量子神经网络及在发酵过程中的应用.pdfVIP

2010 No．2 · · Serial No．215 China Brewing 154 Experience Exchanges 量子神经网络及在发酵过程中的应用 侯 勇1，2 (1. 山东经贸职业学院，山东潍坊 261011 ；2. 北京科技大学信息工程学院，北京 100083） 摘 要：给出了一种量子神经网络模型，组成该模型的全于神经元对信息的处理分为两阶段．第一阶段为宏观信息收集部分，产生 控制量子比特；第二阶段为微观信息处理部分，根据控制量子比特，改变工作量子比特，即神经元的状态。以谷氨酸发酵过程产物 （谷氨酸）浓度数据为检验样本，将量子神经网络应用到以Verhulst方程为菌体的生长模型中，进行发酵模型参数估计。实验结果表 明，基于量子神经网络算法的参数估计方法具有精度高、编程实现简单、计算量小等优点。 关 键 词：量子神经网络；量子神经元；发酵模型；参数估计；优化 中图分类号：TQ920.1 文献标识码：B 文章编号：0254-5071 （2010）02-0154-04 Quantum neural network and its application in fermentation HOU Yong1,2 ( 1. Shandong Vocational College of Economics and Business, Weifang 2610 11, China; 2. School of Information Engineering, Beij ing University of Science and Technology, Beij ing 100083, China) Abstract: A quantum neural network to simulate quantum controlled NOT gate was presented in the study. At the first stage, the neural cell collected macroscopical information of samples and generated controlled quantum bit. The second stage, was named quantum controlled NOT gate simulation, which means the quantum neural cell will be changed according to the corresponding control of quantum bit. Quantum neural network with quantum neural cell is applied in bacterium growth model of Verhulst equation for parameters estimation of fermentation model for glutamic acid production. It is showed that the method of parameters estimation based on quantum neural network has the advantages of high precision, feasible program and low computation. Key words: quantum neural network; quantum neural cell; ferment model; parameter estimation; optimization 发酵过程是一个非线性、多变量、时