用多目标FSQP算法实现涡扇发动机多变量非线性控制  

On a Novel Multi-Objective FSQP Algorithm for Multi-Variable Nonlinear Control of Turbofan Engine

作　　者：戚学锋[1] 樊丁[1] 陈耀楚[1] 

机构地区：[1]西北工业大学动力与能源学院,陕西西安710072

出　　处：《西北工业大学学报》2004年第5期644-648,共5页Journal of Northwestern Polytechnical University

基　　金：航空科学基金 (99C5 30 2 2 )资助

摘　　要：涡扇发动机是一个复杂的非线性系统 ,用常规的多变量控制方法难以获得理想的控制效果。提出了一种自适应变权重的多目标 FSQP( Feasible Sequential Quadratic Programming)算法 ,以及基于此算法的涡扇发动机多变量非线性控制方法。该控制方法可以有效地解决涡扇发动机的多变量非线性控制问题 ,同时在满足控制过程中各项约束的前提下 ,达到调节时间最短的目标。Traditional methods for multi-variable control of turbofan engine are, in our opinion, not quite satisfactory. Having failed to find any paper in the open literature on applying the computationally efficient FSQP (Feasible Sequential Quadratic Programming) algorithm to multi-variable control, we have worked out such an algorithm for control of turbofan engine. The core idea of our algorithm consists of four parts: (1) we convert the multi-variable control problem of turbofan engine into the problem of multi-variable nonlinear optimization, thus obviating any need for linearization of mathematical model of turbofan engine;(2) in our multi-objective FSQP algorithm, relatively reasonable self-adaptive weights are assigned to different objective functions;(3) the adjusting time of control process should be and is made quite short;(4) all constraints are always satisfied during the control process. Simulation results indicate preliminarily that our novel multi-objective nonlinear-control algorithm is feasible and gives quite good performance.

关 键 词：涡扇发动机 多变量非线性控制 FSQP算法 

分 类 号：V233.7[航空宇航科学与技术—航空宇航推进理论与工程]