北京航空航天大学学报 ›› 2015, Vol. 41 ›› Issue (3): 530-537.doi: 10.13700/j.bh.1001-5965.2014.0182

• 论文 • 上一篇    下一篇

基于IFA-ELM的航空发动机自适应PID控制新方法

焦洋, 李秋红, 李业波   

  1. 南京航空航天大学 能源与动力学院, 江苏省航空动力系统重点实验室, 南京 210016
  • 收稿日期:2014-04-03 出版日期:2015-03-20 发布日期:2015-04-02
  • 通讯作者: 李秋红(1972—),女,辽宁葫芦岛人,副教授,lqh203@nuaa.edu.cn,研究方向为航空发动机建模、控制与故障诊断. E-mail:lqh203@nuaa.edu.cn
  • 作者简介:焦洋(1991—),男,河北秦皇岛人,硕士生,jy13661155288@163.com

New adaptive PID control method based on IFA-ELM for aero-engine

JIAO Yang, LI Qiuhong, LI Yebo   

  1. Jiangsu Province Key Laboratory of Aerospace Power System, College of Energy and Power Engineering, Nanjing University of Aeronautics and Astronautics, Nanjing 210016, China
  • Received:2014-04-03 Online:2015-03-20 Published:2015-04-02

摘要: 针对大涵道比涡扇发动机强非线性、变参数的特点,提出了一种基于优化极端学习机(ELM)对发动机参数进行预测的自适应PID控制方法.为提高ELM的预测精度和实时性,采用适用于多峰值寻优的改进萤火虫算法(IFA)优化ELM网络参数,形成优化的ELM训练方法IFA-ELM.该算法在保证预测精度的前提下,有效简化了网络规模,并提高了其泛化能力.利用该算法建立发动机风扇转速预测模型,基于该模型,采用梯度下降法在线调整PID参数,提升发动机动态性能.数字仿真验证表明,与常规PID控制相比,基于IFA-ELM的自适应PID法调节时间减少了0.2~1.4s,超调量降低了0.2%~1.5%,验证了该控制方法的有效性.

关键词: 航空发动机, PID, 极端学习机, 萤火虫算法, 自适应控制

Abstract: For the strong nonlinear and variable parameters properties of high bypass ratio turbofan engine, an adaptive PID control method based on optimized extreme learning machine (ELM) was proposed to predict the engine's parameters. To improve the prediction accuracy and the real-time property of ELM, an improved Firefly algorithm (IFA) for multi-peak optimization was adopted to optimize the network parameters of the ELM, and formed an optimized ELM training method IFA-ELM. Under the premise of ensuring prediction accuracy, the algorithm effectively simplified the network scale and improved its generalization capability. The engine fan speed prediction model was built by this algorithm, and gradient descent method was adopted to adjust the PID parameters online based on the model to improve the dynamic performance of engine. Digital simulation results show that compared with conventional PID control, IFA-ELM based adaptive PID method shortens the settling time by 0.2~1.4s, and reduces the overshoot by 0.2%~1.5%, which demonstrates the effectiveness of the proposed control method.

Key words: aero-engine, PID, extreme learning machine, Firefly algorithm, adaptive control

中图分类号: 


版权所有 © 《北京航空航天大学学报》编辑部
通讯地址:北京市海淀区学院路37号 北京航空航天大学学报编辑部 邮编:100191 E-mail:jbuaa@buaa.edu.cn
本系统由北京玛格泰克科技发展有限公司设计开发