用遗传算法优化的约束背景双线性化方法处理含未知背景组分干扰的色谱二维谱图体系

Numeric Genetic Algorithm Applying in Constrained Background Bilinearization for Two-Way Chromatography

数值遗传算法是全局优化方法.本文将其引入约束背景双线性化问题的优化求解过程,以避免陷入局部最优.用本方法处理了模拟数据和两个实际含未知背景干扰的色谱二维谱图体系,并探讨了如何提高遗传算法在优化平台区域的寻优速度,结果令人满意.

Numeric genetic algorithm (NGA) is an optimization technique for locating the global optimun. In this paper NGA was used as the optimization procedure in the constrained background bilinearization (CBBL) of two-way bilinear data in order to reduce the possibility of sinking into local optima. The behaviour of the algorithm was studied by simulations and real two - way chromatographic data. The results show that, with this algorithm, correct concentrations of the mixture can be determined in the presence of unknown interference. Some details in NGA are also discussed.