The Optimai Operation Research Of Cascade Reservoirs Based On Improved Ant Colony Optimization

Water is the source of the life, production to, the foundation of ecological, the government work report of two sessions in2012points out that to actively develop the hydropower and raise the proportion of renewable energy sources, as well as that the Twelfth Five-Year Plan report by China Electricity Council points out that hydropower should be developed preferentially with the cooperation of different developing scales, and promote the comprehensive development of cascade hydropower system and reasonable acceptance of renewable energy with the improvement of efficiency and flexibility of power system operation. Therefore, in the high speed development period of hydropower, how to carry out effective research and optimal dispatching management of cascade hydropower stations, and give full play to the potential benefits of non-structural measures, is a major issue of social economy today.Carrying out the scientific optimization and economic dispatch of cascaded reservoirs group, giving full play to the regulatory effect of hydropower station and improving accuracy of water regime prediction are important means to improve the waterpower utilization rate of basin hydropower reservoirs and realize the maximum of water resources utilization. Adhering to the scientific dispatch and making the best use of scientific methods is an important way to realize the national policy of energy saving and emission reduction, which changes the reservoirs dispatch work from "experienced, extensive and qualitative" type to "scientific, careful and quantified" one. Taking Jinsha River middle reaches hydropower stations as example, the paper gives a further discussion on the problems of the optimization dispatching management of cascaded reservoirs group. Aiming at different length of dispatching period, the discussion is centering on the basic theory of reservoirs, from different angles such as the construction of the optimization model, the formulation of the dispatching rules, the analysis of load curve and so on. The main results are as follows:(1) The optimization algorithm of swarm intelligence has the advantage of solving problems quickly. When choosing the route according to the pseudo-random ratio rules, the variation of traditional ant colony algorithm is blindly random and the pheromone volatilizes too quickly. Thus, the algorithm will be trapped in local optimum or slow convergence when the population size is large. The improved ant colony algorithm POM-ACO based on the progressive optimized mutation operator is put forward. Using the variation module of guiding type to improve the performance of the ant colony, and making use of the mechanism of self-adaption pheromones volatile factor will improve the convergence speed of the optimization dispatching of reservoirs.(2) As the routine operation chart of annual regulation reservoir is made out according to typical annual runoff material, it is hard to make full use of the maximum operation efficiency of reservoirs. Because of this disadvantage, the establishment model of the reservoir operation chart based on the improved ant colony algorithm is proposed. Making full use of all the long serial hydrologic data that is given, the effectiveness of using the improved ant colony algorithm based on the mutation operator of gradual optimization to solve the problem of the optimization dispatching of reservoirs is proved. This can be proved through the case study of the guaranteed output and the generated energy.(3) The decision of dispatching function of the hydropower station is clear and the operation efficiency is obvious. Because the traditional dispatching function is generated of pure numerical simulation without considering the characteristics of reservoirs, and there is certain space for optimization, the dispatching function optimization model based on the ant colony algorithm is proposed. This model operates the Iterative optimization of the value of the coefficient perturbation sequence of the initial dispatching function. After optimization, the dispatching function will improve the operation efficiency of the reservoir hydropower station obviously. Thus, it will fully reflect the practicality of the model, and the model will effectively guide the actual dispatching operation of hydropower station. Moreover, it provides new ideas for the optimization dispatching research of the reservoir hydropower station.(4) Starting from the basic principle of reservoirs and combining with the optimization idea, the mathematical modeling power dispatching operation rules of the discharge forecast of the cascade reservoir of hydropower stations is proposed and applied into the practical operation project of cascade hydropower stations in Jinsha River middle reaches, which makes full consideration of the runoff forecast material. These rules are not only more practical and convenient in operation, and the effect of it is second only to the operation optimization dispatching solution, which has an obvious improvement compared with the routine operation scheme. In addition, it is easier to correct the bad influence of the prediction error by the method that the reservoir prediction discharge operation rules of cascade hydropower stations is respectively determined by different period. Moreover, it is rolling scheduled according to the size of the forecasted incoming flow, and it will improve the accuracy of the results. In addition, the reservoir prediction discharge operation rules have a good practicability and strong in operation. It is suitable to apply in the actual dispatching. Not only ensure the safety of flood control, but also give consideration to maximize the power benefit of reservoir. It realizes the goal of combining the power benefit and flood control benefit dynamically with the characteristics of innovative, scientific, reasonable and practical.(5) The safe and stable operation of the power network requires the output characteristics anastomosed with the system load characteristics as more as possible. With adequate consideration of requirements of both generation side and demand side, the short-term power optimization-dispatching model of cascaded reservoirs based on the load curve is established the output process of the power station is distributed according to the shape of the load curve. Using the daily and weekly regulating storage of reservoirs properly makes the output process in accordance with the load curve. In the meanwhile, the maximization of the generated energy is pursued.