针对产品的性能要求制定合理的热轧工艺，提出将组织性能预测与控制技术应用于热轧工艺的优化设计。基于大量生产数据，建立了包含10个BP神经网络的模型组以描述化学成分、工艺和力学性能的对应关系，屈服强度、抗拉强度和伸长率的预测精度分别达到了±6%、±6%和±4%。结合多目标粒子群优化算法，针对客户对性能的需求，在化学成分和工艺约束已知的条件下，对热轧工艺进行了优化计算。工艺优化计算结果与现场生产数据吻合良好，验证了工艺优化设计的有效性，从而为热轧最优工艺设计提供指导。
In order to design reasonable hot rolling process according to required mechanical properties, the SPPC (Structure Properties Prediction and Control Technology) was proposed to design hot rolling process for SPAH steel. Based on large quantity of production data, a model group including ten BP neural networks was established to describe the relationship between chemical composition, process parameters and mechanical properties. The prediction precision of yield strength, tensile strength and elongation reached ±6%, ±6% and ±4% respectively. Combined with the multiobjective particle swarm optimization algorithm, the optimization of processing parameters was conducted to meet the customers’ requirements of mechanical properties with preset chemical composition and process constraints. Good agreement was achieved between calculated and production data, which proves the validity of this method and provides guidelines for optimal design of hot rolling process.
参考文献
- 下载量()
- 访问量()
- 您的评分：
-
10%
-
20%
-
30%
-
40%
-
50%