摘要
目的确定黄连厚朴汤抗流感活性成分的提取工艺。方法以液料比、提取时间、提取温度和提取次数为考察因素,进行正交设计,以盐酸小檗碱与厚朴酚含量、浸膏得率及病毒抑制率为考察指标,熵权法赋权得到综合评分,再分别采用多目标遗传算法(multiple objective genetic algorithm,MOGA)和BP神经网络结合遗传算法(BP neural network combined with genetic algorithm,BPNN-GA)进行建模并目标寻优,比较2种方法的优劣性,以确定黄连厚朴汤抗流感活性成分的最佳提取工艺。结果正交法、MOGA和BPNN-GA的综合评分依次为22.63、23.57和24.20,BPNN-GA得到的综合评分最高,且与正交结果存在显著性差异。故选用BPNN-GA所得为最佳提取工艺条件,即:药材加9倍量的水,于72℃提取3次,每次54 min,平均综合评分为24.20。结论MOGA和BPNN-GA均可用于黄连厚朴汤的提取工艺优化,且BPNN-GA优化所得的工艺更为节能高效。结果可为中药多种活性成分的提取提供参考。
Objective To determine the extraction process of anti-influenza active ingredients of Huanglian Houpo Decoction(黄连厚朴汤).Methods Taking liquid-to-material ratio,extraction time,extraction temperature and extraction times as investigating factors,an orthogonal design was carried out.The contents of berberine hydrochloride and magnolol,the yield of extract and the virus inhibition rate were used as the inspection indicators,and entropy weighting was used.The comprehensive score was obtained by entropy weight method,and then the Multi-Objective Genetic Algorithm(MOGA)and BP neural network combined with genetic algorithm(BPNN-GA)were used to model and optimize the target.The advantages and disadvantages of the two methods were compared to determine the best extraction process of anti-influenza active components of Huanglian Houpu Decoction.Results The comprehensive scores of orthogonal method,MOGA and BPNN-GA were 22.63,23.57 and 24.20.The comprehensive scores obtained by BPNN-GA were the highest,and there was a significant difference from the orthogonal results.Therefore,the best extraction process adopted the best extraction conditions obtained by BPNN-GA,namely medicinal materials plus 9 times the amount of water,extracted 3 times at 72℃,54 min for each time,and the average comprehensive score was 24.20.Conclusion Both MOGA and BPNN-GA can be used to optimize the extraction process of Huanglian Houpo Decoction,and the process optimized by BPNN-GA is more energy-efficient and efficient.The results of this article can provide references for the extraction of many active ingredients of traditional Chinese medicine.
作者
吴巧凤
严云良
孙瑶
曹志明
WU Qiaofeng;YAN Yunliang;SUN Yao;CAO Zhiming(Zhejiang Chinese Medical University,Hangzhou 310053,Zhejiang,China)
出处
《中华中医药学刊》
CAS
北大核心
2022年第4期1-5,共5页
Chinese Archives of Traditional Chinese Medicine
基金
国家自然科学基金(81473335/H2803,81803346/H3001)
浙江省自然科学基金(LY18H280007)。
关键词
黄连厚朴汤
正交试验
多目标遗传算法模型
BP神经网络结合遗传算法模型
提取工艺
Huanglian Houpo Decoction(黄连厚朴汤)
multiple objective genetic algorithm model
BP neural network combined with genetic algorithm model
extraction process