BP神经网络结合遗传算法用于丹参提取工艺的多目标优化

Back-propagation neural network and genetic algorithm for multi-objective optimization of extraction technology of Salvia miltiorrhiza Bge.

目的:使用BP神经网络结合遗传算法用于丹参提取工艺的多目标优化。方法:通过已知文献的丹参提取工艺优化实例,采用均匀设计法优化BP神经网络模型参数,并建立网络模型,再利用遗传算法对网络进行多目标寻优,获得丹参最佳提取工艺。结果:BP神经网络结合遗传算法用于丹参提取工艺的多目标优化,模型拟合度和预测性均高于文献采用的多元回归法。结论:BP神经网络结合遗传算法可用于丹参提取工艺的多目标优化。

Objective： To introduce back-propagation（BP） neural network and genetic algorithm for multi-objective optimization of extraction technology of Salvia miltiorrhiza Bge.. Methods： BP neural network was established and optimized with uniform design method based on the data in literature of optimization of extraction process of Salvia miltiorrhiza Bge. Genetic algorithm was used for multi-objective optimization to obtain optimal extraction technology of Salvia miltiorrhiza Bge.. Results： With BP neural network and genetic algorithm for multi-objective optimization of extraction technology of Salvia milti- orrhiza Bge. , the fitting and prediction were better than those of the multiple regression reported in literature. Conclusion： BP neural network and genetic algorithm for multi-objective optimization of extraction technology of Salvia miltiorrhiza Bge. is advisable.