径向基神经网络在非线性非定常气动力建模中的应用研究

The research of RBFNN in modeling of nonlinear unsteady aerodynamics

基于南航NH-2风洞中某飞机模型大迎角大振幅单自由度偏航、滚转及偏航-滚转耦合的谐波、阶跃运动实验数据,应用径向基神经网络,研究人工神经网络描述非线性非定常气动力特性的能力。研究结果表明,所建立的径向基神经网络模型的预测结果与训练数据和验证数据都符合得很好,说明神经网络建模方法可以有效地对高度非线性的气动力进行建模。研究还表明,用神经网络建立模型时所需要的风洞实验数据可以减少,从而提高风洞实验效率、减少风洞实验的时间和成本。

Based on the large amplitude harmonic oscillation and ramp motion experimental data of an aircraft model with yawing,rolling and yawing-rolling motion tested in the NH-2 wind tunnel,using the RBF Neural Network,the capability of Artificial Neural Network in modeling aerodynamics are researched.The results show that,there is a good agreement between the mathematically predicted results and the teaching data and the validating data.It indicates that neural network modeling method can be effectively used in the highly nonlinear aerodynamic modeling.The research results also show that,the test data can be reduced using the neural network aerodynamics model,so it can save the test time and costs.