基于遗传算法的神经网络对难溶硫化物K_(sp)的预测

K_(sp) Research of Insoluble Sulfides by an Artificial Neural Network Based on Genetic Algorithms

采用遗传算法训练神经网络的权重系数 ,并将该神经网络用于对 13种难溶硫化物Ksp的预测 ,预测Ksp值和实验Ksp值的相关系数为 0 .9985 7。

The K sp values of 13 insoluble sulfides were p redcited by a neural network trained by genetic algorithms. Good relationship b etween predicted values and experimental values was established. The results sho wed that the neural network based on genetic algorithms was successful in predic ting K sp of insoluble sulfides.