应用BP神经网络预测热处理温度对高钒高速钢中残余奥氏体含量的影响

Prediction for the Effect of Heat Treatment Temperature on Residual Austenite Content of High Vanadium High Speed Steel using BP Neural Network

对含钒10 %的高速钢,利用铁磁性法测量了经90 0℃～110 0℃淬火、2 5 0℃～6 0 0℃回火后其残余奥氏体含量。基于测量的实验数据,利用BP神经网络建立了残余奥氏体含量与热处理温度的非线型关系模型。结果表明:良好训练的BP网络模型可以较准确预测不同淬火、回火温度条件下残余奥氏体的含量。预测结果揭示了淬火、回火温度对残余奥氏体含量的影响规律,为生产中优化热处理工艺、控制残余奥氏体含量提供了一种新的方法。

The residual austenite contents of a high vanadium high speed steel (V=10%) quenched from 900°C to 1100°C, and tempered from 250°C to 600°C were measured by ferromagnetism method. By the use of backpropagation (BP) network, the non-linear relationship between the residual austenite contents (A) and quenching temperature, tempering temperature (T1, T2) is established on the basis of the experimental data. The results show that the well-trained BP neural network can predict the residual austenite contents precisely according to quenching temperature and tempering temperature. Therefore, a new way of optimizing heat treatment technology and controlling residual austenite content is provided.