摘要
针对干扰严重的两组分体系——As(Ⅲ)和As(Ⅴ)难以利用分光光度法实现同时测定的问题,采用粒子群优化算法训练多层前向神经网络权值,有效地克服了传统反向传播算法误差收敛速度慢、易陷入局部极小值的缺点。结果表明,基于PSO的神经网络方法显著增强了数据处理的准确性和稳定性。
A neural network based on Particle Swarm Optimization （PSO） is adopted to solve the problem of simulta- neously analyzing As（ ⅢI and As（ Ⅴ ） using Speetrophotometer Analysis. The weights of multi-layer forward neural network are trained by the PSO algorithm, which overcomes the disadvantages of error converging slowly and dropping into the local minimum inhering in traditional Back Propagation （BP） neural network. And the results show that the method improves the accuracy and stability of data processing greatly.
出处
《新乡学院学报》
2009年第2期57-59,共3页
Journal of Xinxiang University
关键词
分光光度法
PSO
神经网络
speetrophotometer analysis
simultaneously analysis
PSO neural network
