|
|||||
|
|
| 基于并行神经网络集成的多步预测方法 | |
| 引用本文： | 谢景新,程春田,秦颖. 基于并行神经网络集成的多步预测方法[J]. 计算机工程与应用, 2006, 42(29): 75-77,80 |
| 作者姓名： | 谢景新 程春田 秦颖 |
| 作者单位： | 1. 大连理工大学电子信息学院,辽宁,大连,116024 2. 大连理工大学水利信息研究所,辽宁,大连,116024 3. 大连理工大学管理学院,辽宁,大连,116024 |
| 摘 要： | 神经网络集成通过训练多个神经网络并将其结论进行合成,可以显著地提高学习系统的泛化能力。该文提出了一种基于特征提取的并行神经网络集成多步预测模型ECPNN(ExtractionofCharacteristicsParallelNeuralNetwork)。从单因素时间序列中提取出代表内在机制的特征,采取并行TDNN(Time-delayNeuralNetwork)集成的方式实现时间序列多步预测。实验结果表明了该模型在多步预测方面的可行性和有效性。 |
| 关 键 词： | 键词 神经网络集成 特征提取 TDNN 多步预测 |
| 文章编号： | 1002-8331（2006）29-0075-03 |
| 收稿时间： | 2006-06-01 |
| 修稿时间： | 2006-06-01 |
Multi-step-ahead Prediction Based on Parallel Neural Network Ensemble |
|
| XIE Jing-xin,CHENG Chun-tian,QIN Ying. Multi-step-ahead Prediction Based on Parallel Neural Network Ensemble[J]. Computer Engineering and Applications, 2006, 42(29): 75-77,80 | |
| Authors: | XIE Jing-xin CHENG Chun-tian QIN Ying |
| Affiliation: | 1.School of Electronic and Information Engineering,Dalian University of Technology, Dalian, Liaoning 116024;2.Institute of Hydro Informatics, Dalian University of Technology, Dalian, Liaoning 116024;3.School of Management,Dalian University of Technology,Dalian,Liaoning 116024 |
| Abstract: | By training a finite number of neural networks and then combining their results,neural network ensemble can significantly improve the generalization ability of learning systems.In this paper,a new model,Extraction of Characteristics Parallel Neural Network(ECPNN),is proposed for multi-step-ahead prediction.The framework of the model is composed of parallel Time Delay Neural Networks(TDNN) to process characteristic and remainder sequences extracted from single factor time series.Tested on time series of sunspot prediction,the model provides more accurate result for multi-step prediction than single TDNN. |
| Keywords: | TDNN |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ | |