【摘要】 在实数编码免疫算法的基础上引进了模糊控制技术,对免疫算法中的两个关键参数实现了模糊自适应调整,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。模糊免疫算法被用于优化BP神经网络的结构和参数。结果表明,不但网络的结构得到控制,而且泛化性能有了较大的提高;同时,算法在优化神经网络上的有效性也在溶剂脱水塔醋酸浓度软测量建模中得到很好地证实。更多还原

【Abstract】 In this paper,fuzzy control technology is introduced into the real number coded immune algorithm to realize the adaptive selection of two key parameters.This method is proved to be effective which can well address the problems of slow convergence speed and low calculation precision in the basic immune algorithm.Then,the fuzzy immune algorithm is applied to optimize the structure and parameters in BPNN.And the availability of algorithm optimizing neural network is proved by applying BPNN in acid concentration soft sensor modeling of distillation column.更多还原