Sensors & Transducers Journal (ISSN 1726- 5479) |
|
|
Vol. 17, Special Issue, December 2012, pp. 98-109
Selected papers from the 4th International Conference on Intelligent & Advanced Systems (ICIAS' 2012),
12-14 June 2012, Malaysia, Kuala Lumpur
Development of NOx Emission Model Using
Particle Swarm Optimized Least-Squared SVR (PSO-LSSVR) Hybrid
AlgorithmElangeshwaran PATHMANATHAN, Rosdiazli IBRAHIM, Vijanth ASIRVADAMElectrical and Electronic Engineering Department, Universiti Teknologi PETRONAS, Bandar Seri Iskandar, 31720 Tronoh, Perak, Malaysia Tel.: +605 368 7821, fax: +605 364 7443 E-mail: elanz117@gmail.com, rosdiazli@petronas.com.my, vijanth_sagayan@petronas.com.my
Received: 29 September 2012 /Accepted: 29 October 2012 /Published: 18 December 2012 |
Abstract: This paper aims to develop a NOx emission model of acid gas incinerator using a hybrid of particle swarm optimization (PSO) and least squares support vector regression (LSSVR). Malaysia DOE is actively imposing the Clean Air Regulation to mandate the installation of analytical instrumentation known as Continuous Emission Monitoring System (CEMS) to report emission level online to DOE office. As hardware based analyzer, CEMS is expensive, maintenance intensive and often unreliable. Therefore, software predictive techniques are often preferred and considered as a feasible alternative to replace the CEMS for regulatory compliance. The LSSVR model is built based on the emissions from an acid gas incinerator that operates in a Liquefied Natural Gas (LNG) Complex. PSO is used to optimize the hyperparameters used in training of the LSSVR model. The model is shown to outperform previously developed LSSVR models that were optimized using a combination of Nelder-Mead (NM) simplex and Coupled Simulated Annealing (CSA) algorithms.
Keywords: Industrial pollution, Particle swarm optimization, Predictive algorithms, Support vector machines
Buy this article online (it will be send to you in the pdf format by e-mail) or subscribe Sensors & Transducers journal
(12 issues per year plus special issues; 40 % discount for payment IFSA Members):
Buy this journal issue in pdf format |
Sensors & Transducers journal subscription 450 $ US per year: |
Buy this
article for |
Alternatively we accept a money transfer to our bank account. Please contact for details: sales@sensorsportal.com |
Download <here> the Library Journal Recommendation Form
|
1999 - 2012 Copyright ©, International Frequency Sensor Association (IFSA). All Rights Reserved.
Home - News - Links - Archives - Tools - Voltage-to-Frequency Converters - Standardization - Patents - Marketplace - Projects - Wish List - e-Shop - Sensor Jobs - Membership - Videos - Publishing - Site Map - Subscribe - Search
Members Area -Sensors Portal -Training Courses - S&T Digest - For advertisers - Bookstore - Forums - Polls - Submit Press Release - Submit White Paper - Testimonies - Twitter - Facebook - LinkedIn