电力系统连锁故障的能量演变特征及其防御策略研究

Cascading failure is always of the great concern to modern power systems. In the process of failures propagation, some complex phenomenons such as power flow transfer, oscillation, voltage variation, usually appear in different time scales. With the evolution of structure and parameters state, the power system eventually moves into catastrophic criticality. The popular analysis methods of cascading failures generally focus on the specific parameters, frequency, voltage or active power for instance, but fail to study and grasp the more integrated evolution situation and its significant features. A novel research framework of cascading failures is proposed in this project. A series of important electrical parameters are included into Lyapunov energy theory. Moreover quasi-steady and transient energy function models are derivated with clear physical meanings. By performing alternate computing of power flow and time-domain simulation, the value of different energy functions are obtained, which are designed to study the evolution rules and other important characteristics of disturbances spread. Controlling impact factors based on energy functions are proposed to identify the most severe situation among voltage variation, overload and active power oscillation. Three key performance indices, namely energy accumulation factor, entropy distribution, and energy density are defined to quantify and research dynamic response of power system suffering cascading perturbations in the viewpoint of structure-preserving energy theory. Based on the two kinds of energy functions and feature performance indices, a new algorithm of reducing cascading failure risk and severity by mitigating integrated energy effect is planned to be studied. The basic principles for preventive control strategy decision of generation redispatch and load shedding are explained and tested. A number of benchmark cases and practical power system data are prepared as simulation example to prove and validate the proposed theory. This research project is expected to provide a new academic support for power system security defense in both planning and operation level.

连锁故障是现代大型电力系统必须面对的重大风险。事故扩散往往伴随着潮流转移、功率振荡、电压波动等诸多复杂现象，随着结构与状态不断演进出现"一触即溃"的临界状态。然而当前研究多关注某项电气参数，失去了考察系统宏观演变趋势和关键特征的功能。提出将连锁故障过程中涉及诸多电气参数的系统状态统一至Lyapunov能量框架下的研究思路，建立面向状态演变的暂态与准稳态能量函数，采用相继事件间2类能量指标交替计算方式，探索连锁故障中系统能量的演变规律，利用相关能量项从电压波动、过载、振荡3种现象中判定主导冲击因素，并从累积程度、熵分布、单位密度3个角度揭示随故障扩散的关键能量响应特征。在此基础上，提出利用削弱能量综合效应来降低连锁故障风险、防御系统崩溃的可能性，阐明降低能量效应的发电再调度与切负荷策略基本原理，并通过标准和实际系统仿真算例进行验证。本研究有望为大电网安全防御工具提供新的理论支撑。

连锁故障是现代大型电力系统必须面对的重大风险。 事故扩散往往伴随着潮流转移、功率振荡、电压波动等诸多复杂现象， 随着结构与状态不断演进出现“一触即溃”的.临界状态。 然而当前研究多关注某项电气参数， 失去了考察系统宏观演变趋势和关键特征的功能。 提出将连锁故障过程中涉及诸多电气参数的系统状态统一至 Lyapunov 能量框架下的研究思路， 建立面向状态演变的暂态与准稳态能量函数， 采用相继事件间 2 类能量指标交替计算方式， 探索连锁故障中系统能量的演变规律， 利用相关能量项从电压波动、过载、振荡3 种现象中判定主导冲击因素， 并从累积程度、熵分布、 单位密度 3 个角度揭示随故障扩散的关键能量响应特征。 在此基础上， 提出利用削弱能量综合效应来降低连锁故障风险、防御系统崩溃的可能性， 阐明降低能量效应的发电再调度与切负荷策略基本原理， 并通过标准和实际系统仿真算例进行验证。本研究有望为大电网安全防御工具提供新的理论支撑。