基于主从问题的电力系统最优主动解列

发布时间：2018-11-16 20:20

【摘要】：系统主动解列的完整模型为大规模非线性混合整数规划问题,大多采取完全分解方法,因而只能求得近似解,针对此问题提出了基于主从问题交替求解电力系统最优主动解列断面的新策略,从而求得更优解。该策略基于图背包理论(connected graph constrained knapsack problem,CGKP)将完整主动解列模型转化为主从问题;主问题为图的最优平衡分割问题,采用CGKP技术进行求解;从问题为基于直流最优潮流(optimal power flow,OPF)的调度问题,采用OPF技术进行求解;主从问题之间通过节点负荷的调节量实现耦合。通过主从问题之间的交替迭代获得更优的解列方案,同时也使得解列方案更接近于完整模型的最优解。算例分析表明,该算法相对于其他近似求解策略,所获得的解列方案可使系统总切机切负荷量更少,从而证明了该方法的有效性和可行性。
[Abstract]:The complete model of active solution of the system is a large-scale nonlinear mixed integer programming problem, most of which are completely decomposed, so the approximate solution can only be obtained. In order to solve this problem, a new strategy of alternately solving the optimal active discontinuous section of power system based on master-slave problem is proposed, and a better solution is obtained. Based on the graph knapsack theory (connected graph constrained knapsack problem,CGKP), the complete active solution model is transformed into a master-slave problem, and the main problem is a graph optimal equilibrium partition problem, which is solved by CGKP technique. The slave problem is a scheduling problem based on DC optimal power flow (optimal power flow,OPF), which is solved by OPF technology, and the master-slave problem is coupled by adjusting the load of nodes. A better solution is obtained by alternating iteration between master and slave problems, and the solution is closer to the optimal solution of the complete model. Numerical examples show that compared with other approximate solution strategies, the proposed algorithm can reduce the total cutting load of the system, which proves the effectiveness and feasibility of the method.
【作者单位】：智能电网教育部重点实验室(天津大学);中国电力科学研究院;山西电力科学研究院;东北电力调度通信中心;
【基金】：国家自然科学基金项目(51177107)~~
【分类号】：TM712

【参考文献】