Abstract:

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.