RSIS International

Enhancing The Performance of Pipeline Leakage Detection System Using Artificial Neural Network

Submission Deadline: 30th December 2024
Last Issue of 2024 : Publication Fee: 30$ USD Submit Now
Submission Deadline: 21st January 2025
Special Issue on Education & Public Health: Publication Fee: 30$ USD Submit Now
Submission Deadline: 05th January 2025
Special Issue on Economics, Management, Psychology, Sociology & Communication: Publication Fee: 30$ USD Submit Now

International Journal of Research and Innovation in Applied Science (IJRIAS) | Volume VI, Issue XII, December 2021 | ISSN 2454–6194

Enhancing The Performance of Pipeline Leakage Detection System Using Artificial Neural Network

Obaji C.M.1, Okonkwo O.R.2, Chidiebere U.3
1,2Enugu State University of Science and Technology
3Destinet Smart Technologies

IJRISS Call for paper

Abstract
This research presents enhancing the performance of pipeline leakage detection system using artificial neural network. The study reviewed many literatures and singled out electrochemical sensor as the best when compared to the rest in sensing gas leakages. This was used as a sensing element to collected data of the gas leakages and then a feed forward neural network was modeled and trained with the gas data using back propagation algorithm. A gas reference model was generated after the training and deployed with the sensing element as an improved gas leakage detection system using Matlab. The system was simulated and the result showed high gas leakage detection accuracy of 98.95% was achieved, near real time detection of 11.23ms, MSE value of 0.000103Mu and a regression 0.999 was recorded. The result was also compared with another work developed with neural network which recorded 94.7% and the result showed that the new system achieved 4.75% improvement in detection accuracy which may look small but is very good when considering the delicate nature of natural gas leakages and the big problem it can cause.

Keywords: pipeline, natural gas leakages, feed forward neural network, training, detection, Matlab

I.INTRODUCTION

Black gold (crude oil) as some will call it is one of the most precious natural deposits on the mother earth. It is so important and valuable that even the bye products are employed today in various chemical industries for production reasons and globally all countries blessed with crude as a natural deposit have depended majorly on it as source of economic growth and foreign exchange. This crude oil is a viscous black liquid substance deposited under the earth crust and is mined through exploration by licensed oil industries. They are processed through fractional distillation in refineries to output lots of products ranging from natural gas to petroleum jelly. After the refining process, this crude output is transported to various depots or even offshore for marketing and commercial purposes through pipeline channels.





Subscribe to Our Newsletter

Sign up for our newsletter, to get updates regarding the Call for Paper, Papers & Research.