论文部分内容阅读
运用偏最小二乘(PLS)和遗传算法(GA)预测含能材料的爆炸性能。利用GA在“分子结构—爆炸性能(QSDR)”数据中选取较少的变量个数,以较少的变量个数包含较多的变量信息,再用PLS进行结构性能模型的建立和性能预测。将这种方法应用于呋咱和芳香类含能材料的性能预测当中,可以验证方法的有效性。
Predict the explosive performance of energetic materials by using partial least squares (PLS) and genetic algorithm (GA). GA is used to select a smaller number of variables in the “Molecular Structure - Explosion Performance (QSDR)” data and include more variable information in a smaller number of variables, and then use PLS to establish and performance structural performance model prediction. Applying this method to the performance prediction of furazan and aromatic energetic materials can verify the effectiveness of the method.