Unit commitment'UC)has commonly been formulated as a large-scale,mixed-integer optimization problem which is with the characteristic of high-dimensional, discrete and nonlinear and is known as NP-hard problem in mathematics. In order to solve the problems of time-consuming and easy to fall into local optimum that the ant colony optimization algorithm(ACO)met,the multi colony ant optimization algorithms was investigated. After the analysis of the use of the multi colony ant optimization algorithms in unit commitment, the detect ant and ergate were presented beside the search ant, the new principles of information exchange were set,and the new update method for pheromone was established. The feasibility of the algorithm was verified, the rationality and effectiveness were analyzed by the modified IEEE30. The result shows that the proposed multi colony ant algorithm is reasonable and effective.%电力系统机组组合问题是一个大规模混合整数规划问题,具有高维、离散、非线性等特点,在数学上被称为NP-hard问题.为解决蚁群算法在解决机组组合问题中遇到的计算速度慢、易陷入局部最优等问题,将多种群蚁群算法应用到解决机组组合的问题中.开展了多种群蚁群算法在机组组合问题中的应用分析,新建了除搜索蚁之外的侦察蚁和工蚁,设定了3种蚁群之间的信息交互原理,提出了各蚁群的信息素更新方法.在修正后的IEEE30节点系统对算法可行性作了验证,并对算法的合理性和有效性进行了分析.研究结果表明,所提出的多种群蚁群算法是合理、有效的.
展开▼
