Research On Optimal Control Method Of Turbofan Engine Based On Mathematical Programming

With the continuous development of modern science and technology and the continuous improvement of production requirements,the application of optimal control theory has become increasingly widespread in many fields such as national defense,aviation,and aerospace,and it has gradually become the core of modern control theory.The problem of optimal control of civil aero-engines is to study the selection of a suitable optimal control method for an aero-engine system in order to optimize certain performance indicators of the system.This thesis focused on the topic of optimal control methods for aero-engines,and mainly studied the linear quadratic regulation?LQR?method and the sequential quadratic programming?SQP?algorithm among the different methods of solving optimal control problems.The main research contents are as follows:Firstly,in the steady state or small deviation state of turbofan engines,the state variable model at the cruise point of the engine was obtained by using the component-level model.Aiming at the core problem of the LQR method,that is,the selection of Q and R coefficient matrices,a kind of hybrid algorithm GA-PSO was used in the LQR control strategy.The results showed that the proposed GA-PSO hybrid algorithm has realized the optimization of the LQR regulation system and tracking system.The low-speed rotor speed output curve of the turbofan engine could effectively and quickly converge,and the performance indicators of the curve including rise time,adjustment time,and overshoot amount,etc.have been optimized.Compared with the genetic algorithm?GA algorithm?,the GA-PSO algorithm is used to optimize the turbofan engine LQR control strategy,which can effectively reduce the number of optimization iterations and reduce the calculation time.Secondly,in order to figure out the effect of the selection of fitness function on the optimization of the LQR control strategy of the GA-PSO algorithm,three different fitness functions were proposed and compared for simulation.The results showed that the selection of fitness function of GA-PSO algorithm will have an influence on the effect of LQR control after optimization.It could be concluded that the optimization effect of GA-PSO algorithm using fitness function f₃ is better than that of f₁ and f₂.Finally,under a wide range of turbofan engine acceleration and deceleration conditions,the control strategy of turbofan engine acceleration and deceleration using the GA-SQP algorithm was proposed,based on the component-level model including dynamic and stable model of turbofan engine.The results showed that this method can maximize the potential and improve the acceleration and deceleration performance of turbofan engines,while satisfying the engine's various constraints.At the same time,when compared with the control strategy under the simple SQP algorithm,the GA-SQP algorithm can reduce the dependence on the initial value,optimize the calculation efficiency,and reduce the calculation time.