电力系统优化控制对于改善电力系统运行性能指标具有重要意义，比如无功优化控制能够有效降低系统网损和改善电压质量，环境经济调度优化能有效达到节能减排的目的等等。关于电力系统优化控制的文献大都基于确定性模型，即假定负荷不变、控制对象可以严格得到精密的控制。但是，负荷实际上不仅是波动的而且还具有预测误差，控制对象的控制精度往往也只能保证其大致符合控制预期。 考虑到负荷的不确定性和控制可能存在的偏差，单纯从数学优化的角度得到的最优控制策略有时可能对于上述不确定性和偏差过于敏感，从而在实际当中表现出很差的性能。为此，需要对控制策略的鲁棒性进行评估，以便于从若干候选的优化控制策略中遴选出对不确定性和偏差的承受能力足够强的控制策略作为实施策略。 为了对电力系统优化控制策略的鲁棒性进行评估，综合考虑了控制变量的可实现精度和状态变量的不确定性，建立了对控制策略的鲁棒性进行评估的指标，论述了单性能指标优化问题和多性能指标优化问题的控制策略鲁棒性满足要求的判断方法，提出一种基于抽样优化的控制策略鲁棒性评估方法，给出了若干多目标优化优解的遴选方法，讨论了在若干控制策略的鲁棒性均满足要求或均不满足要求情况下进一步对各个控制策略的鲁棒性优劣进行比较的方法。针对单性能指标优化问题，以IEEE30节点系统无功优化和6机系统经济调度算例为例，对采用优化方法得出的若干个候选控制策略进行了鲁棒性评估分析；针对多性能指标优化问题，以6机系统多目标环境经济调度算例为例，对采用优化方法得出的若干候选控制策略进行了鲁棒性评估分析。结果表明所建议方法是可行的。

Optimal control of power system has important significance for improving the performance of power system, for example, reactive power optimization control can effectively reduce the network loss and improve the voltage quality, economic/emission dispatch optimization can effectively achieve the purpose of energy saving and emission reduction and so on. Most of the literatures about the optimal control of power system are based on the deterministic model, which assumes that the load is constant and the control object can be strictly controlled. However, in fact, the load not only is fluctuant but also has a prediction error, the control precision of the control object can only guarantee its control roughly in line with expectations. Taking into account the uncertainty of the load and the possible deviation of the control, the optimal control strategy derived from the point of view of mathematical optimization may sometimes be too sensitive for the above uncertainties and deviations, thereby show poor performance in practice. Therefore, it is necessary to evaluate the robustness of the control strategy, and select the control strategy which is strong enough for the uncertainty and deviation. In order to evaluate robustness of optimal control strategies for electric power systems, an index of robustness evaluation considering control accuracy and uncertainty of state variables was established. Methodologies for identifying whether robustness was satisfied for optimization problems with single objective function and multi-objective function were described respectively. An approach based on sampling optimization to evaluate robustness of control strategies was put forward. Some methods were given to selecting the optimal solution of multi-objective optimization. In case that robustness of several control strategies could meet or not meet requirements, method for comparing robustness of control strategies was suggested. For the optimization problem of single performance index, reactive power control problem for IEEE-30 system and economic dispatch problem for 6 machine system were used as example, where some candidate control strategies were obtained with reactive power optimization approach and their robustness was evaluated with the proposed method respectively to find out the best strategy; Aiming at the optimization problem of multi performance index, economic/emission dispatch problem for 6 machine system was used as example, where some candidate control strategies were obtained with sampling optimization method and their robustness was evaluated with the proposed method respectively to find out the best strategy. The results show that the proposed method is feasible.