基于动态优化数据的航空发动机过渡态控制方法

针对一般稳态燃油前馈PID算法在发动机过渡态过程中控制效果差的问题，提出了一种基于动态优化数据的涡轴发动机瞬态控制方法。采用带约束限制的序列二次规划（SQP）优化算法采集包线范围内各点的过渡态参数变化数据作为样本数据，利用稀疏化的最小二乘支持向量机（LSSVM）对样本数据进行训练、测试，将训练得到的LSSVM模型作为前馈与PI构成闭环控制器共同对涡轴发动机进行过渡态控制。通过对民用涡轴发动机部件级模型的包线内某两点不同功率水平进行仿真，结果表明，过渡态过程中动力涡轮转速超调量与下垂量均小于0.4%，稳态误差为0，动力涡轮转速稳定时间小于2s，各参数均未超限，因此，该控制器能有效提高涡轴发动机过渡态控制效果，实现对参数的限制管理。

Transient Control Method of Aero-Engine Based on Dynamic Optimization Data

To solve the problem that the general steady-state fuel feedforward PI algorithm has poor control effect during the transition state of the engine, a transient control method for turbo-shaft engine based on dynamic optimization data is proposed. The sequence quadratic programming (SQP) algorithm with constraint restrictions is used to collect the data of transition state parameter variations at each point in the flight envelope as the sample data, and the sample data is trained and tested by the sparse least squares support vector machine (LSSVM). The trained LSSVM model and PI constitute a closed loop controller to control the turbo shaft engine in transition state. By simulating different power levels of two points in the envelope of a civil turbo-shaft engine component-level model, the results show that the overshoot and sag are less than 0.4% during the transition state, the steady-state error is 0, and the power turbine speed stabilization time is less than 2s, and the parameters do not exceed the limits. Therefore, the controller can effectively improve the transition state control effect of the turbo-shaft engine and realize the limited management of parameters.