论文部分内容阅读
提出了一种在保证优化效果前提下提高发动机在线性能寻优实时性的方法,即一方面基于相似理论建立了适用于非加力状态的发动机复合稳态模型,在优化过程中替代原部件级模型以缩短优化计算时间,另一方面采用先进的可行性序列二次规划算法寻求全局最优解,以提高优化精度。最终实现了上述方案在最小油耗模式及最大推力模式下的数字仿真及半物理模拟验证,相比于传统的基于发动机部件级模型进行性能寻优控制,其优化精度相当,但优化实时性得到大幅度提高,从仿真结果可以看出上述方案的有效性及可行性。
A method to improve the on-line performance of the engine under the precondition of ensuring the optimization effect is proposed. On the one hand, based on the similarity theory, a composite steady-state model of the engine suitable for the non-forcing state is established, replacing the original part level Model to shorten the optimization calculation time, on the other hand, the advanced feasible sequence quadratic programming algorithm is used to seek the global optimal solution to improve the optimization accuracy. Finally, the numerical simulations and semi-physical simulations of the above schemes in the minimum fuel consumption and the maximum thrust modes are finally carried out. Compared with the traditional performance optimization control based on the engine part-level model, the optimization accuracy is quite good, but the real-time optimization is large The magnitude of the increase, the simulation results can be seen from the effectiveness of the above programs and feasibility.