Anomaly detection in gas pipelines using flow and pressure
Gas supply is a process of supreme importance in Singapore because it is also used as a raw material for power generation. Majority of gas pipes in Singapore are underground and is made of iron. Gas Pipelines suffer from worsening due to aging and nosedive to satisfy the stated carrying capacities [...
Saved in:
Main Author: | |
---|---|
Other Authors: | |
Format: | Theses and Dissertations |
Language: | English |
Published: |
2018
|
Subjects: | |
Online Access: | http://hdl.handle.net/10356/73152 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Institution: | Nanyang Technological University |
Language: | English |
id |
sg-ntu-dr.10356-73152 |
---|---|
record_format |
dspace |
spelling |
sg-ntu-dr.10356-731522023-07-04T16:05:25Z Anomaly detection in gas pipelines using flow and pressure Kumar, Abhishek Justin Dauwels School of Electrical and Electronic Engineering DRNTU::Engineering::Electrical and electronic engineering Gas supply is a process of supreme importance in Singapore because it is also used as a raw material for power generation. Majority of gas pipes in Singapore are underground and is made of iron. Gas Pipelines suffer from worsening due to aging and nosedive to satisfy the stated carrying capacities [1]. It also results in problem for customers, and reasons power disruptions with loss in industrial productions and occasionally even detonation hazards. Thus, it is imperious to reduce its total cost to make it inexpensive. Malfunction in gas pipelines system are related with various causes. (e.g., leaks, Water Ingression, & worsening of pipes). Approaches for the recognition of pipelines fault were solely established for three circumstances: leaks of gas, obstruction due to water ingress and corroding in the pipes because of aging. Therefore, Multi- Sensor Real Time Monitoring of gas pipeline for leakage, ingression, and disturbance recognition are part of new pipeline projects. Leakage recognition using distributed fiber-optic sensors can be a inclusive solution for unremitting, in-line, real-time monitoring of various pipelines. The scope of this project is to explore the glitches in gas pipeline, foremost signature parameters like the flow rate, change in pressure, etc. by obtaining data using flow and pressure transmitters at the specified locations. Furthermore, it is needed to explore multi-sensor based discovery of anomaly, leak, and linked problems like water ingress and gas leakage. This will help the manoeuvre team in anticipating failure in anomaly recognition. Real-time experimental data was collected from the live test bed at SP Power grid to simulate the results of the acquired data for limiting and tracking the leakage in gas pipelines. Also, comparison experimental data of gas pipelines above the ground with that of test data of the underground pipelines was analysed on pipelines DI 150 and DI 100. Master of Science (Computer Control and Automation) 2018-01-03T08:37:24Z 2018-01-03T08:37:24Z 2018 Thesis http://hdl.handle.net/10356/73152 en 70 p. application/pdf |
institution |
Nanyang Technological University |
building |
NTU Library |
continent |
Asia |
country |
Singapore Singapore |
content_provider |
NTU Library |
collection |
DR-NTU |
language |
English |
topic |
DRNTU::Engineering::Electrical and electronic engineering |
spellingShingle |
DRNTU::Engineering::Electrical and electronic engineering Kumar, Abhishek Anomaly detection in gas pipelines using flow and pressure |
description |
Gas supply is a process of supreme importance in Singapore because it is also used as a raw material for power generation. Majority of gas pipes in Singapore are underground and is made of iron. Gas Pipelines suffer from worsening due to aging and nosedive to satisfy the stated carrying capacities [1]. It also results in problem for customers, and reasons power disruptions with loss in industrial productions and occasionally even detonation hazards. Thus, it is imperious to reduce its total cost to make it inexpensive. Malfunction in gas pipelines system are related with various causes. (e.g., leaks, Water Ingression, & worsening of pipes). Approaches for the recognition of pipelines fault were solely established for three circumstances: leaks of gas, obstruction due to water ingress and corroding in the pipes because of aging. Therefore, Multi- Sensor Real Time Monitoring of gas pipeline for leakage, ingression, and disturbance recognition are part of new pipeline projects. Leakage recognition using distributed fiber-optic sensors can be a inclusive solution for unremitting, in-line, real-time monitoring of various pipelines. The scope of this project is to explore the glitches in gas pipeline, foremost signature parameters like the flow rate, change in pressure, etc. by obtaining data using flow and pressure transmitters at the specified locations. Furthermore, it is needed to explore multi-sensor based discovery of anomaly, leak, and linked problems like water ingress and gas leakage. This will help the manoeuvre team in anticipating failure in anomaly recognition. Real-time experimental data was collected from the live test bed at SP Power grid to simulate the results of the acquired data for limiting and tracking the leakage in gas pipelines. Also, comparison experimental data of gas pipelines above the ground with that of test data of the underground pipelines was analysed on pipelines DI 150 and DI 100. |
author2 |
Justin Dauwels |
author_facet |
Justin Dauwels Kumar, Abhishek |
format |
Theses and Dissertations |
author |
Kumar, Abhishek |
author_sort |
Kumar, Abhishek |
title |
Anomaly detection in gas pipelines using flow and pressure |
title_short |
Anomaly detection in gas pipelines using flow and pressure |
title_full |
Anomaly detection in gas pipelines using flow and pressure |
title_fullStr |
Anomaly detection in gas pipelines using flow and pressure |
title_full_unstemmed |
Anomaly detection in gas pipelines using flow and pressure |
title_sort |
anomaly detection in gas pipelines using flow and pressure |
publishDate |
2018 |
url |
http://hdl.handle.net/10356/73152 |
_version_ |
1772827858242109440 |