谷氨酸发酵过程神经网络模型的建立

Establishment of a Neural Network Model In Glutamic Acid Fermentation

采用改进的BP神经网络作为建模工具,对实际的生产数据进行量化处理,利用处理后的数据分别对神经网络进行训练及校验,再对校验结果进行反量化处理,建立了关于菌体浓度的数学模型。

The glumatic acid fermentation is key step of the monosodium glutamate production. They are very strong non-linear and time-varying. It is difficult to establish accurate mathematical model. This paper adopt a modified algorithm of BP network to model, quantization pratical data of the productive process. The neural network is trained and verified by quantization data, inverse quantization verifying result finally. So we get the distribution mathematical model about mycelium density.