LIU Gu
,
WANG Liu-ying
,
CHEN Gui-ming
,
HUA Shao-chun
钢铁研究学报(英文版)
Plasma surface hardening process was performed to improve the performance of the AISI 1045 carbon steel. Experiments were carried out to characterize the hardening qualities. A predicting and optimizing model using genetic algorithm-back propagation neural network (GA-BP) was developed based on the experimental results. The non-linear relationship between properties of hardening layers and process parameters was established. The results show that the GA-BP predicting model is reliable since prediction results are in rather good agreement with measured results. The optimal properties of the hardened layer were deduced from GA. And through multi optimizations, the optimum comprehensive performances of the hardened layer were as follows: plasma arc current is 90 A, hardening speed is 22 m/min, plasma gas flow rate is 60 L/min and hardening distance is 43 mm. It concludes that GA-BP mode developed in this study provides a promising method for plasma hardening parameters prediction and optimization.
关键词:
plasma transferred arc
,
surface hardening
,
optimization
,
neural network
,
genetic algorithm
WANG Hong-bing
,
XU An-jun
,
AI Li-xiang
,
TIAN Nai-yuan
钢铁研究学报(英文版)
The hybrid method composed of clustering and predicting stages is proposed to predict the endpoint phosphorus content of molten steel in BOF (Basic Oxygen Furnace). At the clustering stage, the weighted K-means is performed to generate some clusters with homogeneous data. The weights of factors influencing the target are calculated using EWM (Entropy Weight Method). At the predicting stage, one GMDH (Group Method of Data Handling) polynomial neural network is built for each cluster. And the predictive results from all the GMDH polynomial neural networks are integrated into a whole to be the result for the hybrid method. The hybrid method, GMDH polynomial neural network and BP neural network are employed for a comparison. The results show that the proposed hybrid method is effective in predicting the endpoint phosphorus content of molten steel in BOF. Furthermore, the hybrid method outperforms BP neural network and GMDH polynomial neural network.
关键词:
basic oxygen furnace
,
endpoint phosphorus content
,
K-means
,
neural network
,
GMDH
薛涛，杜凤山，孙静娜
钢铁
针对冷连轧机组中的6辊CVC轧机，利用大型非线性有限元软件MSC.Marc建立仿真模型，对多种轧制条件下轧件的弹塑性变形和辊系的弹性变形进行了耦合计算，得到了6辊CVC冷轧机工作辊弯辊、中间辊弯辊、中间辊横移的板形调控功效离散值。研究了不同轧制工艺参数、轧件参数和轧辊参数对各板形调节机构调控功效的影响规律，并得出相应的板形调控功效系数。以有限元计算结果为样本，利用BP神经网络强大的非线性映射功能，建立了板形调控功效的神经网络计算模型，为板形在线闭环控制模型提供高精度的板形调控功效，解决了有限元计算耗时长，难以满足在线控制要求的问题。
关键词:
有限元
,
neural network
,
flatness
,
actuator efficiency
