论文部分内容阅读
为了满足变循环发动机性能寻优控制需求,提出了一种基于序列二次约束二次规划(SQCQP)算法的性能寻优控制方法,通过罚函数将二次约束二次规划(QCQP)子问题转化为适应度函数,并提出一种改进的微分进化算法(IDE)求解QCQP子问题,以获得最优的搜索方向。与序列二次规划(SQP)算法相比,所提出的基于改进微分进化算法求解子问题的序列二次约束二次规划算法(IDE-SQCQP)能在更少的迭代次数下寻到更优的解。将IDE-SQCQP算法应用于变循环发动机的性能寻优控制中,数字仿真结果表明在最大推力寻优控制中,IDE-SQCQP算法用时比SQP算法减少16.81%,优化效果提升了21.50%,在最小油耗寻优控制中比SQP算法用时减少14.90%,优化效果提升了31.03%,达到了算法提出的目的。
In order to meet the demand of variable cycle engine performance optimization control, a performance optimization control method based on sequential quadratic constrained quadratic programming (SQCQP) algorithm is proposed. The quadratic constrained quadratic programming (QCQP) sub-problem is transformed by penalty function Is a fitness function, and an improved differential evolution algorithm (IDE) is proposed to solve QCQP sub-problems to get the optimal search direction. Compared with the Sequential Quadratic Programming (SQP) algorithm, the proposed Quadratic Quadratic Quadratic Quadratic Programming Quadratic Quadratic Programming (IDE-SQCQP) algorithm based on the improved differential evolution algorithm can find better performance with fewer iterations solution. The IDE-SQCQP algorithm is applied to the performance optimization control of variable cycle engine. The results of numerical simulation show that IDE-SQCQP algorithm uses 16.81% less time and optimizes 21.50% than SQP algorithm in the maximum thrust optimization control, Compared with the SQP algorithm, the optimization of fuel consumption optimization can reduce 14.90% of the time and the optimization effect by 31.03%, which achieves the goal of the algorithm.