基于模糊聚类滚动筛选决策集的电压协调控制

Coordinated voltage control with CDSRS based on fuzzy clustering analysis

基于模型预测控制,提出一种决策集滚动筛选(CDSRS)的电压协调控制方法。通过求取预测周期内优化目标节点电压对控制变量的响应特性,将电压协调控制模型的求解转化为混合整数规划问题。考虑电压控制的局部性特点,以模型预测过程中求取的电压预测幅值和响应信息作为聚类特征指标,采用模糊聚类方法确定故障后的优化目标节点,在此基础上根据电压响应显著度滚动筛选决策集。仿真结果表明,该方法在取得全局协调控制效果的前提下,大幅降低了备选决策集规模,显著减少了优化计算时间,有效避免电压协调控制的决策集爆炸问题。

Based on the model predictive control,a method of coordinated voltage control with CDSRS （Candidate Decision Set Roiling Selection） is proposed. By obtaining the response characteristic of optimal objective node voltage to control variable in the prediction period,the task to solve the coordinated voltage control model is converted into that to carry out the mixed-integer linear programming. According to the locality of voltage control and with the predicted voltage amplitude and response information obtained during the model prediction as the clustering feature indexes,the fuzzy clustering method is applied to determine the optimal objective node after fault,based on which,the CDSRS is carried out according to the significance degree of voltage response. Simulative results show that ,under the premise of global coordinated control performance,the scale of candidate decision sets is significantly decreased and the time of optimization calculation is obviously reduced,effectively avoiding the problem of decision set explosion in coordinated voltage control.