基于LSTM神经网络的青藏高原月降水量预测
刘 新(1969— ),女,山东肥城人,博士,副教授,主要从事空间数据挖掘和机器学习等研究。E-mail:xinliu1969@126.com |
收稿日期: 2019-07-16
要求修回日期: 2019-10-01
网络出版日期: 2020-10-25
基金资助
国家自然科学基金项目(41774001)
国家科技基础性工作专项(2015FY310200)
版权
Prediction of Monthly Precipitation over the Tibetan Plateau based on LSTM Neural Network
Received date: 2019-07-16
Request revised date: 2019-10-01
Online published: 2020-10-25
Supported by
National Natural Science Foundation of China(41774001)
The Basic Science and Technology Project of China(2015FY310200)
Copyright
青藏高原的降水量预测不仅为该地区水资源合理规划利用提供依据,同时对中国及周边国家气候变化研究有着重要的意义。论文利用1990—2016年青藏高原降水量数据,采用长短期记忆神经网络(LSTM)对青藏高原月降水量进行预测,主要包括:① 使用青藏高原86个测站1990—2013年的月降水资料,预测各个测站2014—2016年的月降水量,并与传统的RNN、NAR、SSA和ARIMA预测模型相比,平均决定系数R2分别提高了0.07、0.15、0.13和0.36,均方根误差(RMSE)和平均绝对误差(MAE)表现更低;② 分析了降水量预测精度的空间分布特征,将各模型的R2在青藏高原地区内插值,分析R2的空间分布特征,发现所有模型降雨稀少的干旱地区和降雨多的湿润地区R2较低,在气候稳定、降水规律性明显的地区R2较高,且LSTM模型R2≥0.6的空间范围远大于传统模型;③ 分析了不同预测长度对各模型预测精度的影响,发现所有模型会随着预测长度增加而预测精度降低,但在不同的预测长度下LSTM预测的RMSE值都低于其他模型。
刘新 , 赵宁 , 郭金运 , 郭斌 . 基于LSTM神经网络的青藏高原月降水量预测[J]. 地球信息科学学报, 2020 , 22(8) : 1617 -1629 . DOI: 10.12082/dqxxkx.2020.190378
Precipitation prediction on the Qinghai-Tibet Plateau not only provides a basis for rational planning and utilization of water resources, but also has significance for climate change research in China and neighboring countries. In this paper, the Long Short Term Memory neural network (LSTM) was used to predict the monthly precipitation over the Qinghai-Tibet Plateau using data from 1990 to 2016. Firstly, the monthly precipitation data of 86 stations in the Qinghai-Tibet Plateau from 1990 to 2013 were used to predict the monthly precipitation of each station from 2014 to 2016. Comparing with the traditional RNN, NAR, SSA, and ARIMA prediction models, LSTM increased the average coefficient of determination (R2) by 0.07, 0.15, 0.13, and 0.36, respectively. Simultaneously, LSTM had lower Root Mean Squared Error (RMSE) and Mean Absolute Error (MAE). Among them, the observation of station 56106 showed that the LSTM model predicted the period more accurately with less displacement deviation, and that the prediction of the valley between July and September was more accurate with R2 reaching 0.87. Secondly, the spatial distribution characteristics of precipitation prediction accuracy were analyzed. The R2of each model was interpolated in the Qinghai-Tibet Plateau, and the spatial distribution characteristics of R2were analyzed. All the drought areas with rare rainfall and the wet areas with heavy rainfall were of lower R2, while the areas with stable climate and obvious precipitation were of higher R2. Areas of R2 over 0.6 were much larger when using the LSTM model than the traditional model. Finally, influence of different prediction lengths on the prediction accuracy was analyzed for each model. All models showed decreased prediction accuracy as the prediction length increased, yet the RMSE values predicted by LSTM were lower than by other models with the varying prediction lengths.
表1 ARMA模型类型及阶数确定Tab. 1 ARMA model type and order determination |
自相关系数 | 偏相关系数 | 模型定阶 |
---|---|---|
拖尾 | p阶截尾 | AR(p)模型 |
q阶截尾 | 拖尾 | MA(q)模型 |
拖尾 | 拖尾 | ARMA(p, q)模型 |
表2 56106号测站不同模型精度评价指标对比Tab. 2 Comparison of different model accuracy evaluation indexes at station 56106 |
评价指标 | LSTM | RNN | NAR | SSA | ARIMA |
---|---|---|---|---|---|
RMSE | 19.69 | 29.82 | 32.15 | 33.77 | 54.46 |
MAE | 14.06 | 18.57 | 24.17 | 22.17 | 40.67 |
R2 | 0.87 | 0.70 | 0.65 | 0.61 | 0.28 |
图10 青藏高原所有测站不同模型的降水量预测精度对比Fig. 10 Comparison of precipitation prediction accuracy of the different models at 86 stations on the Qinghai-Tibet Plateau |
表3 5种模型在青藏高原降水量预测精度对比Tab. 3 Comparison of prediction accuracy of the five models at 86 stations |
评价指标 | LSTM | RNN | NAR | SSA | ARIMA |
---|---|---|---|---|---|
RMSE | 26.83 | 28.87 | 31.64 | 30.17 | 38.01 |
MAE | 17.91 | 20.65 | 22.77 | 21.04 | 26.47 |
R2 | 0.61 | 0.55 | 0.46 | 0.48 | 0.25 |
图12 不同模型在预测长度分别为6、12、24和36个月时RMSE的变化Fig. 12 RMSE changes of the different models at predicted lengths of 6, 12, 24, and 36 months respectively |
表4 不同模型在不同预测长度下的平均RMSETab. 4 Average RMSEs of the different models at different predicted lengths |
预测长度/个月 | LSTM | RNN | NAR | SSA | ARIMA |
---|---|---|---|---|---|
36 | 26.83 | 28.87 | 31.64 | 30.17 | 38.01 |
24 | 26.23 | 28.16 | 30.45 | 29.37 | 35.82 |
12 | 25.21 | 27.52 | 29.95 | 28.78 | 36.24 |
6 | 19.89 | 23.85 | 26.58 | 25.10 | 27.30 |
[1] |
|
[2] |
|
[3] |
冯松, 汤懋苍, 王冬梅. 青藏高原是我国气候变化启动区的新证据[J]. 科学通报, 1998,43(6):633-636.
[
|
[4] |
潘保田, 李吉均. 青藏高原:全球气候变化的驱动机与放大器-Ⅲ.青藏高原隆起对气候变化的影响[J]. 兰州大学学报, 1996,32(1):108-115.
[
|
[5] |
|
[6] |
|
[7] |
|
[8] |
赵莹. 葠窝水库降雨量灰色预测模型应用分析[J]. 中国水能及电气化, 2016,141(12):68-70.
[
|
[9] |
潘刚, 芦冰, 邹兵, 等. 马尔可夫链在水库主汛期降雨状态预测中的应用[J]. 水利科技与经济, 2011,17(6):33-36.
[
|
[10] |
|
[11] |
|
[12] |
|
[13] |
宋帆, 杨晓华, 武翡翡, 等. 基于聚类分析的模糊马尔科夫链在降雨量预测中的应用[J]. 节水灌溉, 2018,278(10):38-41,46.
[
|
[14] |
|
[15] |
|
[16] |
|
[17] |
|
[18] |
|
[19] |
|
[20] |
|
[21] |
|
[22] |
|
[23] |
|
[24] |
|
[25] |
|
[26] |
|
[27] |
|
[28] |
|
[29] |
|
/
〈 | 〉 |