基于混沌粒子群算法的火电机组热工过程辨识方法

Chaos swarm optimization algorithm based thermal system identification for thermal power units

针对火电厂热工过程的特点及其辨识方法的不足,以及粒子群(PSO)算法易早熟,且无法得到全局最优的问题,将粒子群早熟判断机制和混沌搜索序列添加到PSO算法中构成混沌粒子群优化(CPSO)算法,提出了基于CPSO算法的热工过程辨识方法,并将该方法应用于基于现场实测数据的热工过程模型辨识中。结果表明,该方法能够使粒子群快速摆脱局部极小值,提高了算法的收敛速度和收敛精度,取得了较好的辨识效果。

On the basis of the characteristics of thermal process,deficiency of existing methods for thermal object identification,and the weakness of particle swarm optimization （PSO）algorithm （easy to be prema-ture and can not obtain global optimal solution）,an identification method for thermal object based on chaos particle swarm optimization （CPSO）algorithm has been proposed,by adding the particle swarm premature judgment mechanism and the chaotic search sequence into the PSO algorithm.Furthermore,this method was applied to model identification for actual measured data based thermal process.The results show this method has the ability to jump out of local minimum,improve the convergence rate and precision,and achieve ideal identification effect.