带矢量推力的涡扇发动机实时数学模型及智能控制研究

【作者】 马会民；

【导师】 樊思齐；

【作者基本信息】 西北工业大学 ， 航空宇航推进理论与工程， 2001， 硕士

【摘要】 建立发动机实时数学模型是建立发动机控制系统半物理仿真实验平台的重要基础。本文通过对涡扇发动机部件的热力计算过程的深入分析，提出了一种全新的实时模型建模方法，该方法通过将发动机部件参数间的关系通过解析式表示，从而避免了部件计算过程中复杂而又耗时的迭代过程。这样，不仅使模型完全达到实时性要求，而且使得计算过程思路更加清晰，易于理解。另外，该建模方法还有通用性强，精度高的优点。该模型可以计算多种控制规律下的涡扇发动机稳态和动态过程，也可以进行涡扇发动机多变量控制研究。通过与非实时模型对比仿真，表明该模型在全飞行包线内均有很好的收敛性和满意的精度。 根据矢量喷管流场计算结果，本文建立了带推力矢量的涡扇发动机实时数学模型，该模型可以模拟在不同的矢量角度发动机的工作过程，也可以模拟矢量喷管角度改变时对发动机的影响。因此该模型可作为矢量喷管控制模型。 智能控制的蓬勃发展使其已渗透到各个领域，研究智能控制在航空发动机上的应用对提高整个推进系统的性能具有重大意义。本文，利用智能控制中十分活跃的模糊控制、模糊神经网络和遗传算法研究了涡扇发动机加速过程控制。首先，设计了一种连续型的模糊控制器，接着利用遗传算法对模糊控制器中的三个量化因子进行了寻优。最后，本文还研究了模糊神经网络控制在航空发动机上的应用。仿真结果表明，控制系统表现出良好的性能。研究结果均表明，智能控制在航空发动机控制中具有广阔的应用前景。更多还原

【Abstract】 Real-time mathematical model of the aero-engine is the basis for the engine half physical simulation test platform. In this paper, a novel method of establishing the real-time model is proposed through deeply analyzing to the thermal calculation process of the aero-engine components. In this method, the relation of the thermal parameters of the parts of the engine is presented by analytical formula. So, the complex and time-consuming iterative calculation process is avoided, which makes the calculating process not only reaches the real-time requirements but also becomes clearer and more comprehensible. In addition, this method has good quality of high precision and broad applicability. The model established by this method can be used to calculate the steady and dynamic characteristic of the turbo-fan engine and can also be used to study the MIMO control of the aero-engine. Compared with the calculating results of the non-real time model of the engine, the real-time model has good convergence and satisfying calculation precision within the full flight envelope. According to the results of the CFD a real-time thrust-vectoring nozzle turbofan engine model is established. The model can simulate the working process of the engine in different deflected angle and can be applied to the control of the vectoring nozzle. The intelligent control theory has penetrated into many fields. And to study the application of the intelligent control to the aero-engine control system is very important for the improvement of the performance of the propulsion system. In this paper, three different intelligent control methods, fuzzy logic control, neuro-fuzzy network and genetic algorithm (GA) were used to study the accelerating process control of the turbofan engine. Firstly, a P-D type C-F controller is designed. Secondly, the GA is applied to optimize the three important parameters of the fuzzy Logic controller. Finall~. a neuro-fuzzy controller is studied. Simulation results show that the control systems behave very well, which indicates that the intelligent control ?III ? has wide prospect in the aspect of control of the aero-engine.更多还原