关键词: 短期优化调度；机组组合；加速多蚁群算法；正反馈强度调节；稳定最优表；三峡水电站

Abstract: To solve short - term optimal operation problem of the Three Gorges Hydropower Station, an improved ant colony algorithm with enhanced optimization capability called accelerated multiple ant colony algorithm (AMACA) was proposed. And an overall temporal and spatial search matrix was built by increasing the number of ant colony and dispatching every ant to every unit to work in a way of labor division and collaboration. Accelerated searching strategy was used to accelerate the starting time of searching processes of ant colonies generation by generation. Accelerated searching strategy realizes the partition of pheromone feedback intensity and improves global search capability of the algorithm. Local search strategy was used to realize small - scale optimization and to further improve the reliability of start - up and shutdown strategies and dispatching schemes for hydropower stations. Optimal distribution table was created ahead and called in main program to optimize the distribution of total load among units. The running results of AMACA in the Three Gorges Hydropower Station indicate that: fewer water consumption and better operation strategy were obtained using AMACA compared with GA, BACA and EACA under 77.00, 86.00m and 102.00m water head. At the same time, loads of all units are in the stable operation regions using AMACA, which can avoid cavitation vibration, improve the stability of unit operation and improve the robustness of short - term optimal dispatching scheme.

Key words: short - term optimal operation; unit commitment; accelerated ant colony algorithm; positive feedback intensity adjustment; optimal distribution table; Three Gorges Hydropower Station