| 自适应模糊神经网络预测金属-HEDTA配合物稳定常数的研究 |
| |
| 引用本文： | 董海峰,吴启勋. 自适应模糊神经网络预测金属-HEDTA配合物稳定常数的研究[J]. 计算机与应用化学, 2006, 23(9): 909-912 |
| |
| 作者姓名： | 董海峰 吴启勋 |
| |
| 作者单位： | 1. 青海师范大学民族师范学院化学系,青海,西宁,810008
2. 青海民族学院化学系,青海,西宁,810007 |
| |
| 摘 要： | 采用自适应模糊神经网络的方法,以金属离子的价电子结构、电负性、电荷半径比及失屏参数为参变量,关联金属- HEDTA配合物稳定常数。利用减法聚类算法以确定模糊神经网络的结构,并结合模糊推理系统调整其参数。30种已知的金属-HEDTA配合物稳定常数logK值预测结果令人满意,比函数连接网络要好些。在此基础上,预测了迄今尚缺的22种金属- HEDTA配合物的稳定常数值。
|
| 关 键 词： | 人工神经网络 自适应模糊神经网络 金属离子 HEDTA配合物 稳定常数 |
| 文章编号： | 1001-4160（2006）09-909-912 |
| 收稿时间： | 2006-04-01 |
| 修稿时间： | 2006-04-012006-06-28 |
Studies on prediction of the stabilities of metal-HEDTA complex by using adaptive fuzzy neural network |
| |
| Dong Haifeng,Wu Qixun. Studies on prediction of the stabilities of metal-HEDTA complex by using adaptive fuzzy neural network[J]. Computers and Applied Chemistry, 2006, 23(9): 909-912 |
| |
| Authors: | Dong Haifeng Wu Qixun |
| |
| Abstract: | An adaptive fuzzy neural network was applied to study the relationships between the structural parameters of metal ions and their stability constants of HEDTA complexes,with the variables of electronic numbers of the valence layer(f,d,s),out-shell electronic numbers(n),electronegativities(X_p),electric charge-radius ratio(Z~2/r),parameters of lost electron shielding(R_(?)).Subtractive clus- tering algorithm is used to confirm the structure of fuzzy neural network,and combined fuzzy inference systems to process regulation of the network parameters.Stable constants(logK)data of HEDTA complexes of 30 metallic ions were predicted.The obtained results are satisfactory,and better than that obtained by using functional-link net.And stable constants data of 22 metallic ions without experi- mental data were predicted.. |
| |
| Keywords: | artificial neural network adaptive fuzzy neural network metallic ions HEDTA complexes stable constants |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
