Burns prediction of TC4-Ti-alloy based on scaled conjugate gradient neutral networks
-
摘要: 为预测TC4钛合金高速外圆磨削烧伤的程度,基于磨削烧伤后组织相变导致的工件表面硬度值变化,以表面硬度值来反映工件的磨削烧伤程度。采用尺度共轭梯度算法的神经网络对TC4钛合金高速外圆磨削的表面硬度值进行预测,根据表面硬度值与磨削表面烧伤程度的对应关系预测磨削烧伤。验证实验表明:预测结果与实验结果的误差在5%以内,模型预测效果良好。Abstract: In order to predict the degree of TC4 titanium alloy after high-speed cylindrical grinding,surface hardness is used to differentiate grinding burns of the workpiece based on surface hardness value change resulted by phase transformation after grinding burn.Surface hardness of TC4 titanium alloy after high-speed cylindrical grinding is forecasted using the scaled conjugate gradient algorithm of neural network.Correspondence relationship between the surface hardness value and the degree of grinding burn is used to predict grinding burns.Validation experiment indicates that the error between the experiments and the predictions is within 5%,which means that the model prediction effect is good.
-
Key words:
- TC4 titanium alloy /
- surface hardness /
- burns prediction /
- conjugate gradient method
点击查看大图
计量
- 文章访问数: 168
- HTML全文浏览量: 44
- PDF下载量: 10
- 被引次数: 0