Application of LSSVM-PSO algorithm as a novel tool to predict standard molar chemical exergy for organic materials

ABSTRACT

In the different chemical processes, determination of equilibrium state has great importance. The chemical exergy as one of indictor of chemical stability can be used for this purpose. Due to this fact, in the present work a novel and accurate Least squares support vector machine (LSSVM) algorithm optimized by Particle Swarm optimization was used to predict standard molar chemical exergy of pure organic substances as function of their structures. The predicted standard molar chemical exergy values were compared with the extracted experimental data from literature in the statistical and graphical manners. The coefficients of determination for training and testing phases were determined as 0.9886 and 0.9928 respectively. The determined indexes and graphical analyses expressed the high degree of accuracy of LSSVM in estimation chemical exergy so the LSSVM algorithm can be used as an accurate predictive tool in thermodynamics of different processes.

KEYWORDS:

• Chemical exergy
• LSSVM
• artificial intelligence
• thermodynamic
• predicting model

Additional information

Notes on contributors

Ali Naghshgar

Ali Naghshgar has been graduated as a Master student in Petroleum engineering at department of Petroleum engineering in Marvdasht azad university.

Karim Rouhibakhsh

Karim Rouhibakhsh is a Master student in Petroleum engineering at department of Petroleum engineering in shiraz university.

Mohsen Zare

Mohsen Zare has been graduated as Master student in Petroleum engineering from Marvdasht azad university.

Mohammad Sadegh Goodarzi

Mohammad Sadegh Goodarzi is a MA student in Petroleum engineering at department of Petroleum engineering in Marvdasht azad university.