论文部分内容阅读
文中依据T-S模型的思想,提出了一种加权最小二乘支持向量机辨识算法.它采用模糊c均值(FCM)聚类确定规则数目,通过Gauss型函数将原输入输出空间分成若干子空间,在子空间中使用最小二乘支持向量机(LS-SVM)拟合获得子模型,然后由一个权重机制合成这些子模型,得到系统的模型.文中使用该方法去辨识关键反馈变量难以获得的非线性逆系统.为了得到这类逆系统的有效建模数据,采用了联合逆系统方法.仿真结果表明,加权最小二乘支持向量机辨识方法是有效的,它能够实现这类非线性逆系统的辨识,而且拟合误差平稳,波动幅度小,拟合精度和泛化能力都较好.
According to the idea of TS model, a weighted least squares support vector machine (SVM) identification algorithm is proposed, which uses fuzzy c-means clustering to determine the number of rules, divides the original input-output space into subspaces by Gaussian type function, Sub-models are obtained by LS-SVM fitting in sub-space, and then the sub-models are synthesized by a weighting mechanism to obtain the model of the system.The paper uses the method to identify the nonlinearity which is difficult to obtain by the key feedback variables Inverse system.In order to obtain the effective modeling data of such inverse system, a joint inverse system method is used.The simulation results show that the method of weighted least square support vector machine identification is effective and it can realize the identification of such nonlinear inverse systems , And the fitting error is stable, the fluctuation range is small, the fitting accuracy and generalization ability are better.