摘要
前混合磨料射流理论的不完善性及切割深度与影响因素间存在复杂的非线性关系 ,难以用传统的数学方法建立切割深度模型 .在实验室试验数据的基础上 ,应用神经网络方法建立了切割深度的预测模型 ,模型训练平均误差达到 0 0 1.应用结果表明 ,方法可行 ,模型可靠 ,相对误差小于 7 4 7% ,具有较大的实用价值 .扩大了神经网络方法的应用范围 ,探索出了建立切割深度模型的有效途径 。
Due to the deficiency of DIA jet theory, the complicated non linear relations between the cut depth and its affecting factors are difficult to be established model with traditional mathematical method. On the base of experimental data in laboratory, a neural network model is applied to forecast the cut depth and its average training error is 0 01. The results of actual application indicate that the model is reliable and precise, its relative error is less 7 47%, it has larger practical value, which not only enlarge the extent of the application of neural network, but also finds the short cut to predict the cut depth and further riches the foundation of DIA jet theory.
出处
《煤炭学报》
EI
CAS
CSCD
北大核心
2002年第4期430-433,共4页
Journal of China Coal Society
基金
黑龙江省自然科学基金资助项目 (E0 0- 2 0 )
