| description abstract | Models for short-term optimal scheduling of hydropower stations should consider complex cascade constraints, hydropower station constraints, and unit operation constraints, all of which are inextricably linked. In practical scheduling, schedulers cannot be certain that the constraints they set based on their preferences are necessarily feasible, potentially leading to a no-solution situation. When the model has no solution, it is time-consuming to find conflicting constraints, and it is difficult to take the importance of different constraints into account, which will seriously affect the efficiency of day-ahead planning of hydropower stations. To solve the problem, a constraint grading model for short-term optimal scheduling of cascade hydropower stations is proposed in this paper. In both the flood and dry seasons, the model transforms five violable constraints into soft constraints and ranks them according to their importance, respectively. If the model has no solution, the soft constraints with low importance are automatically violated to obtain a feasible solution. The application example of Beipan cascade demonstrates that this model can effectively solve the problem of conflicting constraints. Moreover, compared with the penalty function method, this model can ensure that more important soft constraints are not violated. | |
