计算机工程与应用 ›› 2007, Vol. 43 ›› Issue (12): 228-230.

• 工程与应用 • 上一篇    下一篇

基于模糊免疫算法在溶剂脱水塔软测量

孟科 李绍军 钱锋   

  1. 华东理工大学 华东理工大学自动化研究所
  • 收稿日期:2006-05-26 修回日期:1900-01-01 出版日期:2007-04-20 发布日期:2007-04-20
  • 通讯作者: 孟科

Application of Real-coded Immune Algorithm in Soft Sensor Modeling of Solvent Tower

  • Received:2006-05-26 Revised:1900-01-01 Online:2007-04-20 Published:2007-04-20

摘要: 本文提出了一种改进的实数编码的免疫算法,其很好地解决了基本免疫算法中收敛精度低和寻优速度慢的缺点。通过比较和分析几个标准测试函数的计算结果,免疫算法的优良性得到了充分地证明。最后,免疫算法被用于优化BP神经网络的结构和参数。结果表明,网络的泛化性能有了较大的提高,算法在优化神经网络上的有效性也在溶剂脱水塔软测量建模中得到很好地证实。

关键词: 免疫算法, BP神经网络, 软测量

Abstract: In this paper, an improved real-coded immune algorithm is presented which well addresses the problems of slow convergence speed and low calculation precision in the basic immune algorithm. By comparing and analyzing the results of several standard test functions, the excellent performance of RIA is proved. Then, the RIA is applied to optimize the structure and parameters in BPNN. And the availability of algorithm optimizing neural network is proved by applying BPNN in soft sensor modeling of solvent tower.

Key words: Immune Algorithm, BPNN, Soft Sensor