CEC2020 Real-World Constrained Engineering Optimization Seven-Problem Suite

We have assembled the “CEC2020 Real-World Constrained Engineering Optimization Seven-Problem Suite,” comprising the following benchmark problems: RC15 – Speed Reducer Weight Minimization RC17 – Tension/Compression Spring Design RC19 – Welded Beam Design RC20 – Three-Bar Truss Design RC23 – Step-Cone Pulley Design RC28 – Rolling Element Bearing Design RC31 – Gear Train Design Using this suite, we evaluated our proposed CFMINFO algorithm against three mainstream metaheuristics—GWO, DE, and SSA—the original INFO algorithm, and EnMODE (the fourth-place finisher in the CEC 2020 Real-World Single-Objective Constrained Optimization Competition). Performance was compared on each problem using the mean objective value, standard deviation, and Friedman ranking.本研究构建了「CEC2020真实世界约束工程优化七问题测试集（CEC2020 Real-World Constrained Engineering Optimization Seven-Problem Suite）」，包含如下基准测试问题： RC15——减速器重量最小化问题 RC17——拉压弹簧设计问题 RC19——焊接梁设计问题 RC20——三杆桁架设计问题 RC23——阶梯带轮设计问题 RC28——滚动轴承设计问题 RC31——齿轮传动系统设计问题 基于该测试集，本研究将所提出的CFMINFO算法与三种主流元启发式算法——灰狼优化算法（Grey Wolf Optimizer, GWO）、差分进化算法（Differential Evolution, DE）、麻雀搜索算法（Sparrow Search Algorithm, SSA）——以及原始INFO算法、EnMODE算法（CEC2020真实世界单目标约束优化竞赛第四名获奖算法）进行了性能对比。 针对每个测试问题，均采用平均目标值、标准差与Friedman排名作为评价指标开展性能对比。