CO<inf>2</inf> emission model development employing particle swarm optimized — Least squared SVR (PSO-LSSVR) hybrid algorithm

Abstract

This paper aims to develop a CO <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> 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). CEMS is used to report emission level online to DOE office. As hardware based analyzer, CEMS is expensive, maintenance intensive and often unreliable. Therefore, software predictive techniques is often preferred and considered as a feasible alternative to replace the CEMS for regulatory compliance. The LSSVR model is developed based on the emissions data from an acid gas incinerator that operates in a LNG Complex. PSO technique is used to optimize the hyperparameters used in training the LSSVR model. Overall, the LSSVR models have shown good performance in certain key areas in comparison with the BPNN model.