一种有效的参数估计方法在预缩聚反应中应用

A Valid Algorithm for Parameter Estimation and Its Application to Prepolymerization Reaction

采用改进的遗传算法解决复杂聚合反应模型的参数估计问题。算法采用排序选择、多点交叉和变异优选策略,有效地提高遗传算法的搜索性能,避免了序贯优化方法有可能存在局部极值的问题。根据文献数据,仿真结果表明,该算法在参数估计中,具有参数搜索范围大、收敛速度快和精度高等特点,它能够有效地解决非线性参数估计问题。

This paper presents a parameter estimation for complex polymerization reaction model using genetic algorithms. Using sorting selection, more point crossover strategy and mutation strategy, the method effectively improves search performance of genetic algorithm. It avoids existing local extremum for sequent optimization methods. Simulation results show that it has characteristic of large scope search, fast convergence and high precision and that it can be applied to the problem of nonlinear parameter estimation.