点击次数：

所属单位：[1]Shenzhen Univ, Normal Coll, Dept Biol & Chem, Shenzhen 518060, Peoples R China;[2]Shenzhen Univ, Dept Sci & Technol, Shenzhen, Peoples R China;[3]Xian Univ Elect Sci & Technol, Coll Elect Engn, Xian, Peoples R China;[4]Univ Elect Sci & Technol, Coll Microelect & Solid Elect, Chengdu, Peoples R China

发表刊物：ZEITSCHRIFT FUR METALLKUNDE

关键字：Artificial Neural Network (ANN); complex data analysis; glass-ceramic

摘要：An artificial neural network was applied to slag glass-ceramic complex data analysis. A three-layer feedforward network was built with a new robust learning algorithm based on a concept of entire error modifying. These studies indicate that the network has an excellent learning ability when its topology is M - (2M + 1) - 1 and when an appropriate study error was chosen. Research shows that this slag glass-ceramic neural network is robust, accurate and stable in complex data analysis, and can disclose the relationship of elemental compositions, processing parameters and material properties of slag glass-ceramic effectively.

文献类型：Article

卷号：95

期号：2

页面范围：97-101

ISSN号：0044-3093

是否译文：否