非预期性与全场景可行性: 应对负荷与可再生能源不确定性的现状, 挑战与未来

Optimal scheduling of power systems with renewable energy is a typical multi-stage stochastic optimization problem. Recently, the importance of all-scenario- feasibility (ASF) and nonanticipativity of the scheduling decisions has been realized by researchers. Some effective security-constrained unit commitment (SCUC) methods have been proposed to guarantee the nonanticipativity and ASF in power system generation scheduling with thermal units and renewable resources. However, the nonanticipativity and ASF related difficulties in the scheduling of cascaded hydropower systems/energy storage systems with uncertain renewable energy/loads are still not resolved. In this paper, the meaning and importance of ASF and nonanticipativity were explained by a simple example. Meanwhile, this paper analyzed the core difficulties in scheduling of thermal units, storage systems and cascaded hydro systems and points out a series of scientific problems needed to be studied. Finally, the influence of nonanticipativity and ASF on the existing methods and concepts such as "reserve" were analyzed and the possible future methodswere envisaged. In conclusion, in order to deal with the uncertainty of renewable energy and loads, it is not only possible but also necessary to establish a new scheduling method satisfying nonanticipativity and ASF. A series of problems involved are needed to be solved in this process.