Runoff prediction based on a VMD-LSTM model considering the decomposition error

• Published in: Journal of physics. Conference series Vol. 2491; no. 1; pp. 12017 - 12025
• Main Authors: Ma, Ya-Rong, Yang, Jing, Li, Hao, Liao, He, Feng, Yu-Xin
• Format: Journal Article
• Language: English
• Published: Bristol IOP Publishing 01.04.2023
• Subjects: Algorithms; Decomposition; Errors; Physics; Prediction models; Runoff
• ISSN: 1742-6588; 1742-6596
• DOI: 10.1088/1742-6596/2491/1/012017

For increasing the forecasting accuracy of runoff, a combined prediction model composed of the variational mode decomposition (VMD) and long short-term memory network (LSTM) is investigated in our manuscript. Firstly, data from the runoff is decomposed into three modal components via the VMD algorithm, for reducing the complexity of the original data. In addition, decomposition error is also been considered in this paper, and the hidden information in the series is extracted. Three subsequences and the decomposition error are handled by the LSTM method, respectively. Superimposing the prediction results, the prediction result of the runoff is thus derived. Experimental analysis is carried out, and a comparison of the VMD-LSTM model and other algorithmic indicates that the model constructed in our manuscript is more valid for predicting the runoff.