Xinanjiang model parameter calibration method based on hydrological similarity and artificial neural network

Abstract

The invention discloses a Xinanjiang model parameter calibration method based on hydrological similarity and an artificial neural network, and the method comprises the steps of analyzing the hydrological similarity, obtaining each feature vector of a drainage basin, and constructing a supervised learning database; establishing parameter mapping: using a supervised learning database to train the hybrid neural network-Xinanjiang model, namely an initial artificial neural network, and minimizing a loss function to update the parameters of the Xinanjiang model; flood forecasting and Xinanjiang model parameter calibration: inputting the feature vectors of the drainage basin into a trained hybrid neural network-Xinanjiang model, namely a trained artificial neural network, and outputting Xinanjiang model parameters by using the trained artificial neural network; and receiving rainfall information updated in real time as input of the trained artificial neural network, training the trained artificial neural network again to reduce forecast errors, and finally obtaining a Xinanjiang model parameter calibration result. The method is high in precision and high in physical interpretability and parameter mobility.

Description

technical field [0001] The invention belongs to the technical field of hydrological forecast combining data mining and traditional physical models, and in particular relates to a method for calibrating Xin'anjiang model parameters based on hydrological similarity and artificial neural network. Background technique [0002] With the development of information technology, the degree of water conservancy informatization has been continuously improved, and a large number of intelligent hydrological observation stations have been established. Massive historical hydrological data and real-time observation data are stored in the hydrological database. The hydrological data in the system has also shown explosive growth. Although it has brought convenience to the research of hydrological information, how to dig out the information behind these hydrological data has also become the most prominent problem in the water conservancy informatization. [0003] The trend can effectively pre...

AI Technical Summary

Problems solved by technology

However, in the application of PSO algorithm, like other global optimization algorithms, it may fall into local optimum, resulting in low convergence accuracy and slow convergence speed in the later stage.

Moreover, the above two parameter calibration methods do not take into account the hydrological similarity of the watershed, and the physical interpretability is low.

Images