| Autor: |
Shenshen Chi, Xuexiang Yu, Lei Wang, Fang Chen |
| Zdroj: |
Fresenius Environmental Bulletin; 7/15/2020, Vol. 29 Issue 7A, p6006-6013, 8p |
| Abstrakt: |
In order to improve the prediction accuracy and reliability of surface subsidence basin angle of critical deformation, a prediction model based on linear combination model (LC) and genetic algorithm (GA) was proposed to optimize ELM neural network. LC-GA-ELM model was established taking the measured data of 126 surface movement observation stations in China as training set and testing set, and the model prediction results were analyzed. The accuracy and reliability of the results show that the average relative error and root mean square error of the rise angle, dip angle and strike angle are not more than 1.02% and 1.79, respectively. The optimized neural network model has higher prediction accuracy and stability, which is useful and meaningful to guide and obtain high-precision surface moving boundary of the area to be studied. [ABSTRACT FROM AUTHOR] |
| Databáze: |
Supplemental Index |
| Externí odkaz: |
|
