
基于ForecastNet的径流模拟及多步预测
刘昱, 闫宝伟, 刘金华, 穆冉, 王浩
基于ForecastNet的径流模拟及多步预测
Runoff Simulation and Multi-step Prediction Based on ForecastNet
径流过程呈现出的强非线性,使得现有水文模型的预测性能受到制约,深度学习等人工智能方法具有较强的非线性拟合能力,一定程度上可以突破现有瓶颈。为有效提取径流序列的非线性时变特征信息,提高径流模拟精度和多步预测性能,以雅砻江上游雅江流域为研究对象,建立了基于具有时变结构的ForecastNet径流预测模型,并与传统水文模型SWAT(Soil and Water Assessnent Teol)和神经网络模型RNN(Recurrent Neural Network)、LSTM(Long Short-Term Memory)及其组合进行对比分析。结果表明,ForcastNet模型在长预见期径流预测中有较强的适用性,能有效提高径流模拟及多步预测精度,为高精度实时径流预测提供了一种技术支撑。
The strong nonlinearity presented by the runoff process constrains the prediction performance of the existing hydrological models. The artificial intelligence methods such as deep learning with strong nonlinear fitting ability can break through the current bottleneck to a certain extent. To effectively extract the nonlinear time-varying feature information of runoff sequences and improve the accuracy of runoff simulation and the multi-step-head forecasting performance, the ForecastNet based runoff prediction model with time-varying structure has been established. The Yajiang River Basin in the upper reaches of the Yalong River is taken as a case study, and comparative analyses among the ForecastNet, traditional hydrological model of SWAT(Soil and Water Assessnent Teol), and neural network models of RNN(Recurrent Neural Network), LSTM(Long Short-Term Memory) and RNN-LSTM are carried out. The results show that the ForcastNet model has strong applicability in long term prediction, and can effectively improve the accuracy of runoff simulation and multi-step-ahead forecasting, thus providing technical support for high-precision real-time runoff prediction.
径流模拟 / 多步预测 / 时变结构 / ForecastNet / SWAT {{custom_keyword}} /
runoff simulation / multi-step-ahead forecasting / time-varying structure / ForecastNet / SWAT {{custom_keyword}} /
表1 ForecastNet模型最优参数表Tab.1 Optimal parameters of ForecastNet |
参数名称 | 参数含义 | 取值 | 取值范围 |
---|---|---|---|
input_size | 输入层节点数 | 9 | 2~32 |
output_size | 输出层节点数 | 1 | 1~12 |
epochs | 最大训练迭代次数 | 500 | 500~2 000 |
dropout_rate | 学习率 | 0.001 | 0.000 1~0.01 |
n | 隐藏层内神经元个数 | 256 | 2~512 |
batch_size | 批大小 | 16 | 2~128 |
timesteps | 输入时间步长 | 12 | 2~24 |
表 2 径流模拟评价指标统计表Tab.2 The statistical table of runoff simulation evaluation index |
模型 | 率定期 | 验证期 | ||||
---|---|---|---|---|---|---|
NSE | RMSE/(m³·s-1) | PBIAS/% | NSE | RMSE/(m³·s-1) | PBIAS/% | |
SWAT | 0.77 | 285.62 | -8.0 | 0.76 | 282.06 | -3.2 |
ForecastNet | 0.88 | 207.70 | 9.1 | 0.82 | 202.64 | 0.9 |
表3 4种模型预测结果对比Tab.3 Comparison of forecasting results of four models |
模型 | 3 d预见期 | 5 d预见期 | 7 d预见期 | ||||||
---|---|---|---|---|---|---|---|---|---|
NSE | RMSE/(m3·s-1) | PBIAS/% | NSE | RMSE/(m3·s-1) | PBIAS/% | NSE | RMSE/(m3·s-1) | PBIAS/% | |
RNN | 0.80 | 208.91 | 2.3 | 0.72 | 282.06 | -3.2 | 0.65 | 323.16 | 5.6 |
LSTM | 0.81 | 186.97 | 3.7 | 0.74 | 223.74 | 1.8 | 0.68 | 314.65 | -2.7 |
RNN-LSTM | 0.83 | 182.96 | 2.5 | 0.75 | 216.06 | 4.1 | 0.69 | 304.34 | 4.8 |
ForecastNet | 0.88 | 170.16 | -1.6 | 0.83 | 194.89 | 2.7 | 0.78 | 215.18 | 3.1 |
1 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
2 |
芮孝芳.论流域水文模型[J].水利水电科技进展,2017,37(4):1-7,58.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
3 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
4 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
5 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
6 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
7 |
殷兆凯,廖卫红,王若佳,等.基于长短时记忆神经网络(LSTM)的降雨径流模拟及预报[J].南水北调与水利科技,2019,17(6):1-9,27.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
8 |
李洁,林永峰.基于多时间尺度RNN的时序数据预测[J].计算机应用与软件,2018,35(7):33-37,62.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
9 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
10 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
11 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
12 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
13 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
14 |
曲星宇, 曾鹏, 李俊鹏. 基于RNN-LSTM的磨矿系统故障诊断技术[J]. 信息与控制, 2019,48(2):179-186.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_ref.label}} |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
/
〈 |
|
〉 |