In view of the difficulty to predict water level time series due to the mixed characteristics of linearity and nonlinearity and high uncertainty, a combined model of EMD-DELM-LSTM based on empirical mode decomposition (EMD), long term memory network (LSTM) and deep extreme learning machine (DELM) is proposed in this paper. DELM and LSTM are connected in parallel and connected in series with EMD. First, the original signal was decomposed into several eigenmode functions (IMFs) with single characteristics using EMD, and then the IMFs were classified into high, medium and low frequency signals, which were input into the DELM-LSTM parallel structure for prediction and reconstruction. An important lake in a university of Guangzhou is taken as an example to verify the effectiveness of the model, the results showed that compared with EMD-LSTM, EMD-DELM, LSTM, DELM and Bilstm models, the prediction performance of this model is significantly improved at different time scales, and the 40min time scale has the most obvious improvement effect. Compared with the comparison model, the improvement was 43.08%, 22.92%, 45.79%, 30.92% and 47.31%, respectively. It can be seen that this model has good reliability and stability for water level prediction at different time scales.
YU Zhou
,
JIANG Tao
,
FAN Peng-hui
,
NIU Chao-qun
,
CHEN Bing
. Multi-time Scale Prediction for Lake Level Based on EMD-DELM-LSTM Combined Model[J]. Journal of Changjiang River Scientific Research Institute, 0
: 0
.
DOI: 10.11988/ckyyb.2230032
[1] 张睿芝, 戴凌全, 戴会超, 等. 基于改进LSTM模型的三峡水库蓄水对洞庭湖水位影响的空间异质性分析[J]. 水利水电技术(中英文), 2022, 53(2): 98-108.
[2] 古奋飞, 胡贤德, 王涛春. 浅析大数据背景下深度学习在水文行业中的应用[J]. 给水排水, 2022, 58(8): 201.
[3] LIU Y, WANG H, FENG W W, et al.Short Term Real-Time Rolling Forecast of Urban River Water Levels Based on LSTM: A Case Study in Fuzhou City, China[J]. International Journal of Environmental Research and Public Health, 2021, 18(17).
[4] SONG C, YAO L.Application of artificial intelligence based on synchrosqueezed wavelet transform and improved deep extreme learning machine in water quality prediction[J]. ENVIRONMENTAL SCIENCE AND POLLUTION RESEARCH, 2022, 29(25): 38066-82.
[5] 刘亚新, 樊启祥, 尚毅梓, 等. 基于LSTM神经网络的水电站短期水位预测方法[J]. 水利水电科技进展, 2019, 39(2): 56-60,78.
[6] 刘艳, 张婷, 康爱卿, 等. Seq2Seq模型的短期水位预测[J]. 水利水电科技进展, 2022, 42(3): 57-63.
[7] 纪国良, 周曼, 刘涛, 等. 基于循环神经网络的水库水位预测方法研究[J]. 长江科学院院报, 2021: 1-7.
[8] 詹灏, 宛曼. 基于GRU-ELMAN并联神经网络的短期水量预测方法研究[J]. 给水排水, 2022, 58(5): 146-50.
[9] 刘惟飞, 陈兵, 余周. 基于GRU-BP组合模型的湖泊水位预测方法探索[J]. 中国农村水利水电, 2022, (11): 58-65.
[10] 要震, 许继平, 孔建磊, 等. 基于GA-Elman的河流水位预测方法研究[J]. 长江科学院院报, 2018, 35(9): 34-7.
[11] XU X-G, XU G-L, WANG X-T, et al.Empirical mode decomposition and its application[J]. Acta Electron Sin (China), 2009, 37(3): 581-5.
[12] XIE Z Q, LIU Q, CAO Y L.Hybrid Deep Learning Modeling for Water Level Prediction in Yangtze River[J]. Intelligent Automation and Soft Computing, 2021, 28(1): 153-66.
[13] 王亦斌, 孙涛, 梁雪春, 等. 基于EMD-LSTM模型的河流水量水位预测[J]. 水利水电科技进展, 2020, 40(6): 40-7.
[14] HUI R, LUND J, ZHAO J S, et al.Optimal Pre-storm Flood Hedging Releases for a Single Reservoir[J]. Water Resources Management, 2016, 30(14): 5113-29.
[15] HSU N S, WEI C C.A multipurpose reservoir real-time operation model for flood control during typhoon invasion[J]. Journal of Hydrology, 2007, 336(3-4): 282-93.
[16] SUDRIANI Y, RIDWANSYAH I, RUSTINI H A. Long short term memory (LSTM) recurrent neural network (RNN) for discharge level prediction and forecast in Cimandiri river, Indonesia [J]. IOP Conf Ser, Earth Environ Sci (UK), 2019, 299: 012037 8 .
[17] 王愈轩, 梁沁雯, 章思远, 等. 基于LSTM-XGboost组合的超短期风电功率预测方法[J]. 科学技术与工程, 2022, 22(14): 5629-35.
[18] 何嘉莉, 陈兵, 姜涛, 等. 城市内涝在线监控与信息服务数字化系统设计[J]. 中国给水排水, 2014, 30(1): 94-8.
[19] 赵印, 姜涛, 陈兵. 智慧城市排水管网云服务系统设计与实现[J]. 中国给水排水, 2017, 33(5): 99-103.
