实码退火遗传算法在厂内经济运行中的应用

Economical Operation of Hydropower Station Based on Real-coded Annealing Genetic Algorithm

针对一般优化算法在大型水电站厂内经济运行中精度低或计算速度慢的缺点,对实数编码退火遗传算法(AGA)进行改进,并用于大型电站厂内经济运行。模拟退火算法是在遗传算法(GA)中引入模拟退火算法(SA),它吸收了遗传算法速度快和模拟退火精度高的优点。此外,对传统退火搜索方法的改进,进一步提高了退火遗传算法解决大型优化问题的能力。为了体现退火遗传算法的特点,对某一大型水电站分别采用退火遗传算法、动态规划(DP)、加速遗传算法(AG)、标准遗传算法(SGA)和模拟退火进行了经济运行计算,计算结果表明退火遗传算法易于实现,精度高,收敛速度较快,有一定实用价值。

The paper presents the application of an improved annealing genetic algorithm（AGA） to solve the economical operation problem. In order to overcome the shortcoming of current optimal algorithms in the economical operation problem, such as poor precision or slow speed, the AGA is improved and used in the economical operation of the large scale hydropower station. It integrates the simulated annealing（SA） into the genetic algorithm（GA）, and absorbs their virtues, GA＇s fast speed and SA＇s high accuracy. The AGA＇s search method is improved to elevate its performances for the optimal problem. To reveal its outstanding characters, the AGA are contrasted with several algorithms, including dynamic programming（DP）, accelerating genetic algorithm（AG）, standard genetic algorithm（SGA） and simulated annealing （SA）. The results show that it is easy to implement the AGA in the economical operation problem, moreover, its convergence is quite fast and the solution quality is very good. So the AGA is useful for the economical operation.