Solution Selection from a Pareto Optimal Set of Multi-Objective Reservoir Operation via Clustering Operation Processes and Objective Values

Multi-objective evolutionary algorithms (MOEAs) are widely used to optimize multi-purpose reservoir operations. Considering that most outcomes of MOEAs Pareto optimal sets with a large number incomparable solutions, it is not trivial task for decision-makers (DMs) select compromise solution application purposes. Due the increasing popularity data-driven decision-making, we introduce clustering-based decision-making method into multi-objective operation optimization problem. Traditionally, selection has been conducted based on trade-off ranking in objective space, and characteristics decision space have ignored. In our work, processes were innovatively clustered groups unique properties surfaces analyzed via clustering space. To attain suitable performance, new similarity measure, referred as Mei–Wang fluctuation measure (MWFSM), was tailored processes. This describes time series terms both their shape quantitative variation. Then, selected joint use two results. A case study Three Gorges cascade reservoirs system under small medium floods investigated verify applicability proposed method. The results revealed MWFSM effectively distinguishes Two more patterns similar positions but different shapes identified when compared Euclidean distance dynamic warping Furthermore, decreased range from whole set containing relatively few solutions. Finally, selected.