基于ADEMO/D-ENS的飞控系统性能指标分配方法

Performance Indexes Allocation Method of Flight Control System Based on ADEMO/D-ENS

针对目前飞行控制系统设计中部件/组件性能参数的确定存在反复多次迭代的问题,对飞控系统性能指标的分配进行了研究。通过对性能指标分配过程进行建模,确定了分配过程属于多目标优化问题。基于Tchebycheff方法将多目标优化问题转化为单目标优化子问题集合,基于自适应差分进化算法得到的单目标优化子问题集合的最优解即为多目标优化问题Pareto最优解,同时采用惩罚因子保持差分进化算法种群的多样性。通过仿真与性能指标未分配的系统进行对比,结果表明分配后的系统具有更好的动态性和跟踪性,说明所提出的分配方法是正确的、可行的,并能够为工程应用提供一定的理论指导。

To solve the problem that the performance parameters of components/assemblies are repeatedly selected and replaced in the current flight control system design,the allocation of performance indexes of theflight control system is studied.By modeling the allocation process of the performance indexes,it was determined that the allocation process was a multi-objective optimization problem.The multi-objective optimization problem was transformed into a set of single-objective optimization sub-problems based on Tchebycheff method.The optimal solution of the set of single-objective optimization sub-problems based on adaptive differential evolution algorithm was the Pareto optimal solution of the multi-objective optimization problem.Meanwhile,penalty factors were used to maintain the diversities of the differential evolution algorithm population.Compared with the system with unallocated performance indexes,the results show that the allocated system has better dynamic and tracking performance,which illustrates the proposed allocation method is correct and feasible,and can provide certain theoretical guidance for practical application.