Flood flow prediction method of sequential network based on self-attention mechanism

A technology of flood flow and time series network, which is applied in the direction of forecasting, neural learning methods, biological neural network models, etc., and can solve the problem of parameter sensitivity

Inactive Publication Date: 2021-05-18
HOHAI UNIV
View PDF4 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, this kind of model is often very sensitive to internal parameters, and requires relevant researchers to have rele...

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Flood flow prediction method of sequential network based on self-attention mechanism
  • Flood flow prediction method of sequential network based on self-attention mechanism
  • Flood flow prediction method of sequential network based on self-attention mechanism

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0045] The present invention will be described in further detail below in conjunction with the accompanying drawings.

[0046] Such as figure 1 Shown, the present invention a kind of flood flow prediction method based on the time series network of self-attention mechanism, comprises the following steps:

[0047] Step 1, preprocessing historical flood data, including flood flow and related flood characteristic factors;

[0048] Step 2, using temporal convolutional network and long short-term memory network to build a flood prediction model in parallel. The model chooses temporal convolutional network and long short-term memory network to extract features in parallel. Among them, the temporal convolutional network can perform convolution calculations on the input sequence to obtain the hidden state of the sequence. Use the self-attention mechanism to calculate the weighted feature vector feature S extracted by the temporal convolutional network using the calculated results of...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention discloses a flood flow prediction method of a sequential network based on a self-attention mechanism, and belongs to the technical field of flood flow prediction. The method comprises the following steps: 1, preprocessing historical flood data including flood flow and related flood characteristic factors; 2, constructing a flood prediction model in parallel by using a time convolutional network and a long-short-term memory network; 3, setting related hyper-parameters of the flood prediction model; 4, training the flood prediction model by using the historical flood data, and storing a final model; and 5, applying the trained model to flood prediction, and evaluating a result according to a corresponding index. The prediction method provided by the invention has better robustness, has higher accuracy compared with a traditional long-short-term memory network prediction method, and can effectively complete flood flow prediction of middle and small river basins.

Description

technical field [0001] The invention relates to a flood flow prediction method based on a time series network of a self-attention mechanism, and belongs to the technical field of flood flow prediction. Background technique [0002] Flood is one of the common and widely distributed disasters in nature. It often causes all kinds of huge damage to modern society. The economic loss it brings is incalculable, and it seriously endangers the lives and property safety of the people. Due to the complex mechanism of the flood itself, it has the characteristics of great difficulty in predicting the intensity. Therefore, research on flood flow forecasting has been a hot spot in the past few decades. [0003] Common flood prediction models are generally divided into two categories, hydrological models and data-driven models. Traditional hydrological models explain complex hydrological processes through physical processes. However, this kind of model is often very sensitive to internal...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
IPC IPC(8): G06Q10/04G06F30/27G06N3/04G06N3/08
CPCG06Q10/04G06F30/27G06N3/08G06N3/047Y02A10/40
Inventor 巫义锐孙珺毅
Owner HOHAI UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap