(西北工业大学机械系,西安 710072)
摘 要: 金属液体凝固中直接挤压工艺是近几年发展起来的一种高效成形管、棒、型材新工艺。利用人工神经网络方法研究了该工艺的参数协调和变形力预测等问题,建立了神经网络预测模型。实验结果与预测值吻合良好,从而为预测和控制该工艺成形质量提供了有效的手段。
关键字: 神经网络;参数协调;液态挤压
(Department of Mechanics, Northwestern Polytechnical University, Xi'an 710072, P.R.China)
Abstract: The direct extrusion during solidification of liquid metal is a kind of new process for forming tube, bar or shaped products. This process is distinguished for its characteristics of high efficiency and energy saving. The method of artificial neural network was utilized to research the coordinating of the process parameters and the prediction of the forming pressure, and the model was established. The predicted values were in accordance with the experimental results. So an effective measure for foreseeing and controlling the forming quality of the process was provided.
Key words: neural network; parameters coordination; liquid extrusion