基于粒子群算法的重油热解模型参数估计

针对标准粒子群算法在进化过程中种群多样性降低而早熟的问题,提出了一种根据种群多样性测度动态改变惯性权重系数的自适应粒子群算法,该算法能够平衡算法的全局探索和局部开发能力,不仅有效地避免早熟,而且具有较快的收敛速度.两个经典的测试函数的仿真结果表明了算法的有效性.将改进的粒子群算法应用于重油热解模型参数估计中,效果明显.

Parameter estimation of the thermal cracking model for heavy oil based on particle swarm optimization

To overcome premature searching by standard particle swarm optimization(PSO)algorithm for the great loss of population diversity,a novel adaptive PSO is proposed,which can dynamically change inertia weight coefficients by the measure of population diversity.It not only effectively solves the problem of premature convergence but also has high convergence speed for the trade-off between global exploration and local exploitation.The effectiveness of the novel adaptive PSO is demonstrated by the simulation results of two classic test functions.The novel PSO is successfully applied to the nonlinear parameter estimation of the thermal cracking model for heavy oil.