Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Runoff simulation method and system based on SOM-BPNN model

A simulation method and runoff technology, applied in neural learning methods, biological neural network models, CAD numerical modeling, etc., can solve the problems of ignoring various characteristics of runoff, affecting the accuracy of runoff simulation and prediction, and achieve the effect of improving prediction performance

Inactive Publication Date: 2021-06-25
INST OF GEOGRAPHICAL SCI & NATURAL RESOURCE RES CAS
View PDF7 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

When simulating and predicting runoff based on neural networks, the existing methods mostly use a single neural network method to directly simulate runoff, ignoring the various characteristics of runoff (such as flood peak flow, seasonality, interannual, etc.), which affects the accuracy of runoff simulation and prediction. precision

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 simulation method and system based on SOM-BPNN model
  • Runoff simulation method and system based on SOM-BPNN model
  • Runoff simulation method and system based on SOM-BPNN model

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0030] This embodiment is a runoff simulation method based on the SOM-BPNN model, such as figure 1 shown, including the following steps:

[0031] Step 1, multi-source data acquisition and processing: collect and download the flow data of hydrological stations, meteorological factor data and related remote sensing data in a certain watershed; the meteorological factors include rainfall, temperature, sunshine hours, relative humidity and wind speed; Remote sensing products include evapotranspiration and soil moisture data; outlier processing and missing value interpolation are performed on the collected data; the outlier processing and missing value interpolation are processed sequentially with a sliding window of length n=5, and the outlier is determined The threshold size ε is set according to watershed data.

[0032] Step 2, screening of key influencing factors: Based on the random forest algorithm, the key influencing factors of the simulated predictor (runoff) are screened...

Embodiment 2

[0062] This embodiment provides a runoff simulation system based on the SOM-BPNN model, such as Figure 4 As shown, the system includes:

[0063] Multi-source data acquisition and processing module 1, acquires and processes the multi-source data needed by the method of the present invention, mainly collects flow data of hydrological stations in the watershed, meteorological factor data of meteorological stations (rainfall, temperature, sunshine hours, relative humidity and wind speed) online And relevant remote sensing data (evapotranspiration and soil moisture data), and then use the length of n = 5 sliding windows to process outliers and imputation of missing values ​​in turn on the collected data.

[0064] The key impact factor screening module 2 is used to screen out the relevant impact factors for input and generate a total sample set based on the hydrological data of the watershed. The data collected and processed by module 1 (runoff impact Factors) to measure the impor...

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 relates to a runoff simulation method and system based on an SOM-BPNN model, belongs to the technical field of runoff simulation, and aims to capture multi-aspect characteristics of runoff and improve runoff simulation precision. The method comprises the following steps: acquiring and processing multi-source data; screening key influence factors; constructing an SOM neural network clustering model; constructing an SOM-BPNN runoff simulation model; and carrying out runoff simulation. The method comprises the following steps: firstly, carrying out unsupervised clustering on a sample data set by using an SOM model, and constructing a back propagation artificial neural network model according to a clustered sub-sample set to carry out runoff simulation. According to the method, the two artificial neural network models are coupled together, and the clustered sub-sample set is beneficial to multi-aspect feature learning of complex data by the back propagation artificial neural network, so that the runoff simulation precision of traditional machine learning can be further improved, and an effective auxiliary decision-making means and a solid theoretical basis can be provided for basin water resource planning, flood control, disaster reduction and comprehensive treatment.

Description

technical field [0001] The invention relates to a runoff simulation method and system based on a SOM-BPNN model, and is a watershed runoff simulation calculation method and system. Background technique [0002] Runoff is one of the key links in the water cycle and the basic element of water balance. Affected by global climate change and high-intensity human activities, the runoff process presents significant spatiotemporal heterogeneity and non-stationarity. The simulation and prediction of watershed runoff change is an important content in the field of hydrological research. High-precision runoff simulation and prediction have very important guiding significance for flood control and drought relief, water resource management system formulation and reservoir optimization scheduling. [0003] For a long time, scholars in the field of hydrological simulation and forecasting have mainly focused on using parametric methods to describe the temporal and spatial distribution, boun...

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): G06F30/27G06F30/28G06N3/04G06N3/08G06F111/10G06F113/08
CPCG06F30/27G06F30/28G06N3/084G06F2113/08G06F2111/10G06N3/044G06N3/045
Inventor 邹磊沈建明王飞宇夏瑞刘成建
Owner INST OF GEOGRAPHICAL SCI & NATURAL RESOURCE RES CAS
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products