| 引用本文: | 赵书强,王磊,马燕峰,张昕刚,周玮.基于改进遗传算法的非线性励磁系统参数辨识[J].电力自动化设备,2007,27(7):1-4 |
| .Parameter identification of nonlinear excitation system based on improved genetic algorithm[J].Electric Power Automation Equipment,2007,27(7):1-4 |
|
|
| 摘要: |
| 将大变异遗传算法应用于非线性发电机励磁系统的参数辨识,利用其较强的全局寻优能力辨识出发电机励磁系统参数估计值。其原理为:当某一代中所有个体集中在一起时就以一个远大于通常变异概率的概率执行一次变异操作,随机、独立地产生许多新的个体,使种群脱离早熟。比较每代中所有个体的最大适应度与平均适应度的接近程度,判断当代中所有个体的集中程度;对当代适应度最高的2个个体不进行大变异操作,以保证具有最大适应度的个体不被破坏掉。采用Matlab的Simulink模块建立仿真模型,算例试验结果表明,基于大变异遗传算法的励磁系统参数辨识方法速度快、精度高。 |
| 关键词: 非线性励磁系统,参数辨识,大变异遗传算法,全局搜索能力 |
| DOI： |
| 分类号:TM76 |
| 基金项目: |
|
| Parameter identification of nonlinear excitation system based on improved genetic algorithm |
|
ZHAO Shu-qiang WANG Lei MA Yan-feng ZHANG Xin-gang ZHOU Wei
|
| Abstract: |
| Because of its strong global searching ability,the improved genetic algorithm based on big mutation is applied to the parameter identification of nonlinear generator excitation systems.When all units of one generation focus together,a mutation operation is carried out with a mutation probability much bigger than usual.Many new units are thus generated randomly and independently,making the group deviate from earliness.The difference between the maximum fitness function of all units in one generation and the average fitness function is used to judge the concentration degree of all units in one generation.The two units which have the maximum fitness function should not be operated by the big mutation operation to keep these two units undestroyed.The simulation model is set up using Simulink of Matlab.The fast speed and high precision of the proposed approach is demonstrated by experiment. |
| Key words: nonlinear excitation system,parameter identification,genetic algorithm based on big mutation,global searching ability |
