摘要
针对前馈神经网络隐含层神经元不能在线调整的问题,提出了一种自适应增长修剪算法(AGP),利用增长和修剪相结合对神经网络隐含层神经元进行调整,实现神经网络结构的自组织,从而提高神经网络的性能.同时,将该算法应用于污水处理生化需氧量(BOD)软测量,仿真实验结果表明,与其他自组织神经网络相比,AGP具有较好的泛化能力及较高的拟合精度,能够实现出水BOD的预测.
Due to the unchangable on-line problem of hidden neurons in feed-forward neural networks,an adaptive growing and pruning algorithm（AGP） was presented in this paper.This algorithm can insert and prune hidden neurons during the training process to adjust the structure of the network and achieve self organization of neural network structure,which can improve the performance of the neural network.Additionally,this algorithm has been applied to the biochemical oxygen demand（BOD） soft measurement of the wastewater treatment process.Experimental results show that the proposed algorithm can forecast the effluent BOD with better generalization ability and higher accuracy than other self-organizing neural networks.
出处
《智能系统学报》
2011年第2期101-106,共6页
CAAI Transactions on Intelligent Systems
基金
国家"863"计划资助项目(2007AA04Z160)
国家自然科学基金资助项目(60873043)
北京市自然科学基金资助项目(4092010)
高等学校博士点专项科研基金资助项目(200800050004)
关键词
自适应增长修剪算法
BOD软测量
神经网络
自组织
adaptive growing and pruning（AGP）
BOD soft-measurement
neural network
self organization
