论文部分内容阅读
本文以神经网络为工具，探讨了Ni基喷涂材料Si、Cr、B等元素对其涂层硬度的影响规律，在工程和实验的实际样本的基础上，以均匀设计方法优化了网络结构和学习对参数，得到能正确反映元素-硬度相关性的人工神经网络，计算结果表明：神经网络法不失为正确复现这一复杂相关性有力的新工具，文中指出：以各无素偏相关指数绝对值的大小判别其对Ni基喷涂材料硬度影响程度，效果亦佳。
In this paper, the influence of Si, Cr, B and other elements on the hardness of Ni-based coatings was discussed based on the neural network. Based on the actual samples of engineering and experiment, the network structure and learning were optimized by uniform design The results show that the neural network method is a powerful new tool to correctly reproduce this complex correlation. It is pointed out in this paper that the absolute value of each element’s partial correlation index The size of the Ni-based paint to determine the extent of its impact on the hardness, the effect is also good.