| 崔震,郭生练,王俊,张俊,周研来.基于混合深度学习模型的洪水过程概率预报研究[J].水利学报,2023,54(8):889-897,909 |
| 基于混合深度学习模型的洪水过程概率预报研究 |
| Probabilistic forecasting of flood processes based on hybrid deep learning models |
| 投稿时间：2022-09-23 |
| DOI：10.13243/j.cnki.slxb.20220726 |
| 中文关键词: 概率预报 不确定性分析 长短期记忆神经网络 编码-解码结构 混合密度网络 |
| 英文关键词: probabilistic forecasting uncertainty analysis long and short-term memory neural networks encoder-decoder structure mixture density networks |
| 基金项目:国家重点研发计划项目(2021YFC3200301);国家自然科学基金项目(U20A20317) |
|
| 摘要点击次数: 2573 |
| 全文下载次数: 2532 |
| 中文摘要: |
| 传统的人工神经网络模型无法量化洪水预报的不确定性，而且在多时段连续预报中未考虑输出的时间相关性。本文通过融合新安江(XAJ)模型、基于外源输入编码-解码(EDE)结构的长短期记忆(LSTM)神经网络和混合密度网络(MDN)，构建了XAJ-LSTM-EDE-MDN混合深度学习模型，以实现洪水过程概率预报。该模型在考虑预报洪水时间相关性的前提下，将解码过程产生的点估计转化为条件概率分布的估计;进一步采用最大似然估计法建立了损失函数，通过自适应矩估计(Adam)算法优选模型参数。在陆水和建溪两个流域的研究结果表明：该模型在不降低XAJ-LSTM-EDE模型预报精度的前提下，可有效反映预报洪水过程的不确定性，获得合理可靠的置信区间和优良的概率预报性能，为水库防洪调度等决策提供更多的风险信息，同时为研究深度学习在洪水概率预报中的应用提供参考。 |
| 英文摘要: |
| The traditional artificial neural network model cannot quantify the uncertainty of flood forecasting and is unable to consider the temporal correlation of flood process forecasting in multi-time continuous forecasting.In this paper，a XAJ-LSTM-EDE-MDN model is constructed by fusing the Xinanjiang (XAJ) model，the long short-term memory (XAJ-LSTM-EDE) neural network based on the exogenous input encoder-decoder structure，and the mixture density network (MDN) to achieve probabilistic forecasting of the flood process.The model transforms the point estimates generated by the decoding process into the estimates of conditional probability distributions while considering the temporal correlation of the forecasted flood.The loss function is further established using the maximum likelihood estimation method，and the model parameters are trained by an adaptive moment estimation algorithm.The study results in the two river basins of Lushui and Jianxishow that the model can effectively reflect the uncertainty of the forecast flood without reducing the forecast accuracy of the XAJ-LSTM-EDE model，and obtain reasonable and reliable confidence intervals and excellent probabilistic forecast performance.It provides more risk information for decision-making such as reservoir flood control and scheduling，and also provides a reference for studying the application of deep learning in probabilistic flood forecasting. |
|
查看全文
查看/发表评论 下载PDF阅读器 |
| 关闭 |