Parameter estimation of methanol transformation into olefins through improved particle swarm optimization with attenuation function

A four component lumped kinetic model and a six component lumped kinetic model have been proposed to verify the capacity of the new functions for fitting the experimental data. When the power of influencing function is bigger than 1, the values of error objective function are much smaller than the power equal to 1. This is because the influence of water in the reaction medium changes nonlinearly with the oxygenate transforming. The influence decreases first and then increases. With the power increasing, the trend of the influence is similar and the improvement of the values is minimal.