摘要
探讨苹果籽油的乳化工艺,建立乳化工艺的人工神经网络模型,研究工艺模型的遗传算法优化技术。结果表明:影响乳化稳定性的因素有乳化剂的配比、壁材用量、壁材比、乳化剂量、乳化温度及乳化时间;神经网络结构为4-9-1的模型能较为精确地拟合输入的样本数据,其对测试样本的输出值与试验结果的相关系数为0.988;遗传算法优化出的最佳工艺参数为壁材含量5%、乳化剂量0.59%、温度58℃、时间13min,该工艺参数下乳化稳定性明显大于单因素和二次组合试验的结果,比最好的大12.1%。用神经网络模型描述乳化工艺参数与乳化稳定性之间的关系,用遗传算法优化,能设计出最佳的乳化工艺参数。
The Artificial neural network （ANN） model for emulsification process was established, and the ANN model was optimized using genetic algorithm （GA） in this study. Results showed that emulsifier proportion, wall materials amount and ratio between them, emulsifier amout, temperature and time can affect emulsion stability. The trained network whose structure was 4-9-1 has a high generalization, and the correlation coefficient between simulating outputs and the test data is 0.988. The highest peffonmnce, which is 12.1% greater than that in experiments, is obtained on the conditions of wall materials amount 5%, emulsifier amount was 0.59%, temperature 58 ＂C and time 13 rain. ANN model can be used to describe relationships between process parameters and emulsion stability and GA can be used to optimize process parameters.
出处
《食品科学》
EI
CAS
CSCD
北大核心
2008年第9期166-169,共4页
Food Science
关键词
苹果籽油
乳化
神经网络
遗传算法
apple seed oil
emulsification
artificial neural network
genetic algorithm
