Runoff prediction method and system based on VMD decomposition and IHHO optimization LSTM

Abstract

The invention discloses a runoff prediction method and system based on VMD decomposition and IHHO optimization LSTM, and the method comprises the steps: firstly, selecting historical runoff data of a hydrometric station as experimental data, and carrying out preprocessing and normalization of the data; then decomposing the processed data into a plurality of sub-modes with different frequencies by using variational mode decomposition; optimizing the Harris hawks algorithm by using chaos initialization and a hill climbing algorithm, and optimizing two parameters, namely the number of hidden layer nodes and the learning rate, of along-short-term memory network by using the optimized Harris hawks algorithm; then, assigning the obtained optimal parameters to the LSTM, training each sub-mode, and establishing an IHHO-LSTM sub-model; and finally, testing each sub-mode to obtain a predicted value, carrying out aggregation and reverse normalization on the predicted values of the obtained sub-modes to obtain a final predicted value of the VMD-IHHO-LSTM, carrying out error analysis, and carrying out performance evaluation by using an error index. According to the method, the runoff forecasting precision can be improved, and a relatively accurate forecasting result can be obtained.

Description

technical field [0001] The invention belongs to the field of runoff prediction, and in particular relates to a runoff prediction method and system based on VMD decomposition and IHHO optimized LSTM. Background technique [0002] In recent years, the hydrological system has become more complex under the influence of global climate change and human activities, and medium and long-term runoff forecasting has become more and more important in hydrological forecasting. The improvement of forecasting accuracy can play a role in flood control forecasting, agricultural irrigation, hydropower station operation and other fields. valid reference. Therefore, how to improve the accuracy of medium and long-term runoff forecast under the influence of weather and geographical factors has become particularly important. [0003] The currently used runoff prediction method is mainly artificial neural network, but it is easy to fall into the situation of over-fitting or under-learning during t...

AI Technical Summary

Problems solved by technology

[0003] The currently used runoff prediction method is mainly artificial neural network, but it is easy to fall into the situation of over-fitting or under-learning during training. The development of deep learning provides new methods for runoff prediction, such as cyclic neural network RNN, which can be effective Use past input information, but the influence of the input of the hidden layer on the output will gradually decrease with iterations, and it is easy to fall into gradient disappearance

Images