Buck-Boost变换器PI+PBC控制器参数的多目标优化  

Multi-objective Optimization of PI+PBC Controller Parameters for Buck-Boost Converter

作　　者：王路 王久和[1] 赵燕[2] 李建国 张雅静 WANG Lu;WANG Jiuhe;ZHAO Yan;LI Jianguo;ZHANG Yajing(School of Automation,Beijing Information Science&Technology University,Beijing 100192,China;School of Information Science and Technology,Yanching Institute of Technology,Sanhe 065201,China)

机构地区：[1]北京信息科技大学自动化学院,北京100192 [2]燕京理工学院信息科学与技术学院,三河065201

出　　处：《电力系统及其自动化学报》2022年第11期84-91,99,共9页Proceedings of the CSU-EPSA

基　　金：国家自然科学基金资助项目(51777012);北京市自然科学基金-市教委联合资助项目(KZ201911232045)。

摘　　要：针对Buck-Boost变换器的PI+PBC(passivity-based control)控制器多个参数难以确定的问题,提出采用非支配排序遗传算法Ⅲ即NSGA-Ⅲ(non-dominated sorting genetic algorithmⅢ)算法进行参数多目标优化,可使变换器获得良好的动、静态性能。首先,建立Buck-Boost变换器的欧拉-拉格朗日EL(Euler-Lagrange)模型,设计PI控制与无源控制相结合的控制器;选用时间乘绝对误差积分、输出电压的超调量、电感电流的超调量作为目标函数,注入阻尼、比例系数及积分系数这3个参数作为约束条件,建立参数优化模型。然后,采用NSGA-Ⅲ算法对参数优化模型进行了多目标优化,并与NSGA-Ⅱ算法、PESA-Ⅱ(Pareto envelope based selection algorithmⅡ)算法进行比较,用Hyper-volume指标来评价各解集质量。最后,仿真结果表明,NSGA-Ⅲ算法收敛性与分布性都优于NSGA-Ⅱ和PESA-Ⅱ算法,可使变换器获得好的动态和静态性能。Aimed at the problem that it is difficult to determine the multiple parameters of a PI + passivity-based control(PBC)controller for a Buck-Boost converter,the non-dominated sorting genetic algorithm Ⅲ(NSGA-Ⅲ)is proposed for multi-objective parameter optimization in this paper,which can make the converter obtain satisfying dynamic and static performances. First,an Euler-Lagrange(EL)model of the Buck-Boost converter is established,and the PI+PBC controller is designed. A parameter optimization model is established with the integral of time weighted absolute error,the overshoot of output voltage and the overshoot of inductor current as its objective function,and the injection damping,proportional coefficient and integral coefficient as its constraints. Then,the multi-objective optimization of the parameter optimization model is performed using NSGA-Ⅲ,and a comparison with NSGA-Ⅱ and the Pareto envelope based selection algorithm Ⅱ(PESA-Ⅱ)is made. The Hyper-volume index is used to evaluate the quality of each solution set. Finally,simulation results show that both the convergence and distribution of NSGA-Ⅲ are better than those of NSGA-Ⅱ and PESA-Ⅱ,and the converter can obtain satisfying dynamic and static performances.

关 键 词：BUCK-BOOST变换器 PI控制 无源控制 非支配排序遗传算法Ⅲ Hyper-volume指标 

分 类 号：TM46[电气工程—电器]