Hydropower Generation Scheduling Of Large-Scale Cascaded Hydropower Stations

There are some characteristics, such as multi-dimensional, stochastic inflow, non-uniform service and scheduling subjects, in optimal scheduling of power generation. The problem is also subjected to hydrological and meteorological conditions, operation control, water demand, and loads of the power grid. As a result, hydropower genertion scheduling problem is a typical type of complex optimization problem with multi-constraints and multi-dimensions. With the continuous development of Chinese hydropower, large-scale cascade hydropower stations have been gradually formed, which has played an irreplaceable role to support floods and droughts control, water and energy supply. With the increase of cascade hydropower stations, requirements of comprehensive utilization have become more and more complex. The contradictions between single reservoir and the cascade, basin and the whole are sharpening. Traditional optimization theories and methods have been unable to meet the requirements of efficient use of large-scale cascade hydropower stations and water resources optimization, it’s urgent to find out some new theories and methods to adapt and solve the subject about optimal scheduling of generation dispatch in large-scale cascade hydropower stations. This paper focused on the key problems and technologic difficulties in hydropower generation shecheduling of large-scale cascade hydropower stations. In the context of great development on the upper reaches of Yangtze River Basin, the paper pursued comprehensive economic benefits and great efficient of water utilization. Combined with hydropower energy science, multi-objective optimization theories and system science, we carried out researches about modeling and solving the subject of hydropower generation dispatch of large-scale cascade hydropower stations. Some results which have theoretical and application values has been achieved, which could be used as reference in practical scheduling. The main contents and innovations of this paper is as follows:(1) Hydropower operation in dry season is an important problem that influences the comprehensive economic benefits of cascade hydropower stations. In order to increase the total benefit of power generation and decrease the waste of the water, our research analyzed the rule of runoffs in dry season and pre-flood season of a representative hydrologic station-PingShan hydrologic station. Then a method to control water level in May and June is proposed. On this basis, the paper analyzed the influences of different drawdown depth of Xiluodu and Xiangjiaba cascade. Feremore, different operation schemes about level control of Xiluodu and Xiangjiaba is disscussed to find the best strategy of different runoffs. Finally, this paper focused on cascade hydropower stations consist of Wudongde, Baihetan, Xiluodu and Xiangjiaba in lowerer reaches of the Jinsha River. The method of level control in dry season was studied in the long run. The results of this paper can be applied to cascade hydropower stations in Jinsha River.(2) The optimal operation of cascade hydropower stations is a problem with multi-dimensional, stochastic inflow, non-uniform service and scheduling subjects. From this perspective, we developped the model of optimal operation for Jinsha River cascaded as well as the model of large-scale cascaded hydropower stations. Then virtual partition based on the relative location and hydraulic connection can help partition optimal operation of sub-layer in each virtual partition. To overcome technical bottlenecks when applied to practical engineering problems, an elite-guide particle swarm optimization (EGPSO) is proposed. External archive set which can preserve elite solutions along the evolution process is employed to provide flying directions for particles. In this way, EGPSO algorithm can figure out solutions with high accuracy. It is also proved it’s an effective approach to solve the multiple constrained coupling problems when used the constraint-corridor method. Applications show that the proposed method is of high accuracy, good efficiency and strong engineering practicality, which provides a new method to deal with high-dimensional and complex operation problem of large scaled hydropower scheduling.(3) In order to explore the compensation benefits and rule of power generation in lower reaches of Jinsha River and Three Goreges cascade, optimal operation models of single station and cascade stations are founded respectively. These models are developped to pursure maximum hydropower generation. With different input of typical annual runoff, EGPSO algorithm referred above is applied to solving the model. Then, the differences between two models about power generation and discarded water in spatial and temporal distribution were analyzed. The comparative analysis reveals that the joint dispatching of lower reaches of Jinsha River can significantly improve the whole generating capacity and reduce discarded water. Its compensation benefits are significant. In condition of cascade operation, Jinsha River has a power compensation for Three Goreges cascade, which contributes to improving total output in dry season. Meanwhile, scheduling risk of centralized water-level fluctuation is alleviated in flood season.(4) Based on engineering requirement of hydropower station and hydro-thermal system, we analyzed competition or collision relationship among power generation, firmed output and navigation discharge of cascaded hydro plant. The relationship between operation cost of hydro-thermal coordinated system and pollution emission is also condsidered. In this way, we established a generation multi-objective scheduling model of hydropower stations and a short-term multi-objective operation model of regional hydro-thermal system. In order to solve multi-objective problems efficiently, we proposes an improved multi-objective particle swarm optimization (MOPSO) based on the existing single-objective algorithm. In this new algorithm, external archive set is employed to provide flying directions for particles, which can not only improve the accuracy of solution but also enhance the uniformity of the of Pareto optimal solution set. By applying the method to the multi-objective model, the simulation experiment shows that it can get a set of Pareto optimal solution rapidly. On this basis, energy benefit, capacity benefit and navigation benefit of hydropower were analyzed, as well as the operation cost and emission reduction. The reasearch provided technical support for cascade hydropower stations and interconnected hydro-thermal coordinated system.