River chaotic characteristic identification method on basis of multivariate time series

Abstract

The invention relates to a river chaotic characteristic identification method on the basis of multivariate time series, which comprises the steps: identifying chaotic characteristics of the multivariate time series; carrying out weight calculation on outside conditions influencing river evolution the basis of information entropy and calculating influence weights of width-depth ratio, runoff volume and sand content time series on river evolution, which particularly comprises the steps of constructing an initial matrix of q factors influencing river evolution in n years, carrying out normalization processing on the factors influencing river evolution, calculating the information entropy and calculating information entropy weights of the factors influencing river evolution; calculating a chaotic characteristic weighted average of the width-depth ratio, runoff volume and sand content time series. According to the invention, by utilizing the river chaotic characteristic identification method on the basis of the multivariate time series, the chaotic characteristics of a river can be identified and intensity of the chaotic characteristics of different river types can be judged so as to provide theoretical guidance for river evolution forecast and channel improvement.

Description

Technical field [0001] The invention relates to a method for identifying chaotic characteristics of rivers. In particular, it relates to a method for identifying chaotic characteristics of rivers based on multivariate time series. Background technique [0002] The river is a very complex nonlinear dynamic system. Whether the evolution of the river is a chaotic phenomenon, there has been no clear answer. In the past, long-term predictions of river evolution, including the sedimentation process of reservoirs, such as decades or even hundreds of years of evolution or siltation process, no matter whether mathematical or physical models are used, the prediction results are often different from each other, which is far from the actual situation. , Where is this problem? According to the chaos theory, if the river evolution is a chaotic phenomenon, then the river evolution prediction is feasible in the short-term, and it is difficult to predict or even unpredictable in the long-term. ...

AI Technical Summary

Problems solved by technology

[0002] The river is a very complex nonlinear dynamical system. Whether the evolution of the river is a chaotic phenomenon has not been clearly answered.

In the past, the long-term prediction of river evolution, including reservoir sedimentation process, such as the evolution or deposition process of decades or even hundreds of years, whether using mathematical models or physical models, the prediction results are often different from each other, which is far from the actual situation. , where is the problem? According to chaos theory, if river evolution is a chaotic phenomenon, then the prediction of river evolution is feasible in the short term, but difficult to predict in the long term, or even unpredictable

There is no practical significance in making long-term predictions of river evolution or reservoir sedimentation. We can only find ways to improve the accuracy of short-term predictions.

Images