基于小波分析与RBF神经网络的甘蔗制糖结晶过程预测

基金项目: 国家自然科学基金资助项目（69864001）

详细信息

PREDICT IN CRYSTALLIZING PROCESS OF CANE SUGAR BASED ON WAVELET ANALYSIS AND RBF NEURAL NETWORKS

College of Electronic & Information Engineering, South China University of Technology, Guangzhou 510640

摘要: 本文针对甘蔗制糖结晶过程中由于传感器失灵引起的变量及参数不可测的问题，建立一个甘蔗制糖结晶过程的RBF（径向基函数）神经网络模型，该模型把小波分析和RBF神经网络紧密结合起来．实验结果表明该模型能达到逼近实际煮糖过程的目的．
- 小波分析 /
- 神经网络 /
- RBF网络 /
- 学习 /
- 权值
Abstract: According to the problems such as the unaffected sensor, the unavailable parameters in crystallizing process of cane sugar, a kind of RBF neural network model is built in this paper. In the model, the RBF networks are combined with wavelet analysis. The empirical test indicates the model can approximate the realistic process of cane sugar.
- wavelet analysis /
- neural networks /
- RBF networks /
- learn /
- weight

[1]	B.T.卡略夏等著,杜松龄等译校.煮糖过程的强化.轻工业出版社,1989
[2]	无锡轻工业学院,华南工学院编著.甘蔗制糖工艺学.轻工业出版社,1982
[3]	广东省糖纸食品工业公司编.糖膏煮炼与助晶.轻工业出版社,1985
[4]	陈树功编著.现代制糖工艺理论.轻工业出版社,1988
[5]	陈哲等.小波分析与神经网络结合的研究发展.电子科学学刊,22(3)
[6]	董新洲等.小波变换.继电器学刊,1999,27(1)