基于飞行参数的直升机桨叶载荷评估方法

为了解决传统应变计测量获得直升机桨叶载荷谱存在的周期长、研制成本高、占用空间等问题,提出了采用BP神经网络建立飞行参数与桨叶载荷的关系模型。首先通过相关性分析验证了直升机桨叶载荷与操纵、姿态、过载等飞行参数存在着较强的相关性,其次通过悬停、爬升、转弯等直升机飞行状态下的桨叶载荷实测数据,确定了模型中的参数值,并对模型进行了验证,最后将模型应用于平飞状态下桨叶载荷的评估。结果表明,通过飞行参数来评估桨叶载荷是可行的,评估模型的评估精度高,评估值和实测值的相对误差为6.67%,满足工程评估精度的要求。所提方法不仅可以解决传统方法的不足,还可应用于大机动、大过载试飞科目的桨叶载荷预评估,可有效提高飞行安全水平,具有重要的工程应用价值。

Evaluation method of helicopter blade load based on flight parameters

In order to solve the problem of long period,high development cost and space occupation of helicopter blade load spectrum measured by traditional strain gauge,the BP neural network was used to establish the relationship model between flight parameters and blade load. Firstly,the correlation analysis was conducted to verify the strong correlation between helicopter blade load and flight parameters such as control,attitude and overload. Then,the parameters of the model were determined by the measured data of the blade load in the helicopter flight state such as hovering,climbing and turning,and the model was verified. Finally,this method was applied to the evaluation of blade load in forward flying state. The results show that it is feasible to evaluate blade load by flight parameters,and the evaluation model has high evaluation accuracy. The relative error between the evaluation value and the real value is 6.67,which meets the requirements of engineering evaluation accuracy. The method can not only solve the shortcomings of traditional method,but also be applied to the pre-evaluation of blade load in large maneuver and large overload flight test subjects to improve flight safety,which has a high value in engineering application.