The Application Of Rapid Niche Adaptive Genetic Algorithm To AGC Of Cascaded Hydroelectric System

The cascaded hydroelectric stations are playing a more and more significant role in the operation of the power system. The research for combined operation of hydroelectric systems will bring society and economic benefits enormously. As the core section of the whole optimized scheduling among cascaded hydroelectric stations, the study for AGC (Automatic Generation Control) of cascaded hydroelectric systems should be emphasized enough. The main aim of the research works on AGC is to reduce the cost of generating electricity and improve the benefits. It aims at distributing the cascaded total loads to each station among the cascaded hydroelectric system economically and controlling the running state and load dispatching of the units for each cascaded hydroelectric station in order to meet the optimal economic mode while satisfying various constraints as well as ensuring the safety operation on the cascaded hydroelectric system. In this paper, the rapid niche adaptive genetic algorithm (RNAGA) is proposed for scheduling of cascaded AGC systems based on the existing research fruits after reviewing the correlative algorithms to optimal operation of hydroelectric plant. Then, according to the research goal of cascaded AGC, the complex mathematical models of AGC are discussed systematically. Aiming at the actual restriction of Qingjiang cascaded hydroelectric system, the emphasis is put on the implementation of AGC utilizing RNAGA to solve each complex mathematical models. At last, The case study for the Qingjiang cascaded AGC is executed to show the efficiencies of the proposed algorithm and adopted optimal mathematical models for cascaded AGC. In detail, in chapter 1, it introduces background details of research, literature survey, and contributions of the present studies and organization of the thesis in brief. In chapter 2, based on some reported achievements for improving GA, RNAGA is proposed by introducing niche idea to basic genetic algorithm. And the efficiency of the proposed algorithm is demonstrated through the simulation examples. In chapter 3, based on the features of Qingjiang hydroelectric system and the optimal rule of minimizing the water resource consumption, the mathematical models for unit commitment and loads dispatching are built. And based on the proposed generic algorithm, the implementations of solving two kinds of models are expatiated respectively. In chapter 4, the main tasks for AGC are discussed firstly. Then according to the actual situations of Qingjiang cascaded power system, the maximum storage rule is adopted to set up the optimal operation models for AGC between two plants. Lastly, thinking to actual system constraints, each step of operation for utilizing the proposed algorithm to solve these models is elaborated. In chapter 5, utilizing the actual data provided by Qingjiang, we implement the case calculation for Qingjiang cascaded AGC and achieve ideal results. At last, a conclusion is drawn and some promising aspects are mentioned in chapter 6.