用改进的人工蜂群算法设计AVR系统最优分数阶PID控制器

分数阶PID控制器（FOPID）是标准PID控制器的一般形式.与PID控制器相比，FOPID有更多的参数，其参数整定也更复杂.本文提出一种基于环交换邻域和混沌的人工蜂群算法（CNC-ABC），用于FOPID控制器的参数整定.CNC-ABC算法由于应用了环交换邻域，增加了解的搜索范围，从而能加快人工蜂群算法的收敛速度；同时利用混沌的遍历性使算法跳出局部最优解.用CNC-ABC算法优化AVR系统的FOPID控制器的参数.仿真结果表明，CNC-ABC算法整定的FOPID控制器比其它FOPID及PID控制器有较好的性能.

Optimum Design of Fractional Order PID Controller for an AVR System Using an Improved Artificial Bee Colony Algorithm

Fractional order proportional-integral-derivative(FOPID) controller generalizes the standard PID controller.Compared to PID controller, FOPID controller has more parameters and the tuning of parameters is more complex. In this paper, an improved artificial bee colony algorithm, which combines cyclic exchange neighborhood with chaos(CNC-ABC), is proposed for the sake of tuning the parameters of FOPID controller. The characteristic of the proposed CNC-ABC exists in two folds: one is that it enlarges the search scope of the solution by utilizing cyclic exchange neighborhood techniques, speeds up the convergence of artificial bee colony algorithm(ABC). The other is that it has potential to get out of local optima by exploiting the ergodicity of chaos. The proposed CNC-ABC algorithm is used to optimize the parameters of the FOPID controller for an automatic voltage regulator(AVR) system. Numerical simulations show that the CNC-ABC FOPID controller has better performance than other FOPID and PID controllers.