由于固有的间歇性、波动性和低可预测性,大规模可再生能源接入电网给电力系统调度带来了极大挑战。其中关键问题是在可再生能源出力严重偏离预测时,如何以最经济的手段保证系统安全运行。基于此考虑,文中提出了电力系统鲁棒经济调度的基本概念、模型和实现框架,其理论基础可归结为“鲁棒可行性”问题。它本质上是一个自然决策的不确定性和人工决策的调度系统之间的两阶段零和博弈问题。以鲁棒可行性检测为基础,提出了鲁棒经济调度问题的系统化求解算法,通过逐次产生割平面不断排除非鲁棒可行点,最终获得可靠的调度策略,对可再生能源在给定集合内的任意出力组合均可保证存在可行的校正策略使系统安全运行,并使调度成本最低。
Owing to the inherent intermittence, fluctuation, and low predictability, integrating large-scale renewable power generation into bulk power grids has posed significant challenges to system operators. One of the key issues is how to guarantee safe operation at minimum cost where renewable power generation seriously deviates from prediction, Motivated by such considerations, this paper proposes a generalized concept and implementation framework of robust and economical scheduling, whose theoretical foundation is boiled clown to the robust feasibility problem, which is actually a two-move zero sum game between nature and the decision maker. Moreover, a systematic solution algorithm is developed for the robust and economical scheduling problem. By gradually cutting off non-robust feasible solutions, the final robust solution will be immunized against all possible realization of uncertain renewable power generation and minimize the scheduling cost.