论文部分内容阅读
依据独立共同可别粒子体系的熵与配分函数的关系,采用自适应模糊神经网络的方法,以元素原子量和其电子层数为参数,关联阳离子标准熵。利用减法聚类算法确定模糊神经网络的结构,并结合模糊推理系统调整网络参数,仿真的结果令人满意。成功地关联了固体化合物中70种阳离子的标准熵。在此基础上,预报目前尚缺的17种阳离子的标准熵。自适应模糊神经网络可望成为研究元素和化合物构效关系的辅助手段。
According to the relationship between the entropy and the partition function of independent co-discernable particle system, an adaptive fuzzy neural network method was used to correlate the standard entropy of the cation with the atomic weight of the element and its electron number as parameters. Subtractive clustering algorithm is used to determine the structure of fuzzy neural network, and the fuzzy inference system is used to adjust the network parameters. The simulation results are satisfactory. The standard entropy of 70 cations in solid compounds has been successfully correlated. Based on this, the standard entropy of the 17 kinds of cations still lacking is forecasted. Adaptive fuzzy neural network is expected to be an aid to the study of the structure-activity relationship between elements and compounds.