Prediction of Pore water Pressure Responses to Rainfall Using Auto Regressive Integrated Moving Average Method (ARIMA)

Pore water pressure (PWP) is the pressure of groundwater contain within a soil or rock, in gaps between particles (pores). Information of pore water pressure is required for slope stability analysis. Field instrumentation to collect PWP data is time consuming and expensive exercise. The objective of...

Full description

Main Author: SATHIVEL, SHALINI
Format: Final Year Project
Language: English
Institution: Universiti Teknologi Petronas
Record Id / ISBN-0: utp-utpedia.17988 /
Published: IRC 2016
Subjects:
TA Engineering (General). Civil engineering (General)
Online Access: http://utpedia.utp.edu.my/17988/1/final%20dissertation.pdf
http://utpedia.utp.edu.my/17988/
Tags: Add Tag
No Tags, Be the first to tag this record!
id utp-utpedia.17988
recordtype eprints
spelling utp-utpedia.179882018-08-01T09:51:54Z http://utpedia.utp.edu.my/17988/ Prediction of Pore water Pressure Responses to Rainfall Using Auto Regressive Integrated Moving Average Method (ARIMA) SATHIVEL, SHALINI TA Engineering (General). Civil engineering (General) Pore water pressure (PWP) is the pressure of groundwater contain within a soil or rock, in gaps between particles (pores). Information of pore water pressure is required for slope stability analysis. Field instrumentation to collect PWP data is time consuming and expensive exercise. The objective of this study was to predict soil pore water pressure responses to rainfall. Time series of PWP and rainfall for one month of 10 minutes resolution were used to develop ARIMA model. Autocorrelation and partial autocorrelation were performed to select appropriate input for the model. The ARIMA modelling was performed in three stages i.e. identification and estimatation of modelling parameters; diagnostic , and forecasting. Statgraphic software was used for this analysis. Performances of the ARIMA model was evaluated using root mean square (RMSE) and mean square error (MAE). ARIMA model with configuration of (3,1,3) was chosen to predict the PWP based on lowest error achieved (AIC = 0.065). The RMSE and MAE were 0.863 and 0.52 respectively. The predicted PWP data were observed is very closed to experimental PWP data. The study recommended that use of ARIMA for other hydrological analysis could also be advantageous. Additionally, correlation analysis is advantageous for appropriate selection of antecedent values and can be helpful to improve the model efficiency. This study was performed one hour interval of data and further research can be carried out using different time interval to estimate how long the model can give a good prediction. IRC 2016-05 Final Year Project NonPeerReviewed application/pdf en http://utpedia.utp.edu.my/17988/1/final%20dissertation.pdf SATHIVEL, SHALINI (2016) Prediction of Pore water Pressure Responses to Rainfall Using Auto Regressive Integrated Moving Average Method (ARIMA). IRC, Universiti Teknologi PETRONAS.
institution Universiti Teknologi Petronas
collection UTPedia
language English
topic TA Engineering (General). Civil engineering (General)
spellingShingle TA Engineering (General). Civil engineering (General)
SATHIVEL, SHALINI
Prediction of Pore water Pressure Responses to Rainfall Using Auto Regressive Integrated Moving Average Method (ARIMA)
description Pore water pressure (PWP) is the pressure of groundwater contain within a soil or rock, in gaps between particles (pores). Information of pore water pressure is required for slope stability analysis. Field instrumentation to collect PWP data is time consuming and expensive exercise. The objective of this study was to predict soil pore water pressure responses to rainfall. Time series of PWP and rainfall for one month of 10 minutes resolution were used to develop ARIMA model. Autocorrelation and partial autocorrelation were performed to select appropriate input for the model. The ARIMA modelling was performed in three stages i.e. identification and estimatation of modelling parameters; diagnostic , and forecasting. Statgraphic software was used for this analysis. Performances of the ARIMA model was evaluated using root mean square (RMSE) and mean square error (MAE). ARIMA model with configuration of (3,1,3) was chosen to predict the PWP based on lowest error achieved (AIC = 0.065). The RMSE and MAE were 0.863 and 0.52 respectively. The predicted PWP data were observed is very closed to experimental PWP data. The study recommended that use of ARIMA for other hydrological analysis could also be advantageous. Additionally, correlation analysis is advantageous for appropriate selection of antecedent values and can be helpful to improve the model efficiency. This study was performed one hour interval of data and further research can be carried out using different time interval to estimate how long the model can give a good prediction.
format Final Year Project
author SATHIVEL, SHALINI
author_sort SATHIVEL, SHALINI
title Prediction of Pore water Pressure Responses to Rainfall Using Auto Regressive Integrated Moving Average Method (ARIMA)
title_short Prediction of Pore water Pressure Responses to Rainfall Using Auto Regressive Integrated Moving Average Method (ARIMA)
title_full Prediction of Pore water Pressure Responses to Rainfall Using Auto Regressive Integrated Moving Average Method (ARIMA)
title_fullStr Prediction of Pore water Pressure Responses to Rainfall Using Auto Regressive Integrated Moving Average Method (ARIMA)
title_full_unstemmed Prediction of Pore water Pressure Responses to Rainfall Using Auto Regressive Integrated Moving Average Method (ARIMA)
title_sort prediction of pore water pressure responses to rainfall using auto regressive integrated moving average method (arima)
publisher IRC
publishDate 2016
url http://utpedia.utp.edu.my/17988/1/final%20dissertation.pdf
http://utpedia.utp.edu.my/17988/
_version_ 1741195277681819648
score 11.62408

Similar Items