Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
A Dam Deformation Prediction Model Based on ARIMA-LSTM
Dam deformation prediction is an important part of the dam safety supervision system, which can effectively judge the operation status of the dam. The dam safety monitoring data is time-series data with both trend and seasonality. In order to obtain more accurate prediction results, this paper proposes a prediction model based on ARIMA-LSTM, which solves the problem that the single prediction algorithm cannot solve the linear component and the nonlinear component in the dam deformation data at the same time. The model uses Autoregressive Integrated Moving Average (ARIMA) model to predict linear components in deformation data, and uses Long Short-Term Memory (LSTM) model to predict nonlinear components in dam deformation data. Combined with the dam monitoring data of Manwan Hydropower Station, we use the ARIMA-LSTM model to predict the future sequence and compared with the predicted results of ARIMA model, LSTM model and traditional-combined model. We found that the model has the smallest RMSE value and the FA value of this model is closest to 1. The results proved that the model can improve the accuracy of prediction effectively.
A Dam Deformation Prediction Model Based on ARIMA-LSTM
Dam deformation prediction is an important part of the dam safety supervision system, which can effectively judge the operation status of the dam. The dam safety monitoring data is time-series data with both trend and seasonality. In order to obtain more accurate prediction results, this paper proposes a prediction model based on ARIMA-LSTM, which solves the problem that the single prediction algorithm cannot solve the linear component and the nonlinear component in the dam deformation data at the same time. The model uses Autoregressive Integrated Moving Average (ARIMA) model to predict linear components in deformation data, and uses Long Short-Term Memory (LSTM) model to predict nonlinear components in dam deformation data. Combined with the dam monitoring data of Manwan Hydropower Station, we use the ARIMA-LSTM model to predict the future sequence and compared with the predicted results of ARIMA model, LSTM model and traditional-combined model. We found that the model has the smallest RMSE value and the FA value of this model is closest to 1. The results proved that the model can improve the accuracy of prediction effectively.
A Dam Deformation Prediction Model Based on ARIMA-LSTM
Xu, Guoyan (Autor:in) / Jing, Zixu (Autor:in) / Mao, Yingchi (Autor:in) / Su, Xinyue (Autor:in)
01.08.2020
1093035 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Multi-scale combined prediction model of concrete dam deformation based on VMD-LSTM-ARIMA
| Springer Verlag | 2024
Automatic Concrete Dam Deformation Prediction Model Based on TPE-STL-LSTM
| DOAJ | 2023
Prediction for the Sluice Deformation Based on SOA-LSTM-Weighted Markov Model
| DOAJ | 2023