考虑小干扰稳定约束的最优潮流求解

Computation of Optimal Power Flow with Small Signal Stability Constraints

针对电力系统已有控制器不能有效抑制低频振荡的情况,提出了一种通过潮流调整提高小干扰稳定性的新方法。它是在定义了小干扰稳定指标、介绍了指标灵敏度计算方法的基础上提出的考虑小干扰稳定约束的最优潮流(OPFSC)模型和算法。由于在模型中使用小干扰稳定指标及其灵敏度组成的代数不等式作为稳定约束条件,简化了计算;考虑了多种可控参数的调整,能有效提高系统的小干扰稳定性;根据指标灵敏度提出的主导控制节点概念及对优化过程中成本费用的修正使算法能够快速收敛至满足约束的最优解。最后通过对4机2区域系统的仿真验证了所提出模型和算法的有效性。

The security of large power system has been seriously threatened by low frequency oscillations. The method of power flow rescheduling could be applied to enhance the small signal stability when the poor damping can not be improved by controllers placed in power system. Based on above background, the small signal stability index is defined and the calculation method of the index sensitivity to voltage or reactive/active power is introduced based on structure preserved model. According to the sequence of index sensitivity, the concept of key controllable bus is proposed. Then a new type of model of Optimal Power Flow with small signal Stability Constraints （OPFSC） is presented to obtain the power flow rescheduling. In this model, the stability constraints represented by the differential-algebraic equations are converted to the algebraic inequality constraints of the small signal stability index and its sensitivity with respect to the control variables in this approach. Furthermore, by considering that load rescheduling in the model is as same as generation re-dispatching, the model improves the ability of small signal stability enhancement. The cost coefficient of the object function is dynamically adjusted according to the key controllable bus to accelerate the convergence of the OPFSC algorithm. The converted OPFSC problem can be easily solved by the nonlinear programming approach, which not only takes a satisfied convergence, but also can enhance the small signal stability effectively. As an illustration, the proposed method is applied to the 4-machine 2-area power system. The optimization results and simulations results both demonstrate the effectiveness of proposed model and algorithm .