静止无功发生器递归神经网络逆动力学控制

INVERTDYNAMIC CONTROL OF ASVG BY USING DIAGONAL RECURRENT NEURAL NETWORK

根据对角递归神经网络的特点分析了其非线性映射能力，并根据新型静止无功发生器（ＡＳＶＧ）的特点提出了ＡＳＶＧ递归神经网络逆动力学控制。所构造的控制器由于可以实现ＡＳＶＧ电力系统非线性逆动力学特性的学习，因而根据期望的控制效果可以获得期望的控制量。理论分析与仿真实验表明，这种控制器可以对ＡＳＶＧ实现良好的控制。

The nonlinear mapping ability of diagonal recurent neural network (DRNN) is analyzed. According to the characteristics of ASVG, the invert dynamic controller of ASVG by using DRNN is proposed. The invert dynamic controller can master the nonlinear invert dynamics of power system with ASVG. By inputting the desired voltage on connecting point to the controller, the desired control signal can be achieved. Theoretic analysis and simulation results show that the proposed controller can realize favorable control of ASVG.