中文摘要：

为提高电压预测控制优化问题的求解效率,该文在中心控制模式下,建立基于动态方程的电力系统电压预测模型,应用Gramian平衡降阶技术对预测模型进行降阶处理,构建基于平衡Gramian的电力系统电压多步预测–滚动优化模型。使用面向目标的对偶低秩Cholesky因子交替方向隐方法加快模型降阶的速度。应用原对偶非线性变尺度方法和Warm Start技术求解优化问题,以减少迭代次数。文中还分析相关参数设置和电压设定值的下发控制问题,给出电压预测控制的实施流程。仿真表明,所提方法可极大缩短求解优化问题的时间,具有低预测步数的计算复杂度和高预测步数的控制性能。该方法能够及时响应系统动态变化,维持电压稳定,保证电力系统安全运行。

英文摘要：

In order to improve the solving-efficiency of voltage predictive control optimization problem, this paper formulated power system voltage multi-step prediction and rolling optimization model in the mode of central control. In this process, this paper built the voltage predictive model based on dynamic equations and applied Gramian truncation model order reduction to dealing with the predictive model. The method of goal-oriented dual low rank Cholesky factor alternation direction implicit was used to get reduction model quickly. In order to decrease iterative times of optimization, the primal-dual nonlinear rescaling method and warm start technique were used. Moreover, this paper analyzed the parameter setting and the sending of voltage setting values and gave the implementation process of voltage predictive control. Simulation results reveal that this method can greatly decrease the optimization calculation time and have the complexity with low predictive step and control properties with high predictive step. Meanwhile, this method can respond the predictable dynamic change of the system in advance to keep voltage stable and make power system safe.