针对热轧板带的力学性能预报问题，利用人工神经网络中的BP算法建立原始化学成分和热轧生产的主要工艺参数与产品力学性能之间的关系，并开发出专门的应用软件。软件共分3大部分：数据处理部分、人工神经网络训练部分、运用成熟网络预报部分。本软件的特点是直观、方便、稳定。数据均从稳定生产的现场取得。采用此软件对SS400钢的性能进行预报，经过10万次训练后，产品力学性能的预报值与实际值拟合良好，预报结果的相对误差很小。

In order to predict the mechanical properties of the hot rolled strip, the BP algorithm of ANN was used to reflect the influence of the chemical composition and the main parameters of hot rolling on the mechanical properties of product and the corresponding software was designed. This software has three parts: data, training part of the ANN, predicting part by using the mature ANN. The characteristic of this software is convenient, intuitive and stable. There is an example of predicting mechanical properties of SS400 steel using this software. On the basis of the data obtained randomly from the Bao Steel 2050 hot rolling supervisor, through one hundred thousand times training, the predicted results are in good agreement with the measured values.