A hybrid prediction model for pipeline corrosion using Artificial Neural Network with Particle Swarm Optimization

Ee, L.K. and Aziz, I.A. (2018) A hybrid prediction model for pipeline corrosion using Artificial Neural Network with Particle Swarm Optimization. Journal of Engineering and Applied Sciences, 13 (Specia). pp. 3131-3138.

Full text not available from this repository.
Official URL: https://www.scopus.com/inward/record.uri?eid=2-s2....

Abstract

Pipeline corrosion is one of the most critical and severe cause of pipeline incidents annually. Pipeline incidents bring about disastrous damages not only to human but also to the ecosystem and economy of a country. Pipeline operators are aware of this fact and have deployed a more regular and thorough pipeline inspection program through various sensors for data acquisition that can be analyzed to predict the current state of pipelines. However, there are different factors which cause corrosion and current analytical methods are not specific enough in the prediction process. Therefore, a prediction model that is able to target specific corrosion damage mechanisms needs to be developed. Artificial Neural Networks (ANN) have been selected as the most suitable method to be adopted for such model. A critical study done among existing work on ANN has shown the need to improve time efficiency of the method. This project aims to develop a hybrid prediction Model which can target specific corrosion damage mechanisms. The basic ANN Model will be improved by integrating the Particle Swarm Optimization (PSO) algorithm to achieve a better and optimal performance. The final hybrid model will be put to test with a real world industrial dataset to verify its time efficiency as compared to the basic ANN Model. © Medwell Journals, 2018.

Item Type: Article
Impact Factor: cited By 0
Depositing User: Mr Ahmad Suhairi Mohamed Lazim
Date Deposited: 26 Feb 2019 03:18
Last Modified: 26 Feb 2019 03:18
URI: http://scholars.utp.edu.my/id/eprint/21267

Actions (login required)

View Item
View Item