Chance Constrained Optimal Operation Rules For Hydropower Reservoirs Based On The Stochastic Dynamic Programming

With the development of Chinese hydroelectric stations, operation scheduling for hydropower reservoirs need managements. It not only can play a better engineering benefits of hydropower system, but also can greatly improve the power system stability and security. Hydropower reservoirs can accomplish these important tasks consisted of generating electricity, supplying water, controlling flood, maintaining ecology, shipping and so on. Therefore, research on the optimal operating rules for the hydropower reservoirs has a great significance.The optimal operation rules of hydropower reservoirs are formulated by considering the hydrological constraints, reservoir constraints and the linkage between each reservoir in group. According to previous research on runoff regulation theories, methods of waterpower calculation, techniques and the runoff of hydropower reservoirs for long periods, study the optimal operation rules. In order to meet the requirements of hydropower reservoir operation at all aspects, how to develop a new, more reasonable and more optimal operation rule is the main problem of controlling and managing hydropower reservoir operation.This paper researches on the above optimal operating rules for the hydropower reservoirs. The optimal operating problems for the giant and cascade hydropower reservoirs in the middle or lower reaches of the Lancang River in Yunnan province as the background,5hydropower reservoirs including Xiaowan, Manwan, Dachaoshan, Nuozhadu and Jinghong are chosen as the study objects. Due to cascade hydropower reservoirs’optimal operating problems, this paper studies the hydropower station and reservoir operation chance constrained optimization models used the stochastic dynamic programming. The specific works are as follows:A stochastic dynamic programming model is established for the hydropower station reservoir optimal scheduling in order to gain the maximum average annual generating electricity, with the combination of long-term reservoir regulation capacity or cascade reservoirs as state variables and the average monthly output of hydropower station as the decision variables. In the middle or lower reaches of the Lancang River the cascade hydropower stations system as an example, the results show that this method of hydroelectric power generation scheduling is effective.Due to the problem of hydropower station reliably generating power, an opportunity constraint optimization model of hydropower station and reservoir group regulated by the guarantee rate of power generation and the maximum failure depth is set up. At the same time, combining with Lagrange relaxation method for the penalty function method resolves the problem. The results indicate that this model, based on electricity generation’s indicators, can gain the optimal scheduling rules satisfied the guarantee rate of power generation and the failure depth.In view of the reliability and failure depth constraint of cascade hydropower stations and reservoirs’scheduling, the dynamic opportunity constrained optimization scheduling rules for hydropower stations and reservoirs in different period are recorded. Meanwhile, the stochastic dynamic programming algorithm is used to solve the dispatching rules of each cascade hydropower station and reservoir successively. The case shows that this method can obtain the dispatching rule set related to operation time, historical simulation operation record and storage.Finally, this paper is summarized, while this problem is prospected to be studied in the future.