Optimal Operation Method of Cascade Hydropower Stations Based on Improved Genetic Algorithm

Abstract

梯级水电灵活性较强广泛应用于电网调峰、调频，然而梯级水电站群约束众多，是高度复杂的混合整数非线性规划问题，求解难度较大。为此本文提出一种改进的遗传算法对梯级水电站进行优化求解，针对遗传算法时效性差、容陷入局部最优的缺点，本文从初始解的产生、交叉变异概率选取、精英个体保存三个方面对遗传算法进行改进。以一个二级水电站为例进行验证，结果表明，改进后遗传算法收敛更快、求解效率更高，只需要59.30 s，即可得出结果；在保证其它约束不变的情况下，梯级水电多发45.56万kW•h。能有效满足梯级水电联合优化调度时效性与经济性的要求。 Cascade hydropower has strong flexibility and is widely used in peak load regulation and frequency regulation of power grid. However, there are many constraints of cascade hydropower station group, which is a highly complex mixed integer nonlinear programming problem and difficult to solve. In this paper, an improved genetic algorithm is proposed to solve the problem of cascade hydropower stations. Aiming at the shortcomings of poor timeliness and local optimum of genetic algorithm, this paper improves the genetic algorithm from three aspects: the generation of initial solution, the selection of cross mutation probability and the preservation of elite individuals. Taking a two-stage hydropower station as an example, the results show that the improved genetic algorithm has faster convergence and higher efficiency, and it only takes 59.30 s to get the result; under the condition of keeping other constraints unchanged, the cascade hydropower station can generate 455,600 kW•h more. It can effectively meet the requirements of timeliness and economy of cascade hydropower joint optimal operation.