模式识别技术在生物杀虫剂苏云金杆菌生产中的应用

Application of Neural Network Pattern Recognition to Discriminating the Production Conditions of Bacillus Thuringiensis

采用 BP模型 ,以苏云金杆菌固态发酵为模型对象 ,研究了基于人工神经元网络的模式识别方法在固态发酵工艺条件辨别中的应用。网络自学习结果表明 ,发酵工艺条件分类全部正确 ,Cross Validation方法考察网络预测能力也得到满意的结果。说明人工神经元网络在生物发酵工程中有广泛的应用前景。

Artificial neural network pattern recognition was applied to discriminating the solid state fermentation conditions of Bacillus thuringiensis . The neural network was constitute of three layers and the back propagation algorithm was used. To evaluate the performence of the networks, the Cross Validation strategy was employed and satisfactory results was obtained. It was showed that neural network pattern recognition approach was quite promising in the optimization of fermentation conditions.