基于粒子群算法的燃水比控制器参数优化研究

提出了一种采用粒子群算法优化多模型控制器参数的直流炉燃水比解耦控制方法。在种群初始化、惯性权值和变异率引入方面对基本粒子群算法进行了改进,以提高算法的收敛精度和速度。对解耦后的系统,分别用改进粒子群算法、基本粒子群算法和工程整定法得到了控制器参数,完成了燃水比控制的仿真试验。结果表明,使用基于改进粒子群算法的控制策略的系统较传统控制策略下的系统动、静态特性更好,更能适应深度调峰的需要。

PSO-Based Parameter Optimization for Fuel-Water Ratio Controller

A decoupling method is presented to solve the problem of fuel-water ratio in once-through boiler,using Particle Swarm Optimization(PSO) to optimize the parameters of multi-model controller.In order to enhance PSO algorithm′s convergence precision and speed,improvements are made in the way of population initialization,inertia weights and introduction of mutation rate.As for the the decoupled system,the controller parameters are obtained by means of improved PSO,basic PSO and engineering turning method respectively,and fuel-water ratio is simulated.Simulation results reveal that the proposed control strategy based on improved PSO is superior to the traditional one in the system dynamic and static characteristics,more feasible to the demand of depth peak load cycling.