Moyo Learnmore, Nwulu Nnamdi I, Ekpenyong Uduakobong E
University of Johannesburg.
Aurecon Group PTY Ltd.
MethodsX. 2020 May 23;7:100932. doi: 10.1016/j.mex.2020.100932. eCollection 2020.
The Generator Maintenance Schedule model is formulated mathematically as a highly constrained combinatorial optimization problem and it is obligatory to implement a suitable optimization tool to determine the best feasible maintenance schedule. The maintenance schedule obtained has to meet a number of power system constraints. There is increased research in the development of approximate solution methodologies such as heuristic and meta-heuristic techniques [1]. Unlike mathematical methods, metaheuristics can obtain an optimal solution to a complex problem fast and are not subjected to limitations such as linearity, continuity, differentiability and convexity that are faced by mathematical programs [2]. This work presents the application of Exchange Market Algorithm (EMA) to find an optimal maintenance schedule. The algorithm is customized to achieve the following:•Selecting the initial population within the maintenance window constraint to enable faster convergence.•Adapt the algorithm to give discrete solutions.•Penalty function included for constraint handling.
发电机维护计划模型被数学公式化为一个高度受限的组合优化问题，因此必须实施合适的优化工具来确定最佳可行维护计划。所获得的维护计划必须满足若干电力系统约束条件。在开发近似求解方法（如启发式和元启发式技术）方面的研究不断增加[1]。与数学方法不同，元启发式方法可以快速获得复杂问题的最优解，并且不受数学规划所面临的线性、连续性、可微性和凸性等限制[2]。本文介绍了应用交易市场算法（EMA）来寻找最优维护计划。该算法经过定制以实现以下目标：
在维护窗口约束内选择初始种群，以实现更快的收敛。
使算法能够给出离散解。
包含用于约束处理的惩罚函数。
