摘要
针对现有Oustaloup滤波器拟合精度不佳、结构复杂的缺点,提出了最优精简Oustaloup滤波器。针对粒子群优化算法整定分数阶PI^(λ)D^(μ)控制器参数时学习能力不充分、迭代收敛乏力的问题,提出了一种改进的粒子群优化算法。该算法设计了双异步非线性动态学习因子,以提高粒子的思考能力与信息共享能力,并增加了粒子群质量重心项,用以加速收敛过程。将改进的算法结合最优精简Oustaloup滤波器应用于分数阶PI^(λ)D^(μ)控制器的设计过程,选取了2个分数阶系统模型进行仿真验证。结果表明,改进的算法收敛速度更快且不易陷入局部最优,所设计的控制系统超调量更小、调节时间更短、稳态误差更小,提高了系统的抗干扰能力。
To overcome the shortcomings of the existing Oustaloup filter with poor fitting accuracy and complex structure,the optimal simplified Oustaloup filter is proposed.An improved particle swarm optimization algorithm is proposed to solve the problems of insufficient learning ability and weak iterative convergence when particle swarm optimization algorithm is used to tune parameters of fractional order PI^(λ)D^(μ) controller.A double asynchronous nonlinear dynamic learning factor is designed to improve the particle thinking ability and information sharing ability,and a particle swarm barycenter term is added to accelerate the convergence process.The improved algorithm combined with the optimal streamlined Oustaloup filter is applied to the design process of the fractional order PI^(λ)D^(μ) controller,and two fractional order system models are selected for simulation verification.The results show that the improved algorithm converges faster and is not easy to fall into a local optimum,and the designed control system has smaller overshoot,adjustment time and steady-state error,which improves the anti-interference ability of the system.
作者
王仁明
刘闻仲
鲍刚
张铭锐
杨婕
WANG Renming;LIU Wenzhong;BAO Gang;ZHANG Mingrui;YANG Jie(College of Electrical Engineering&New Energy,China Three Gorges University,Yichang 443002,China)
出处
《控制工程》
CSCD
北大核心
2024年第6期1067-1074,共8页
Control Engineering of China
基金
国家自然科学基金资助项目(61876097)。
