首页 | 本学科首页   官方微博 | 高级检索  
     检索      

基于适应度继承的航空发电机比例-积分-微分参数优化算法
引用本文:李昆,赵理,汪光,客汉宸.基于适应度继承的航空发电机比例-积分-微分参数优化算法[J].科学技术与工程,2021,21(33):14258-14265.
作者姓名:李昆  赵理  汪光  客汉宸
作者单位:北京信息科技大学机电工程学院
基金项目:国家自然科学基金(52077007);北京市教育委员会科技计划项目(KM201811232003);北京信息科技大学研究生科技创新项目(5112110835)
摘    要:三级无刷交流发电机系统运行过程复杂,很难用明确的数学公式进行表达,其PID参数的优化过程需要用模型的运行结果(而不是简单数学公式计算)来确定,这就导致传统的优化算法因仿真过程时间过长而不可行。针对该问题,提出了一种基于适应度继承的遗传算法。该算法首先将PID参数定义为种群内的个体,将上升时间及调节时间组合定义为优化目标,接着在寻优过程中将种群聚类为P个子类,对子类中精英个体利用航空发电机仿真模型运行结果进行适应度计算,对非精英个体利用日志分析器进行适应度估计,最后利用适应度值来对种群内个体进行下一轮循环的选择、交叉、变异等操作,从而实现了估计值与精确值的结合、提高了利用仿真工具进行PID参数优化的效率。仿真分析表明,在保证寻优质量的前提下,提出的基于适应度继承的遗传算法能显著缩短寻优时间,为航空发电机系统仿真设计提供了有效的研究手段和参考依据。

关 键 词:适应度继承  遗传算法  PID参数优化  三级无刷交流发电机  航空电源系统  转速波动  负载波动
收稿时间:2021/4/25 0:00:00
修稿时间:2021/11/9 0:00:00

Optimization of PID Parameters of Aero Generator Based on Fitness Inheritance Algorithm
Li Kun,Zhao Li,Wang Guang,Ke Hanchen.Optimization of PID Parameters of Aero Generator Based on Fitness Inheritance Algorithm[J].Science Technology and Engineering,2021,21(33):14258-14265.
Authors:Li Kun  Zhao Li  Wang Guang  Ke Hanchen
Institution:School of Mechanical and Electrical Engineering,Beijing Information Science and Technology University
Abstract:The operating process of three-stage brushless alternator system is very complicated and it is very difficult to describe the process with clear mathematical formulas. The PID parameters optimization process of it is determined not by a simple mathematical formula, but by the operating results of the model. So as to the traditional genetic algorithm is not feasible due to the long simulation process time. A genetic algorithm based on fitness inheritance was proposed to solve the problem. Firstly, the PID parameter is defined as the individual in the population, and the combination of rise time and adjustment time is defined as the optimization objective. Secondly, in the optimization process, the population was grouped into P sub-population, and the fitness of elite individuals was calculated with aero generator simulation model in the sub-population. At the same time, the fitness of non-elite individuals was estimated with using log analyzer. Finally, the fitness value of the individuals in the population is used to implement the select, cross and mutation operator in the next cycle, so as to the combination of the estimated value and the precise value. In this way, the simulation efficiency of the model in PID parameter optimization process was improved. The simulation results show that the proposed genetic algorithm based on fitness inheritance can significantly shorten the optimization time under the premise of ensuring the optimization quality, which provides an effective research means and reference for the simulation design of aero generator system.
Keywords:Fitness inheritance      genetic algorithm      PID parameters optimization      Three stage brushless alternator      Aero power supply system      Rotational fluctuation      Load fluctuation
点击此处可从《科学技术与工程》浏览原始摘要信息
点击此处可从《科学技术与工程》下载免费的PDF全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号