【作者】 林朝朋； 芮汉明； 许晓春；

【Author】 Lin Chao-peng~1 Rui Han-ming~2 Xu Xiao-chun~1 (1.Food Engineering department,Shaoguan University,Shaoguan 512005)(2.Food College,South China University of Technology,Guangzhou510640)

【机构】 韶关学院食品工程系； 华南理工大学食品学院；

【摘要】 用 BP 神经网络的方法建立联结微生物培养基的性能与主要因素糖、硫酸锌、硫酸铵、硫酸钾、磷酸和硫酸镁的浓度的关系模型,并应用 MATLAB 神经网络工具箱实现对该模型的训练和系统仿真。不论培养基的配方如何变化,只要用一定的试验数据对模型进行训练,然后对拟采用的配方进行仿真,均可得到较为可靠的目标参数。更多还原

【Abstract】 The model of the relationship between the quality of culture medium and the consistence of C₆H₁₂O₆,（NH₄）₂SO₄,ZnSO₄, K₂SO₄,H₃PO₄,MgSO₄is established by the BP neural networks method and the relation model is trained and system simulation is implemented by MATLAB.Whatever the consistence of culture medium change,the satisfied quality targets are achieved as long as training the model with certain experimental data before the system simulation on the adopted consistence of culture medium.更多还原