基于蚁群优化算法的机组最优投入

机组最优投入问题（optimal Unit Commitment,UC)是寻求1个周期内各个负荷水平下机组的最优组合方式及开停机计划，使运行费用为最小。该问题是一个高维数、非凸的、离散的、非线性的优化问题，很难找出理论上的最优解，但由于它能带来显著的经济效益，所以受到了国内外很多学者的广泛关注。作者尝试采用一种新型的模拟进化优化算法－－蚁群优化算法（ACO）来求解该问题。首先，利用状态、决策及作者提出的路径概念把UC设计成类似于旅行商（TSP）问题的模式，从而可以方便地利用ACO来求解。其次，由于ACO处理的是无约束优化问题，对于UC这一约束优化问题，提出了不同的方法来处理各种约束。用tabu表限制不满足旋转备用约束和机组最小启／停时间约束的状态；通过附加惩罚项来处理线路N安全性约束。数值算例验证了此算法的可行性和有效性。

OPTIMAL UNIT COMMITMENT BASED ON ANT COLONY OPTIMIZATION ALGORITHM

The aim of optimal unit commitment (UC) is to determine an optimal schedule which can minimize the system production cost during a period while the load demand,spinning reserve, physical and operation constrains of theindividual unit are simultaneously satisfied. Because UC is a nonlinear, large scale, non-convex, discrete, mixed-integer combinatorial optimization problem with constrains, so it ishard to find out the theoretical optimal solution of UC, but the attentions from the scholars home and abroad are widely paid to this problem for outstanding economic benefit can bebrought about by use of UC. In this paper the Ant Colony Optimization (ACO) algorithm which is a novel optimization method by simulated evolutions is attempted to solve UCproblem. Firstly, in order to expediently solve UC by ACOalgorithm, a calculating context similar to TSP is designedaccording to the concepts of state and decision and the path definition proposed by authors; secondly, because only the problem without constraints can be solved by ACO algorithm, therefore, some corresponding techniques to deal with equality and inequality constrains are developed, i.e., applying the Tabu list to restrict states violated the constrains, such as spinning reserve, minimum start up/shut down time, etc., thetransmission line capacity limits is handled by additive penalty item. The results of numerical calculation examplesdemonstrate that the presented algorithm is feasible andavailable.