DASHBOARD FOR CORROSION PREDICTION ANALYTICS

In the oil and gas industry, pipelines have become crucial for enabling the transport of
flammable and hazardous substances such as crude oil, natural gas, and refined
petroleum products. Compared with trucks and trains, they hold fluids in greater
volume, healthier way, and more environmentally friendly. However, as with any other
equipment, to some extent pipelines can have different failures. Leakage in the
pipelines can cause progressive accidents such as spillage of fluids, fire, and explosion.
Exposure of such incidents results in casualties, even worse, deaths, damage to the
environment and to properties, poor reputations, financial distress, and more negative
impacts. Thus, risk-reducing initiatives that can avoid leakage of the pipelines are
necessary because the interventions will ideally be able to control leakage root causes.
Many accidents have proven that corrosion induces the leakage phenomena in the
pipelines. Therefore, the commitment for safety measures to prevent leakage is crucial
to carry out corrosion assessment. In order to promote decision-making in the
prevention of pipeline leakage, this study analyzes the correlation of depth and length
of corrosion within pipelines. The methodology used in the project includes the
literature review and the study of neural long short term memory (LSTM) network
model and how it behaves with sequential data. Thus, the results of this work can assist
risk assessors in identifying the level of risk and prevent future leakage in the pipelines
effectively.