Tên bài báo:

Development of 48-hour Precipitation Forecasting Model using Nonlinear Autoregressive Neural Network
Tác giả:
Phạm Thái Bình
Tham gia cùng:
Lý Hải Bằng
Tạp chí:
CIGOS 2019, Innovation for Sustainable Infrastructure
Năm xuất bản:
Từ trang 1191 đến trang 1196
Lĩnh vực:
Kỹ thuật xây dựng công trình giao thông
Phạm vi:
Quốc tế

Tóm tắt:

Rainfall intensity has a significant impact on urban drainage infrastructures and the precipitation forecast therefore remains essential in urban areas. In this study, a prediction model using Nonlinear Autoregressive Neural Networks (NANN) was proposed to forecast 48-hour-ahead the rainfall intensity. The proposed NANN model, which is based on a precipitation data of fiveyear time series, was constructed and validated using various parameters such as Coefficient of Determination (R 2 ), Root Mean Squared Error (RMSE) and Mean Absolute Error (MAE). The results exhibited a high statistical correlation between the outputs of NANN model and the measured data for 48 hour ahead prediction, i.e.R 2 =0.8998, RMSE=3.2909 and MAE=1.8672. This indicates that the developed model is very promising for precipitation forecasting and could contribute to improve the urban drainage systems.

Từ khóa:

Nonlinear Autoregressive Neural Networks Precipitation Time Series Forecasting
Thông tin tác giả
Phạm Thái Bình

Phạm Thái Bình

Thạc sĩ kỹ thuật

Lý lịch khoa học