基于改进蚁群算法的发电机组检修计划优化

随着计算机技术的迅速发展，人工智能方法在动态组合优化问题中得到了广泛应用。针对发电机组检修计划优化问题，以系统运行费用最小为目标建立了机组检修计划模型，从检修状态组合的角度将其转化为动态模型，并提出了一种利用改进蚁群算法进行求解的方法。针对基本蚁群算法的不足，从转移概率和信息素动态更新2方面对其进行了改进。利用蚁群算法的状态表记忆机制，较好地处理了检修计划连续性和检修开始时间等约束条件。对于其他约束条件，通过罚函数方式形成增广目标函数。算例结果表明，该方法计算速度较快，灵活性强。

Solving Unit Maintenance Scheduling Problem by Use of Improved Ant Colony Algorithm

Along with the rapid development of computer technology, artificial intelligent methods are widely applied in dynamic combinatorial optimization problems. Aiming at the optimization of unit maintenance scheduling, a unit maintenance scheduling model is built in which the minimum system operation cost is taken as objective function. In view of maintenance state combination, the built model is transformed into dynamic model and by use of improved ant colony algorithm a method to solve the dynamic model is proposed. To remedy its deficiency, the basic ant colony algorithm is improved in both transition probability and dynamic updating of pheromone. Using the state table memory mechanism, the constraints such as the continuity of maintenance scheduling and start time of maintenance are well processes. For other constraints, by means of penalty function the augmented objective function is formed. Case study result shows that the proposed method is flexible and its calculation speed is satisfied.