Unlock instant, AI-driven research and patent intelligence for your innovation.

Runoff volume stability prediction method based on LSTM composite network

A forecasting method and composite network technology, applied in forecasting, biological neural network models, data processing applications, etc., can solve problems such as gradient disappearance, achieve the effects of narrowing the gap, improving the average effect, and improving the worst performance results

Pending Publication Date: 2020-04-07
TIANJIN UNIV
View PDF3 Cites 6 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] Recurrent neural network is suitable for time series data processing, but it has the problem of gradient disappearance

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
  • Runoff volume stability prediction method based on LSTM composite network
  • Runoff volume stability prediction method based on LSTM composite network
  • Runoff volume stability prediction method based on LSTM composite network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0036] The present invention is described in detail below in conjunction with accompanying drawing:

[0037] The present invention uses LSTM to stably predict the daily average runoff of the Jinghe River Basin. Adding meteorological and land type parameters to the input well takes into account the differences in the infiltration capacity of different land types for surface water. At the same time, based on LSTM, two prediction methods are used, namely direct prediction method and differential prediction method, to improve the performance stability of the prediction model and make the prediction results more reliable. In hydrological forecasting, traditional conceptual hydrological models are generally used to forecast stationary sequences, but hydrological data are nonlinear sequences with high uncertainty and complexity, so the forecasting effect of traditional models is not satisfactory. In modern forecasting methods, people use artificial neural network to build hydrologic...

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 runoff volume stability prediction method based on an LSTM composite network. The runoff volume stability prediction method comprises the following steps: selecting hydrological, meteorological and land related data from a public database; respectively establishing a direct prediction model and a differential prediction model according to the selected data; respectively selecting respective prediction results from the direct prediction model and the differential prediction model; combining and calculating the direct prediction result, the differential prediction result and the meteorological data to generate a calibration model; and performing optimal calculation on the calibration model to obtain an optimal calibration model, and outputting the optimal calibration model. According to the method, the performance stability of the prediction model is improved by using two prediction modes based on LSTM.

Description

technical field [0001] The invention relates to the application of deep learning methods to the field of hydrological forecasting to make quantitative and qualitative forecasts on hydrological conditions, and in particular to a method for stabilizing runoff flow forecasting based on LSTM composite networks. Background technique [0002] 1. Average daily runoff [0003] Runoff refers to the amount of water passing through a certain section of a river within a certain period of time. Runoff is the main link in the water cycle, and runoff is one of the most important hydrological elements on land and the basic element of water balance. Runoff will be affected by factors such as the basin's meteorology and land type. By averaging the instantaneous runoff by time, the average flow of a certain period (such as a day, a month, a year, etc.), such as daily average flow, monthly average flow, and annual average flow, can be obtained. The total amount of water passing through in a ...

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
Patent Type & Authority Applications(China)
IPC IPC(8): G06Q10/04G06N3/04
CPCG06Q10/04G06N3/044G06N3/045
Inventor 李幼萌王雨晴章亦葵
Owner TIANJIN UNIV