Swarm Intelligence for Estimating Model Parameters in Thermodynamic Systems

Parameter estimation from VLE data has drawn ample amount of attention in nonlinear vapor–liquid thermodynamic modeling problems. Traditional optimization methods are very sensitive to the initial guesses of unknown parameters and often fail to converge to the global optimum of the parameter estimation nonlinear mathematical programming problems. In this work, we demonstrate the application of a swarm intelligence-based algorithm called the Particle Swarm Optimization (PSO) algorithm. It can solve efficiently the nonlinear parameter estimation problems and finds the global optimum with high probability. Furthermore, it is not sensitive to the initial estimates of unknown parameters. PSO is easy to execute and is experimentally proven to perform well on many optimization problems. The results obtained using PSO are compared with those reported in literature. Also, the new results obtained using PSO are presented.