旋翼多维振动最优调整方法

针对直升机旋翼系统非线性、难以建模的特点,采用径向基函数(RBF,Ra-dial Basis Function)神经网络建立直升机旋翼动平衡调整模型.根据约束条件以直升机机身振动值作为目标函数建立适应度函数,以旋翼系统的调整参数为优化变量,进行神经网络学习和优化.利用粒子群优化(PSO,Particle Swarm Optimization)算法对适应度函数进行寻优,获得当直升机振动最小时的桨叶的调整参数.实验结果表明:PSO算法寻优效率方面高于遗传算法;RBF神经网络和PSO算法相结合可以有效地实现直升机旋翼动平衡调整.

Optimal adjustment method of rotor multi-dimensional vibration

Considering characteristics of non-linear and difficult to model for the rotor system, helicopter rotor dynamic balancing adjustment models was established by radial basis function(RBF)neural network. According to the constraints, the fitness function was established by using the helicopter vibration as objective function and the optimization variables were used by the adjustment parameters of rotor. The radial basis function neural network learning and optimization were used by the helicopter vibration and the adjustment parameters of rotor. Particle swarm optimization(PSO) algorithm was used to make a global optimization to find the suitable rotor adjustments corresponding to the minimum vibrations. The experimental results indicate that the particle swarm optimization algorithm is higher than the genetic algorithm in the aspect of efficiency optimization and the radial basis function combined with the particle swarm optimization algorithm can effectively achieve the helicopter rotor dynamic balance adjustment.