基于支持向量机的电力系统紧急控制实时决策方法

发布时间：2018-03-05 14:10

本文选题：电力系统　切入点：紧急控制　出处：《中国电机工程学报》2017年16期 　论文类型：期刊论文

【摘要】：电力系统紧急控制是故障消除后保持系统稳定性、防止事故扩大的重要手段。相比于目前电网普遍使用的离线决策,实时紧急控制决策能够提供更准确有效的控制措施。但实时决策对计算速度的要求极高,传统方法无法胜任。基于支持向量机技术,该文研究了电力系统紧急控制实时决策方法,关键在于利用支持向量机从大量的仿真数据中挖掘系统稳定规则,并用于构造可以实时求取的稳定性约束,由此将大量的仿真搜索从实时阶段转移到离线阶段。该文首先建立紧急控制决策的优化模型;然后利用灵敏度方法将模型中的稳定性约束线性化;接着通过改造支持向量机的分类表达式将其用于拟合,提出能够实时求取的稳定裕度指标,实现灵敏度求解;最后在IEEE39节点系统中进行算例仿真,仿真结果显示:所提出的稳定性指标能够准确反映系统的稳定裕度,同时针对不同的失稳状况该文提出的决策方法能有效恢复系统稳定性。
[Abstract]:The emergency control of power system is an important means to maintain the stability of the system after the fault is eliminated and to prevent the expansion of the accident. Compared with the off-line decision, which is widely used in the power network at present, Real-time emergency control decision can provide more accurate and effective control measures. However, real-time decision making requires very high computational speed and the traditional method is not competent. In this paper, the real-time decision-making method of power system emergency control is studied. The key point is to mine system stability rules from a large amount of simulation data by using support vector machine (SVM), and to construct stability constraints that can be obtained in real time. Therefore, a large number of simulation searches are transferred from real-time stage to off-line stage. Firstly, the optimization model of emergency control decision is established, and then the stability constraints in the model are linearized by sensitivity method. Then, by modifying the classification expression of support vector machine to fit it, the stability margin index which can be obtained in real time is put forward, and the sensitivity is solved. Finally, the simulation example is carried out in the IEEE39 node system. The simulation results show that the proposed stability index can accurately reflect the stability margin of the system, and the decision method proposed in this paper can effectively restore the stability of the system.
【作者单位】：电力系统及发电设备控制和仿真国家重点实验室(清华大学电机系);国家电网公司西南分部;中国电力科学研究院;
【基金】：国家电网公司科技项目清华大学电力系统国家重点实验自主课题(SKLD16Z08)~~
【分类号】：TM712;TP18

【相似文献】