Short Term Hydropower Scheduling Model Considering Constraint Priorities

The mixed integer linear programming(MILP)model for short-term optimal scheduling of hydropower stations needs to consider complex cascade constraints,power station constraints and unit operation constraints,which are closely coupled with each other.The dispatchers often set the constraints and objectives of the model according to the scheduling experience,which may cause the contradiction between constraints and lead to no feasible solution of the model.When the model has no solution,it is time-consuming to find conflicting constraints and difficult to consider the importance of different constraints,which will seriously affect the efficiency of the hydropower station day-ahead scheduling.To solve this problem,this paper proposes a constraint grading model for short-term optimal scheduling of cascade hydropower stations.When the original scheduling model is unsolvable due to conflicting constraints,the low priority soft constraints are destroyed as much as possible to seek feasible solutions,and reduce the damage to soft constraints.Firstly,the destructible constraints are sorted by priority and converted into soft constraints by introducing destructive indicator factor and destructive degree variables to make the model have solutions.The destructive factor term and destructive degree term are set in the objective function.The destructive factor term is used to strictly reflect the priority of soft constraints and the destructive degree term is used to reduce the destructive degree of soft constraints.Finally,the model is converted into mixed integer linear programming model(MILP)and solved by Lingo solver.The application example of Beipanjiang cascade hydropower station verifies the feasibility and effectiveness of the proposed model.(1)To address the problem of conflicting constraints in the economic operation model of hydropower station,a constraint graded economic operation model of hydropower station considering unit operation constraints is proposed.Five levels of soft constraints are set in order of priority from lowest to highest: minimum water consumption,lower limit of water level operation,unit climbing capacity,upper limit of water level,and unit vibration zone.Taking the Guangzhao hydropower station as the research object,the results show that the proposed model destroys the three low priority soft constraints of minimum water consumption,lower water level limit and unit climbing capacity,and strictly meets the two high priority soft constraints of upper water level operation limit and unit vibration zone.(2)To address the problem of conflicting constraints in the peak shaving optimization model of cascade hydropower stations,a peak shaving optimization scheduling model of cascade hydropower stations considering the grading of constraints is proposed.According to the order of priority from low to high,five soft constraints are set in the dry season,including peak shaving,the last period water level of the Guangzhao hydropower station,the lower limit of cascade output,the lower limit of water level of the Guangzhao hydropower station and the minimum discharge of the lighting Guangzhao hydropower station,five soft constraints are set in the wet season,including peak shaving,the last period water level of the Guangzhao hydropower station,the upper limit of cascade output,the water abandoned by the Guangzhao and downstream hydropower stations and the upper limit of the water level of the Guangzhao hydropower station.Taking Beipanjiang cascade hydropower station as the research object,in the dry season,the proposed model destroys the soft constraints of peak shaving,the last period water level of the Guangzhao hydropower station and the lower limit of cascade output,and satisfies the soft constraints of the lower limit of water level of the Guangzhao hydropower station and the minimum discharge of the lighting Guangzhao hydropower station at high priority level.In the wet season,the proposed model destroys the soft constraints of peak shaving,the last period water level of the Guangzhao hydropower station and the upper limit of cascade output,and satisfies the soft constraints of the water abandoned by the Guangzhao and downstream hydropower stations and the upper limit of the water level of the Guangzhao hydropower station.In summary,the application example of Beipanjiang cascade hydropower station shows that the proposed model can solve the problem that the short-term optimal scheduling model of cascade hydropower stations has no solution due to conflicting constraints.It can destroy some soft constraints to obtain feasible solutions when the original model has no solution,and strictly follow the priority level of soft constraints,which improves the efficiency of the day-ahead scheduling of hydropower stations and has strong practical value.