基于离散纵横交叉算法的含风电电力系统机组组合优化

摘要

Randomness and volatility of wind power brings serious challenges to safe and economic operation of the power system and reasonable wind power uncertainty model and unit commitment (UC)optimization method are critical to ensure safety and economy of day-ahead dispatching of the power system.Therefore,this paper presents a kind of UC optimization method for the power system with wind power based on discrete crisscross algorithm.According to day-ahead predictive val-ues and error distribution of wind power,this method adopts Latin hypercube sampling technique and Cholesky decomposi-tion to generate large numbers of wind power scenarios and uses synchronous back generation reduction method to reduce those wind power scenarios to small numbers of most likely scenarios.Aiming at minimizing the expected operational costs of thermal power generating units under the most likely scenarios,it establishes the UC optimization model,meanwhile it pro-poses the optimized method based on discrete crisscross algorithm to solve difficulties in UC optimization.Experimental sim-ulation verifies effectiveness of the proposed method.%风电的随机性和波动性给电力系统的安全经济运行带来了严峻的挑战,合理的风电不确定性模型及机组组合优化方法是保证电力系统日前调度安全性和经济性的关键.为此,提出一种基于离散纵横交叉算法的含风电电力系统机组组合优化方法.根据风电功率的日前预测值及其误差分布,采用拉丁超立方抽样技术和 Chol-esky分解生成大量的风电场景,利用同步回代削减法将生成的风电场景削减为少量最有可能发生的场景.以最有可能发生的场景下火电机组的期望运行成本最低为优化目标,建立机组组合优化模型.针对机组组合难以求解的问题,提出了离散纵横交叉算法的优化方法.通过实验仿真验证了所提方法的有效性.