Coal mine surface deformation prediction method based on AWC-LSTM model

The invention discloses a coal mine surface deformation prediction method based on an AWC-LSTM model, and belongs to the field of remote sensing algorithm development, and the method comprises the steps: obtaining time sequence deformation data of a coal mine surface; an ARIMA model is constructed,...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: WANG SHUAI, CHEN YU, CHENG HUIBIN, SUO ZHIHUI, FENG XIAOJUN, CHEN XINLONG, LI HUAIZHAN, TIAN JINZE, TIAN XIAOLONG
Format: Patent
Sprache:chi ; eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:The invention discloses a coal mine surface deformation prediction method based on an AWC-LSTM model, and belongs to the field of remote sensing algorithm development, and the method comprises the steps: obtaining time sequence deformation data of a coal mine surface; an ARIMA model is constructed, the time sequence deformation data are input into the ARIMA model, first data and second data are obtained, the first data are linear time sequence deformation prediction data, and the second data are nonlinear time sequence deformation data; a CNN-LSTM model is constructed, the second data is input to the CNN-LSTM model, a nonlinear prediction result is obtained, and the CNN-LSTM model is an optimized LSTM model; and obtaining a deformation prediction result of the coal mine surface based on the first data and the nonlinear prediction result. The method is simple to operate, avoids subjectivity of parameter setting, can effectively improve the accuracy of mining area surface deformation prediction, and realizes sh