量子微粒群优化算法(QPSO)是一种改进的微粒群优化算法(PSO),克服了PSO算法搜索空间有限和易陷入局部极值的不足,同时该算法具有参数少、易实现、收敛速度快等优点。应用量子微粒群优化算法,以谷氨酸发酵过程产物(谷氨酸)浓度数据为检验样本,以Verhulst方程为菌体生长模型,进行发酵模型参数估计。实验结果表明,基于QPSO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。
Quantum-behaved particle swarm optimization (QPSO) algorithm is an improved PSO algorithm, which can avoid the shorts of finite sampling space and easily getting in local extremum. In addition, QPSO algorithm is simple implementation and fast convergence with few parameters. QPSO algorithm is applied to estimate the parameters of fermentation model; using product (glutamic acid) concentration data of fermentation process ofglutamic acid as test samples, using Verhulst equation as bacterium growth model. The experimental results show that this method features high precision with low computation, simple-operation and realizable.