存在约束的线性网络灵敏度计算

Sensitivity Calculation of Linear Network under Constraints

灵敏度是利用系统中某些物理量的微分关系，来获得因变量对自变量敏感程度的方法。它在系统稳定分析与控制中得到了广泛的应用。传统灵敏度方法没有考虑系统的约束条件，造成灵敏度指标不能反映实际物理状况。本文把线性网络中实际存在的约束分为两类：等式约束和不等式约束，并通过实例计算了不同约束条件下的灵敏度。同时，由于功率网络灵敏度计算是基于运行点处的线性化，它与线性网络灵敏度计算本质上是一致的，因此，本文方法也为存在约束的功率网络灵敏度计算奠定了理论基础。

Sensitivity method is a method to obtain the responsivity of dependent variable to independent variable by utilizing the differential relations of some physical quantities in a system. It has been widely used in analysis and control of system stability. Traditional sensitivity methods don＇ t take into account constrained conditions so that sensitivity index cannot reflect actual physical condition. This paper categorizes the constraints existing in linear network into two types, equality constraints and inequality constraints, and calculates the sensitivity under different constraints through an example. Sensitivity calculation of power network is based on the linearization at operation points with the same nature as that of linear network. Therefore, this paper also lays a theoretical foundation for sensitivity calculation of power network under constraints.