一种基于云模型优化的飞行器参数辨识算法

An Aircraft's Parameter Identification Algorithm Based on Cloud Model Optimization

极大似然估计方法(ML)在飞行器参数辨识中得到了广泛应用,该方法需要预先推导灵敏度方程,进而求解灵敏度矩阵,在应用过程中比较繁杂,且容易陷入局部最优。提出一种基于云模型优化的飞行器参数辨识算法,根据极大似然估计原理,利用云模型的优化理论对极大似然函数进行优化,从而得到待辨识参数值。该算法不必推导灵敏度矩阵,对初值要求不高,应用便捷,且保留了云模型优化的特点,收敛速度较快、不易陷入局部最优。以Twin Otter飞机为例对算法进行验证。结果表明:算法易于实现、辨识结果精度较高、收敛速度较快,不易陷入局部最优。

The Maximum Likelihood（ML） estimation method has been extensively applied to identifying the pa- rameters of an aircraft, but it has to derive sensitivity equations in advance and solve sensitivity matrices, thus being inconvenient for its application and easily reaching locally optimal solutions. An aircraft＇s parameter iden- tification algorithm by combining the cloud model optimization with the ML estimation method is proposed. The algorithm uses the global optimization algorithm based on the cloud model to optimize the ML function, and then obtains the identified parameters. The algorithm has neither to derive sensitivity equations nor to calculate sensitivity matrices. A Twin Otter airplane is taken as an example to verify the method. The numerical results show that the parameter identification algorithm is easy to implement, has good identification precision and fast convergence and is not likely to reach locally optimal solutions.