一种商用大涵道比涡扇发动机无模型自适应性能优化控制方法

A Model-Free Adaptive Performance Optimization Control Method for High-Bypass Ratio Commercial Turbofan Engine

关键词:  涡扇发动机  性能控制  无模型自适应控制  敏感性分析  参数优化

Key words:  Turbofan engine  Performance control  Model free adaptive control  Sensitivity analysis  Parameter optimization

摘要:

针对大涵道比涡扇发动机全包线分段线性设计控制器的过程复杂、自适应能力差等问题，本文提出了一种发动机无模型自适应性能优化控制方法。基于无模型自适应控制（Model Free Adaptive Control，MFAC）算法设计了一种推力控制器以实现发动机全包线范围内任意工作点的自适应控制，解决了传统逐段全程设计控制器的繁琐过程；为了获得实现最优性能的控制器参数，本文通过控制变量法对控制器调节参数进行了敏感性分析，获得了敏感性参数和变化规律；然后，设计了基于收缩粒子群优化算法（Constriction Particle Swarm Optimization，CPSO）的敏感性参数寻优方法，实现了控制器参数的自整定。仿真结果表明：相比当前抗积分饱和PI控制，所述控制方法从慢车到最大状态的动态调节时间缩短了4.2%，提高了动态加速性能，且无超调，无稳态误差。同时，克服了发动机未知非线性动态变化，实现了全包线全状态下任意工作点的性能优化控制，并始终保持良好的控制效果和鲁棒性，具有较好的工程应用前景。

Abstract:

A model-free adaptive performance optimization control method is proposed to deal with the problems of complex processes and poor adaptive ability of the piecewise linear design controller of a high-bypass ratio turbofan engine in the full envelope. Firstly， based on the model-free adaptive control （MFAC） algorithm， a thrust controller is designed to realize the adaptive control of any operating point in the whole envelope of the engine. It solves the cumbersome process of traditional controller design step by step. Secondly， in order to get the controller parameters to achieve optimal performance， the sensitivity analysis of the controller tuning parameters is carried out through the control variates method， and the sensitivity parameters and change rules are obtained. Then， a sensitivity parameter optimization method based on contraction particle swarm optimization（CPSO） is also designed to realize the self-tuning of controller parameters. Finally， the simulation results show that compared with the current anti-windup PI control， the dynamic tuning time from idle to maximum state is shortened by 4.2%， and the dynamic acceleration performance is improved without overshoot and steady-state error. Meanwhile， it overcomes the unknown nonlinear dynamic changes of the engine， realizes the performance optimization control of any operating point in the full envelope and full state， and always maintains good control effect and robustness. It has a good engineering application prospect.