基于模糊聚类的模糊神经网络在非定常气动力建模中的应用

建立了一种基于模糊聚类的模糊神经网络模型.该模型利用模糊聚类技术确定系统的模糊空间和模糊规则数,利用BP算法调整模糊神经网络的权系数.应用该模型对某飞机模型做俯仰-滚转耦合运动的非定常气动力进行了辨识.结果表明,基于模糊聚类的模糊神经网络计算速度快,辨识结果与实验结果符合较好.用模糊聚类技术可以解决模糊神经网络的结构辨识问题,基于模糊聚类的模糊神经网络可以很好地用于复杂机动飞行的非定常气动力建模.

The application of FNN in unsteady aerodynamics modeling based on fuzzy clustering

In this paper a Fuzzy Neural Network (FNN) model based on fuzzy clustering is developed. The fuzzy space and the number of fuzzy rules of this model are defined by the fuzzy clustering method and weight coefficients of the model are adjusted by the BP algorithm. Using the model the unsteady aerodynamics of one aircraft in pitching-rolling motion is identified. The simulating results are agreement with the experimental results very well. The calculating process can be speeded up. It is suggested that the fuzzy clustering method can be used to design fuzzy neural network structures and the developed model can be used to identify the nonlinear unsteady aerodynamics of many complicated maneuvers.