Underground water chemical seasonal change analysis method based on self-organizing neural network

Abstract

The invention discloses an underground water chemical seasonal change analysis method based on a self-organizing neural network, and the method comprises the steps: obtaining underground water samples, obtaining underground water sample data based on the underground water samples, and obtaining the number of matched neurons based on the number of the underground water samples; constructing an SOM neural network based on the number of the matched neurons; and clustering the groundwater sample data through an SOM neural network, constructing a clustering change diagram based on a clustering result, and analyzing the groundwater sample based on the clustering change diagram to obtain a groundwater chemical seasonal change analysis result. The hydrological parameters in different seasons can be detected, and the spatial-temporal variability of the hydrological parameters can be accurately analyzed.

Description

technical field [0001] The invention relates to the technical field of groundwater evaluation, in particular to a method for analyzing seasonal changes in groundwater chemistry based on a self-organizing neural network. Background technique [0002] Self-organizing map neural network (SOM) is a neural network based on unsupervised learning method, which was first proposed by Kohen, Helsinki University of Technology, Finland in 1981. In hydrology, environment and other related fields, scholars usually apply SOM to hydrogeochemical cluster analysis, but in hydrogeochemical cluster analysis, the existing technology mainly focuses on the hydrological parameters in the same time and in the same area. This type of analysis generally detects the hydrological parameters at the same time to study the parameter distribution in the hydrological environment in a certain area. However, in the above-mentioned technologies, it is impossible to detect the transformation of hydrological para...

AI Technical Summary

Problems solved by technology

[0003] In order to solve the problems existing in the above-mentioned prior art, such as the inability to analyze and detect hydrological parameters at different times, the present invention provides a self-organizing neural network-based method for analyzing seasonal changes in groundwater chemistry, which can analyze groundwater through the SOM neural network. Cluster analysis of hydrochemical parameters, and then analysis of temporal and spatial variability of groundwater in the area

Images