旋翼动平衡调整中GRNN网络和遗传算法的应用探究

针对传统旋翼调整方法没有考虑调整参数与振动信号之间的非线性关系，提出一种结合广义回归神经网络GRNN（GeneralRegressionNeura1Network）和遗传算法的旋翼调整方法，采用GRNN建立旋翼动平衡调整模型，以桨叶调整参数作为GRNN输入，以旋翼转轴3个方向的加速度测量值和机身3个方向加速度测量值作为网络输出，建立调整参数与直升机振动信号之间的模型。以直升机振动作为目标函数，采用改进的遗传算法对桨叶调整参数进行寻优，获得直升机振动最小时的桨叶的调整量。飞行实验表明，通过1到2次飞行调整，可使3个方向机身振动（旋翼的一阶振动）为最小，完成旋翼的动平衡调整。

Considering traditional adjustment method without calculating possible nonlinearity between rotor adjustments and fuselage vibration signals, a new rotor adjustmentmethod based on general regression neural network （GRNN） and genetic algorithm was was employed to mod P el resented. GRNN network the relationship between the rotor adjustments and the fusela taking rotor adjustment parameters as ge vibrations, the inputs and the acceleration measurements along the three axes of rotor shaft and the fuselage as the outpus. With helicopter vibration as an objective function, genetic algorithm was used to make a global optimizationto find the suitable rotor adjustments corresponding to the minimum vibrations. Flight test results indicate that proposed rotor adjust mentmethod can minimize fuselage vibration at fundamental rotor frequency along the three axes,with one or two flight adjustment , and the neural networks may be updated to include new data thus allowing the system to evolve and mature in the course of its use