摘要
介绍基于遗传算法BP神经网络建立的热轧带钢力学性能预报模型,对某带钢厂产品的力学性能预测结果:与检测数据对比,屈服强度、抗拉强度、延伸率最大误差分别为2.6%,2.4%,3.1%,并提出改进措施。
The forecasting model for mechanical properties of hot-rolled strip based on BP neural network and genetic algorithm was introduced; the forecasting results of product mechanical properties from a certain plant showed that: compared with the inspection data, the max. errors of forecasting yield strength,tensile strength and elongation was 2.6%, 2.4% and 3.1% respectively, hence, the corrective measures were put forward.
关键词
遗传算法
BP人工神经网络
热轧带钢
力学性能
预报模型
Genetic Algorithm
BP Artificial Neural Network
Hot-rolled Strip
Mechanical Prop-erty
Forecasting Model