摘要
研究飞行控制系统精度优化问题中,针对系统辨识模型阶次难以确定从而导致辨识精度难以满足要求,提出了一种通过将待辨识系统的非线性隐式方程进行离散化,建立非线性系统辨识模型,克服了传统辨识模型需要确定系统阶次的困难,所用系统辨识模型物理意义明确,算法实现简单。采用最小二乘支持向量机算法,选取合适的径向基核函数,以某型无人机为研究对象,用上述辨识方法进行仿真。仿真结果表明,上述算法具有良好的预报能力和较强的泛化能力,同时具有较高的辨识精度。
This paper put forward a new method, which builds the identification model by the method of discretiz-ing the nonlinear implicit equations of the system. The method can overcome the difficulties that the NARMAX identi-fication model determines the order of the system. Also the identification model has clear physical meaning and simple algorithm to realize it. Based on the least squares support vector machine algorithm, this paper madke the simulation research based on this identification method with a type of UAV as research object. The simulation result shows that the method has good prediction and strong generalization ability, also has high recognition accuracy.
出处
《计算机仿真》
CSCD
北大核心
2012年第9期50-52,135,共4页
Computer Simulation
关键词
系统辨识
最小二乘支持向量机
径向基核函数
飞行控制系统
System identification
Least square support vector machine (LS-SVM)
Radical basis function(RBF)
Flight control system(FCS)