| 本文已被：浏览 728次 下载 0次 |
码上扫一扫！ |
|
|
| 具有周期变化和下降趋势的地下水位的预测 |
|
金菊良1,杨晓华2,金保明3,丁晶1
|
1. 四川大学,水利电力工程学院,四川,成都,610065
2. 河海大学,数学物理系,江苏,南京,210098
3. 福建省南平市水利电力局,福建,南平,353000
|
|
| 摘要: |
| 应用门限自回归 (TRA)模型解决具有周期性变化和下降趋势的地下水位的预测问题 ,可以有效地利用地下水位资料所隐含的时序分段相关性 ,起到限制模型误差 ,保证模型预测性能的稳定性 ,提高预测精度的作用 |
| 关键词: 地下水位,时间序列,门限自回归模型,非线性预测,遗传算法 |
| DOI： |
| 分类号:P641 |
| 基金项目:国家自然科学基金! (49871 0 1 8)，中国博士后科学基金，四川大学高速水力学国家重点实验室开放基金!990 4 |
|
| Pediction of periodic variation and decline tendency of groundwater level |
|
JIN Ju-liang1,YANG Xiao-hua2,JIN Bao-ming3,DING Jing1
|
| Abstract: |
| A simple and general scheme is presented for establishing the threshold auto regressive (TAR) model. With the accelerating genetic algorithm by the authors, both threshold values and auto regressive coefficients can be optimized, and the difficult problem of modeling of TAR is resolved, which gives a strong tool for widely using TAR model. The result of the calculation example shows that the problem can be successfully resolved for the prediction of the periodic variation and decline tendency of groundwater level with the scheme, that the method is practical and efficient, and that TAR model can effectively utilize the important information of the section interdependence during the time series such as groundwater level dates by controlling threshold valves, can reduce model errors, and can ensure good stability and accuracy of the model forecasting. As a general method, the scheme has major theoretic value and wide range application for predicting nonlinear time series. |
| Key words: groundwater level,time series,threshold auto regressive model,nonlinear forecast,genetic algorithm |
