航空人-机复杂系统极值风险评估稳定性分析

Stability analysis of aeronautic man-machine complex system risk evaluation by extremum theory

针对航空人-机复杂系统极值风险评估中存在不确定性、难于验证的理论问题,建立了综合考虑'物理特性'、'随机性'和'非线性'的驾驶员-飞机系统和故障数学模型,通过大量仿真数据分析样本数量、评估次数以及优化算法选取对评估稳定性的影响。为提高极值分布中样本数据序列的拟合精度,针对综合评估模型提出一种随机自适应进化粒子群算法(random adaptive evolutionary particle swarm optimization,RAE-PSO)算法,实现了收敛速度和拟合精度的提高。提出了提高航空人-机系统风险评估稳定性、降低评估误差的方法步骤,并通过案例验证了方法的有效性。

Risk evaluation of a man-machine complex system based on extremum theory is stochastic and is difficult to validate. A pilot-aircraft complex system and failure mathematical model considering physical, stochastic and nonlinear factors are established. The influences of extremum number, assessment time and fitting algorithm on evaluation stability are analyzed based on a large number of simulation tests. To improve the fitting precision of sample sequence in extremum distribution, a random adaptive evolutionary particle swarm optimization （RAE-PSO） algorithm is proposed based on the synthesis evaluation model. The results show that the random adaptive evolutionary particle swarm optimization algorithm has a better convergence and precision. The approach for improving evaluation stability and precision is proposed. Simulation results verify the effectiveness of the proposed method.