针对同步发电机组这个非线性、工况复杂的多变量控制对象,研究了一种基于支持向量机的发电机组逆模型控制方法。根据采集到的输入输出样本数据,首先由支持向量机网络来辨识发电机组的逆模型,以此作为逆控制器设计的基础。随后,分析了2种不同的逆模型控制器,即直接逆模型控制和具有线性补偿的逆模型控制。仿真研究表明,支持向量机逆模型控制器具有优良性能,而且线性补偿逆模型控制比直接逆模型控制的性能更优,鲁棒稳定性更好。
Synchronous generator is a nonlinear, fast acting, multiple-input multiple-output device operating over a wide range of varying conditions. This paper investigated the basis for the inverse model control strategies for synchronous generator based on support vector machines (SVM). Firstly, SVM was used to identify the inverse model of the plant from the input-output data, and this inverse model was served as the basis for the inverse controller design. Then two different inverse model control schemes namely the direct inverse control and the linear-compensated inverse control were analyzed together. Simulations show that the proposed inverse controller has good performance while linear-compensated inverse control has better capability and stability than direct inverse control.