铸铁进行熔凝处理,利用SEM、XRD及显微硬度计对硬化层的组织和性能进行了测试和分析。在此基础上,采用BP人工神经网络建立等离子工艺参数与硼铸铁熔凝硬化层性能之间的神经网络预测模型。结果表明,熔凝层的组织为细小均匀的共晶莱氏体+少量未溶石墨,神经网络预测的硬化层深度和硬度值与试验值相对误差小于4.3%,说明该BP神经网络模型可以较准确预测硼铸铁等离子熔凝硬化层的性能。利用该模型可为实际生产中选择合适的工艺参数提供参考。
Boron cast iron was melted by plasma technique. The microstructure and properties of hardened layer were tested and analyzed by SEM, XRD and microhardness tester. On this basis, BP artificial neural network is used to establish the neural network prediction model between the plasma process parameters and the properties of the molten hardened layer of boron cast iron. The results show that the microstructure of the fused layer is fine and uniform eutectic ledeburite + a small amount of undissolved graphite. The relative error between the predicted depth and the hardness value of the hardened layer and the experimental value is less than 4.3%, which shows that the BP neural network model can compare Accurately predict the performance of boron cast iron plasma fusion hardened layer. The model can be used to select suitable process parameters for actual production.