Elitist Rao Algorithms and R-Method for Optimization of Energy Systems
Elitist Rao Algorithms and R-Method for Optimization of Energy Systems
复制标题
用于能源系统优化的精英 Rao 算法和 R 方法
DOI：
--
复制
发表时间：
2022
影响因子：
2.3
通讯作者：
David Taler
中科院分区：
文献类型：
--
作者：
R. Rao;Hameer Singh Keesari;J. Taler;P. Ocłoń;David Taler
摘要识别、检查和优化可再生能源系统中不同参数的影响可以显著帮助确定其效率。此外，由于这些系统具有多个目标，例如功率输出、系统效率、投资成本、经济和生态因素，因此通常不优选仅考虑一个目标来给出最佳系统参数。针对可再生能源系统的最优系统参数问题，提出了三种改进的Rao算法，即精英Rao算法。此外，提出了一种新的多属性决策方法R-方法，用于从精英Rao算法得到的Pareto前沿中选择最优解。使用30个单目标无约束基准函数对所提算法进行了测试，并使用Friedman统计检验验证了其相对于基本Rao算法的改进意义.最后，以太阳能辅助斯特林热机系统和增压直喷柴油机系统为例，对所提出的算法进行了多目标和多目标优化测试。此外，所提出的算法的有效性方面的超体积，覆盖率和间距度量。此外，所提出的算法在单，多，多目标优化的性能进行了比较，与其他算法从文献中发现是上级或竞争力。
Abstract Identifying, examining, and optimizing the impact of different parameters in a renewable energy system can significantly help determine its efficiency. Furthermore, since these systems have several objectives such as power output, system efficiency, investment cost, economic, and ecological factors, it is often not preferable to present the optimum system parameters considering just one objective. This article proposes three improved versions of recently developed Rao algorithms named elitist Rao algorithms to find optimum system parameters of renewable energy systems. In addition, a new multi-attribute decision-making method named R-method is proposed for selecting the best solution from the Pareto-fronts obtained using the elitist Rao algorithms. The proposed algorithms are tested using 30 single-objective unconstrained benchmark functions, and the significance of improvement over basic Rao algorithms is validated using the Friedman statistical test. Later, the proposed algorithms’ performances are tested in multi- and many-objective optimization scenarios of a solar-assisted Stirling heat engine system and a turbocharged direct injection diesel engine system. Furthermore, the proposed algorithms’ effectiveness is presented in terms of hypervolume, coverage, and spacing metrics. Also, the performances of the proposed algorithms in single-, multi-, and many-objective optimization are compared with the other algorithms from the literature and found to be superior or competitive.