TY  - JOUR
UR  - https://www.scopus.com/inward/record.uri?eid=2-s2.0-85140738980&doi=10.1007%2f978-981-19-1939-8_47&partnerID=40&md5=25c591e3ca53629c35c1b6bbb5530246
JF  - Lecture Notes in Mechanical Engineering
A1  - Shaik, N.B.
A1  - Pedapati, S.R.
A1  - Othman, A.R.
A1  - Dzubir, F.A.B.A.
EP  - 624
Y1  - 2023///
SN  - 21954356
PB  - Springer Science and Business Media Deutschland GmbH
N1  - cited By 3; Conference of 7th International Conference on Production, Energy and Reliability, ICPER 2020 ; Conference Date: 14 July 2020 Through 16 July 2020; Conference Code:284729
N2  - Almost all fluid transportation in oil and gas facilities from one point to another are mainly using simple and complex pipelines system.These pipelines system is vital as it generates significant cash revenue to the facilities.However, the structural health of pipeline deteriorates over time due to various damage mechanism.Any Loss of Primary Containment (LOPC) to the pipelines can lead to heavy business losses.Facility engineers will normally apply the most common approach to mitigate the LOPC by providing schedule maintenance, perform repair activities as well as system replacement.These activities are mostly controlled manually by the engineers.This paper will focus on the prediction of structural health monitoring of gasoline pipeline located in one of the facilities in Malaysia, using historical inspection reports by means Artificial Neural Networks (ANN) and Gaussian Process Regression (GPR) regression techniques.The collected data was analyzed and applied on these two methods to determine the best fitness and performance.It is noticed that both Bayesian-regularization ANN and exponential GPR model demonstrate better performance compared to the others when assessed based on the highest R2 value.The results were deemed satisfactory as the R2 value for both methods were close to 1.0.The results also showed that both ANN and GPR models were almost equal in predicting the structural health of a pipeline with an accuracy of 99 and 97.3, respectively.This work may help in controlling/monitoring the inspection cost and to preplan the maintenance scheduling for gasoline pipeline network in oil and gas industry. © 2023, Institute of Technology PETRONAS Sdn Bhd.
KW  - Gas industry; Gasoline; Neural networks; Petroleum transportation; Pipelines; Regression analysis; Structural health monitoring
KW  -  Case-studies; Gasoline pipeline; Gaussian process regression; Gaussian process regression model; Network process; Oil and gas; Performance; Pipeline systems; Regression; Structural health
KW  -  Forecasting
SP  - 611
ID  - scholars19504
TI  - A Case Study to Predict Structural Health of a Gasoline Pipeline Using ANN and GPR Approaches
AV  - none
ER  -